Complications in Surgery and Trauma

SAVE OFFLINE

Complications in Surgery and Trauma

DK3109_C000a.indd 1

08/11/2006 2:37:51 PM

Complications in Surgery and Trauma Edited by

Stephen M. Cohn

University of Texas Health Science Center San Antonio, Texas, U.S.A.

Associate Editors

New York London

DK3109_C000a.indd 3

Erik Barquist Patricia M. Byers Enrique Ginzburg Fahim A. Habib Mauricio Lynn Mark McKenney Nicholas Namias David Shatz Danny Sleeman

08/11/2006 2:37:52 PM

Informa Healthcare USA, Inc. 270 Madison Avenue New York, NY 10016 © 2007 by Informa Healthcare USA, Inc. Informa Healthcare is an Informa business No claim to original U.S. Government works Printed in the United States of America on acid‑free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number‑10: 0‑8247‑5898‑6 (Hardcover) International Standard Book Number‑13: 978‑0‑8247‑5898‑1 (Hardcover) This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978‑750‑8400. CCC is a not‑for‑profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation with‑ out intent to infringe. Library of Congress Cataloging‑in‑Publication Data Complications in surgery and trauma / edited by Stephen M. Cohn. p. ; cm. Includes bibliographical references and index. ISBN‑13: 978‑0‑8247‑5898‑1 (hardcover : alk. paper) ISBN‑10: 0‑8247‑5898‑6 (hardcover : alk. paper) 1. Surgery‑‑Complications. 2. Wounds and injuries‑‑Complications. I. Cohn, Stephen M. [DNLM: 1. Intraoperative Complications‑‑prevention & control. 2. Postoperative Complications‑‑prevention & control. 3. Wounds and Injuries‑‑complications. WO 181 C7373 2006] RD98.C665 2006 617’.9‑‑dc22

2006044474

Visit the Informa Web site at www.informa.com and the Informa Healthcare Web site at www.informahealthcare.com

DK3109_C000a.indd 4

08/11/2006 2:37:52 PM

This text is dedicated To my supportive parents, Leland and Iris, To my wonderful children, Sam and Elizabeth, To my loving wife, Kelly. Lechaim (To Life!) Steve Cohn, 2006

Foreword

Most surgeons would prefer not to talk about surgical complications, much less write about them. If they are written about, unusual complications are frequently published as case reports. For this reason, I know of no single-author books on this topic; multiple-author works on surgical complications are the norm. This book is no exception. The multiple authors of its chapters have been chosen for their specific expertise in a given anatomic or physiologic field. From the patient’s viewpoint, a surgical complication is any unexpected event that occurs after surgical intervention and causes the patient pain or suffering. From the surgeon’s viewpoint, many of the occurrences that patients would call complications are seen instead as sequelae. A sequela is an untoward occurrence that is out of the surgeon’s control, such as phantom limb pain after amputation, the dumping syndrome after gastric resection, and the onset of diabetes after pancreatic resection. The list is almost endless. Surgeons should warn their patients about potential problems before performing the procedure and should convince the patient that the need for operative intervention outweighs any problems that may result. A classic example of such a problem is wound infection. Wound infection can be a sequela when the surgeon is forced to operate through an infected site. In fact, one wound classification system is based on the potential that infection may develop. More often, however, wound infections are true complications and can be prevented. Among the many available preventive techniques are nutritional support, prophylactic administration of antibiotics, skin preparation, sterility of the operative environment, and sterile technique. For example, before sterile technique was adopted, compound fractures resulted in amputation and a mortality rate of 50% to 80% because of systemic infection, probably by streptococcus. The patients who survived frequently experienced the development of ‘‘laudable pus,’’ which indicated a localized infection, probably due to staphylococcus. War wounds, whether clean contaminated or fully contaminated, were and still are treated with debridement and secondary closure. Appendicitis with rupture is still treated with secondary closure

at many medical centers. Since 1900, long before the development of antibiotics, addressing clean wounds under sterile surgical conditions has been associated with low rates of wound infection. The example given above, wound infection, cuts across all surgical disciplines and demonstrates the need for a system-oriented approach to minimizing complications. The occurrence and severity of complications are affected by factors related to patient, environment, hospital, nursing, and surgeon. It makes little difference whether one uses the ‘‘weakest link’’ or the ‘‘Swiss cheese’’ model to explain the occurrence of complications due to failure of the system if, during each case, the level of care is optimized and adequate communication between all personnel providing patient care is ensured. A strong case can be made for the usefulness of the classic surgical mortality and morbidity conference in analyzing the system failures associated with a given complication. The mortality and morbidity discussion addresses approaches that can be used to obviate a surgical complication in the future. The use of a table can be helpful in placing complications related to system failure into such categories as errors in judgment, errors in technique, and delay in diagnosis, treatment, or both because of the disease progress. Such an analysis serves both educational and quantity-of-care goals. The thrust of this book is to review and classify the complications related to surgery from the standpoints of prevention, recognition, and management so that their impact on the patient’s recovery can be minimized. This book analyzes the complications associated with many types of surgical intervention and dissects the optimal manner of preventing each one. It also discusses diagnostic and therapeutic techniques that can be used to minimize the impact of such complications on the welfare of the patient. Therefore, this book should be a standard reference for all surgeons, regardless of specialty. J. Bradley Aust, M.D., Ph.D. Department of Surgery, University of Texas Health Science Center at San Antonio, San Antonio, Texas, U.S.A.

Preface

‘‘This was not my patient . . . ’’ ‘‘I was not present for this case . . . ’’ ‘‘There is nothing different I would do the next time . . . ’’ ‘‘Let me have the surgical attending tell you why we did this . . . ’’ ‘‘I never saw this patient . . . ’’ ‘‘If you do enough operations you are bound to have this happen’’ ‘‘It was an act of God . . . ’’ ‘‘I did a perfect operation . . . the ungrateful patient died . . . ’’ Anonymous Chief Resident This text was conceived to provide important information regarding the incidence and management of complications encountered in the surgical care of patients. More importantly, the contributing authors have identified methods to prevent or avoid complications. To paraphrase Albert Einstein, ‘‘Geniuses learn from other people’s mistakes.’’ This book is dedicated to all the surgeons who have paved the path for surgical success. Stephen M. Cohn

Contents

Foreword J. Bradley Aust . . . . v Preface . . . . vii Contributors . . . . xvii The Surgical Mortality and Morbidity Review Julie Heimbach, Jyoti Arya and Alden H. Harken . . . . xxi The Surgical Mortality and Morbidity Review: Best Practices and Procedures Robert E. Condon . . . . xxiii PART I: Complications in the Intensive Care Unit 1. Complications of Anesthesia . . . . . . . . . . 1 Miguel A. Cobas and Albert J. Varon Complications of Airway Management . . . . 2 Complications of the Respiratory System . . . . 4 Complications of the Cardiovascular System . . . . 5 Complications of the Renal System . . . . 6 Complications of the Neurologic System . . . . 8 Complications of Regional Anesthesia . . . . 9 Complications of Specific Nerve Blocks . . . . 10 Systemic and Metabolic Complications . . . . 11 Conclusions . . . . 13 References . . . . 13 2. Complications of Acute Fluid Loss and Replacement . . . . . . . . . . . . . . . . . . 17 Juan Carlos Puyana Theoretical Basis of Fluid Distribution . . . . 17 Volume Expansion . . . . 19 Volume Replacement After Acute Blood Loss . . . . 21

Goals of Fluid Replacement and End Points of Resuscitation . . . . 23 Complications of Fluid Losses in Surgical Patients . . . . 24 Inherent Complications of Commonly Used Fluid Replacement Solutions . . . . 25 Conclusions . . . . 26 References . . . . 26 3. Complications of Antibiotic Therapy . . 29 Mohamed Fahim and Nicholas Namias General Complications . . . . 29 Specific Antibiotics and Their Associated Complications . . . . 32 Conclusion . . . . 36 References . . . . 36 4. Complications of Blood and Blood-Product Transfusion . . . . . . . . . . . . . . . . . . . . . . 41 Igor Jeroukhimov and Mauricio Lynn Historical Background . . . . 41 Physiology of Transfusion Therapy . . . . 41 Role of Component Therapy . . . . 42 Pathophysiology of Blood Transfusion . . . . 42 Complication of Massive Transfusion . . . . 47 Volume Overload . . . . 48 Differential Diagnosis and Role of Laboratory Tests . . . . 48

x

Contents

Prevention of Transfusion Complications . . . . 49 Treatment Axioms . . . . 51 References . . . . 51 5. Hypovolemic and Septic Shock . . . . . . 55 Matthew O. Dolich and Don H. Van Boerum Hypovolemic Shock . . . . 55 Septic Shock . . . . 59 Specific Complications of Hypovolemic and Septic Shock . . . . 62 References . . . . 63 6. Complications Associated with the Use of Invasive Devices in the Intensive Care Unit . . . . . . . . . . . . . . . . . . . . . . . . 67 Victor Cruz and J. Martin Perez Central Venous Access . . . . 67 Arterial Lines . . . . 70 Gastrostomy Tubes . . . . 70 Thoracostomy Tubes . . . . 72 Intracranial Pressure Monitoring . . . . 73 Conclusion . . . . 74 References . . . . 74 PART II: Gastrointestinal Surgery Complications 7. Complications of Abdominal Wall Surgery and Hernia Repair . . . . . . . . . . 77 James C. Doherty and Robert W. Bailey Complications of Abdominal Wound Closure . . . . 77 Complications of Ventral Hernia Repair . . . . 79 Complications of Inguinal Hernia Repair . . . . 81 Conclusion . . . . 85 References . . . . 85 8. Complications of Biliary Tract Surgery and Trauma . . . . . . . . . . . . . . . . . . . . . . 87 Akpofure Peter Ekeh and James B. Peoples Common Biliary Operations . . . . 87 Prevention of Potential Pitfalls During Elective Biliary Surgery . . . . 87 Diagnosis of Biliary Complications . . . . 88 Management of Specific Complications . . . . 90 Biliary Trauma . . . . 91 Summary . . . . 92 References . . . . 92

9. Complications of Intestinal Surgery . . . 95 John J. Hong and Sapoora Manshaii General Complications of Intestinal Surgery . . . . 95 Complications of Specific Intestinal Operations . . . . 96 Complications of Intestinal Surgery for Trauma . . . . 98 Complications of Appendectomy . . . . 98 Complications of Colorectal Surgery . . . . 99 Complications of Anal Surgery . . . . 104 Complications of Colorectal Trauma . . . . 105 References . . . . 105 10. Complications of Gastric Surgery . . . . . . . . . . . . . . . . . . . . . . . 111 Dory Jewelewicz, Dean Goldberg, and Erik Barquist Anatomy . . . . 111 Diagnostic Procedures . . . . 111 Peptic Ulcer Disease . . . . 111 Gastric Carcinoma . . . . 113 Gastric Lymphoma . . . . 114 Gastrointestinal Motility . . . . 114 Effect of Surgical Procedures on Gastric Motor Physiology . . . . 114 Acute Complications of Gastric Surgery . . . . 115 Long-Term Complications of Gastric Surgery . . . . 117 Nutritional Consequences of Gastric Surgery . . . . 125 Gastroesophageal Reflux Disease . . . . 127 Infantile Hypertrophic Pyloric Stenosis . . . . 128 References . . . . 129 11. Complications of Hepatic Surgery and Trauma . . . . . . . . . . . . . . . . . . . . 135 Adrian W. Ong and Danny Sleeman Operative Techniques for Hepatic Surgery and Trauma Surgery . . . . 135 Complications . . . . 137 References . . . . 140 12. Complications of Pancreatic Surgery and Trauma . . . . . . . . . . . . . . . . . . . . . . . . 143 Christopher K. Senkowski Anatomy . . . . 143

Contents

General Complications Associated with Pancreatic Surgery . . . . 144 Complications of Operative Treatment for Acute Pancreatitis . . . . 148 Complications of Surgery for Chronic Pancreatitis . . . . 148 Complications of Surgery for Pancreatic Pseudocyst . . . . 149 Complications of Surgery for Pancreatic Trauma . . . . 152 Complications of Surgery for Cancer . . . . 154 References . . . . 157 13. Complications of Splenic Surgery and Splenic Injury . . . . . . . . . . . . . . . 161 David S. Lasko and Louis R. Pizano Categories of Splenic Surgery . . . . 161 Complications of Elective Splenic Surgery . . . . 161 Complications in the Treatment of Splenic Injury . . . . 162 References . . . . 164 14. Complications of Laparoscopy in General Surgery . . . . . . . . . . . . . . . . . 167 David S. Lasko and Robert W. Bailey Physiologic Changes . . . . 167 Complications Related to Laparoscopic Access . . . . 168 Altered Visualization, Unique Instrumentation, and the Role of the Learning Curve . . . . 170 Common Laparoscopic Operations . . . . 171 Conclusions . . . . 174 References . . . . 175 15. Complications of Liver Transplantation . . . . . . . . . . . . . . . . . 179 Gennaro Selvaggi, Andreas G. Tzakis, and David M. Levi Early Complications . . . . 179 Late Complications . . . . 185 Conclusion . . . . 190 References . . . . 190 PART III: Breast and Endocrine Surgery Complications 16. Complications of Breast Surgery . . . . 193 Qammar Rashid, Scott McDonald, and Frederick L. Moffat Complications of Breast Biopsy . . . . 193

xi

Complications of Breast Conservation Therapy for Breast Cancer . . . . 193 Complications of Breast Biopsy After Breast-Conserving Therapy . . . . 195 Complications of Salvage Surgery for Local Recurrence . . . . 195 Complications Associated with Breast Reconstruction . . . . 195 Conclusion . . . . 198 References . . . . 198 17. Complications of Thyroidectomy and Parathyroidectomy . . . . . . . . . . . . . . . 203 George L. Irvin III and Denise M. Carneiro-Pla Postoperative Cervical Hematoma . . . . 203 Laryngeal Nerve Injury . . . . 203 Hypocalcemia . . . . 204 Seroma and Wound Infection . . . . 205 Failure of Parathyroidectomy . . . . 205 References . . . . 205 18. Complications of Adrenal Gland Surgery . . . . . . . . . . . . . . . . . . . . . . . 207 James C. Doherty, Hussein Mazloum, and Danny Sleeman Indications for Adrenalectomy . . . . 207 Endocrine and Metabolic Complications of Adrenalectomy . . . . 208 Technical Complications of Adrenalectomy . . . . 211 Conclusion . . . . 213 References . . . . 213 19. Complications Associated with Surgery for Enteropancreatic Neuroendocrine Tumors . . . . . . . . . . . . . . . . . . . . . . . . 215 Kenji Inaba and Stephen Brower Insulinoma . . . . 215 Gastrinoma . . . . 216 Complications of Hepatic Surgery for Metastatic Pancreatic Endocrine Tumors . . . . 217 References . . . . 218 20. Gastrointestinal Carcinoid Tumors . . 219 Jana B. A. MacLeod, Erik Barquist, and Ha˚kin Ahlman History and Epidemiology of Carcinoid Tumors . . . . 219 Carcinoid Tumors of the Stomach . . . . 220 Carcinoid Tumors of the Small Intestine . . . . 221

xii

Contents

Carcinoid Tumors of the Duodenum . . . . 223 Carcinoid Tumors of the Appendix . . . . 224 Carcinoid Tumors of the Colon and Rectum . . . . 224 Metastasis of Carcinoid Tumor to the Liver . . . . 225 Conclusion . . . . 225 References . . . . 225 PART IV: Complications of Thoracic Surgery 21. Complications of Pulmonary and Chest Wall Resection . . . . . . . . . . . . . 229 Rosemary F. Kelly and Romualdo J. Segurola, Jr. Evaluation . . . . 229 Respiratory Problems . . . . 233 Bronchial Stump Problems . . . . 237 Pleural Space Problems . . . . 238 Cardiac Problems . . . . 240 Chest Wall Problems . . . . 242 Nerve Injury . . . . 245 Conclusions . . . . 246 References . . . . 246 22. Complications of Esophageal Surgery and Trauma . . . . . . . . . . . . . . . . . . . . 251 Peter P. Lopez Anatomy and Physiology of the Esophagus . . . . 251 Assessment of Operative Risk . . . . 253 Esophageal Perforation . . . . 256 Esophageal Trauma . . . . 261 Complications of Esophageal Resection . . . . 263 Conclusion . . . . 266 References . . . . 266 23. Respiratory Failure After Surgery or Trauma . . . . . . . . . . . . . . . . . . . . . 271 Fahim A. Habib Respiratory Failure . . . . 271 Principles of Mechanical Ventilation . . . . 273 Acute Lung Injury and Adult Respiratory Distress Syndrome . . . . 275 Other Causes of Respiratory Failure After Trauma or Surgery . . . . 280 Conclusion . . . . 282 References . . . . 282

24. Complications of Lung Transplantation . . . . . . . . . . . . . . . . . 287 Xiao-Shi Qi and Debra Fertel Historical Perspective . . . . 287 Recipient Selection . . . . 288 Donor Selection . . . . 289 Intraoperative Management . . . . 289 Complications Related to Immediate Graft Dysfunction . . . . 290 Perioperative Complications . . . . 290 Complications of Immunosuppression . . . . 293 Living-Donor Lung Transplantation . . . . 294 Conclusion . . . . 294 References . . . . 294 PART V: Cardiovascular Surgery Complications 25. Complications After Cardiopulmonary Resuscitation and Cardiac Arrest . . . 299 Abhijit S. Pathak, Amy J. Goldberg, and Robert F. Buckman, Jr. Causes of and Factors Underlying Cardiac Arrest . . . . 299 Pathophysiology of Cardiac Arrest . . . . 299 Physiology of Standard Closed-Chest Cardiopulmonary Resuscitation . . . . 300 Management of Cardiac Arrest . . . . 300 Technique of Resuscitative Thoracotomy and Open Cardiac Massage . . . . 301 Complications of Cardiopulmonary Resuscitation . . . . 302 Management After Successful Cardiopulmonary Resuscitation . . . . 303 Conclusion . . . . 303 References . . . . 304 26. Complications of Vascular Surgery . . . . . . . . . . . . . . . . . . . . . . . 305 Frank B. Pomposelli, Jr., Allen D. Hamdan, Malachi G. Sheahan, and Guatam V. Shrikhande Associated Diseases and Their Implications for Complications of Vascular Surgery . . . . 305 Complications Common to All Vascular Procedures . . . . 307 Complications of Lower-Extremity Bypass for Chronic Ischemia . . . . 309

Contents

Complications of Limb Amputation . . . . 313 Complications of Carotid Endarterectomy . . . . 313 Complications of Surgery to the Aorta and its Branches . . . . 315 Venous Diseases . . . . 320 Complications of Surgery for Acute Arterial Insufficiency . . . . 320 Summary . . . . 322 References . . . . 322 27. Acute Complications of Cardiovascular Surgery and Trauma . . . . . . . . . . . . . 327 Riyad C. Karmy-Jones, Edward M. Boyle, and John C. Mullen Technical Complications Related to Cardiopulmonary Bypass . . . . 327 Complications of Repeat Cardiac Operation . . . . 330 Postoperative Myocardial Infarction and Coronary Spasm . . . . 330 Postoperative Bleeding and Tamponade . . . . 331 Postoperative Dysrhythmias . . . . 332 Low Cardiac Output . . . . 333 Neurological Complications . . . . 336 End-Organ Complications of Cardiopulmonary Bypass . . . . 339 Acute Infectious Complications . . . . 340 Acute Complications of Valve Replacement or Repair . . . . 342 Trauma . . . . 344 Conclusion . . . . 349 References . . . . 349 28. Complications of Cardiac Transplantation . . . . . . . . . . . . . . . . . 357 Xiao-Shi Qi, Louis B. Louis IV, and Si M. Pham Indications for Cardiac Transplantation . . . . 357 Operative Techniques . . . . 357 Intraoperative Complications . . . . 358 Early Postoperative Complications . . . . 358 Infectious Complications . . . . 361 Posttransplantation Malignancy . . . . 361 Cardiac Allograft Vasculopathy . . . . 362 Gastrointestinal Complications . . . . 363 Other Complications of Immunosuppression . . . . 363 Conclusion . . . . 363 References . . . . 363

xiii

29. Complications of Mechanical Circulatory Support . . . . . . . . . . . . . . . . . . . . . . . 367 Fotios M. Andreopoulos and Richard J. Kaplon Devices . . . . 367 Complications . . . . 370 Experimental Devices . . . . 375 Summary . . . . 376 References . . . . 376 30. Venous Thromboembolism . . . . . . . . 379 Yoram Klein and Enrique Ginzburg Deep Venous Thrombosis . . . . 379 Pulmonary Embolism . . . . 382 References . . . . 386 PART VI: Complications of Trauma 31. Epidemiological, Organizational, and Educational Aspects of Trauma Care . . . . . . . . . . . . . . . . . . . 389 Michael E. Ivy Background . . . . 389 Trauma Systems . . . . 390 Prehospital Care . . . . 391 Hospital Care . . . . 393 Rural and Urban Trauma Centers . . . . 395 Triage in the Hospital . . . . 396 Management of Pediatric Trauma . . . . 396 Prevention of Trauma . . . . 396 References . . . . 397 32. Competing Priorities in the Trauma Patient . . . . . . . . . . . . . . . . . . 399 Michael E. Ivy Brain and Abdominal Trauma . . . . 399 Injuries to the Thoracic Aorta and the Abdomen . . . . 402 Timing of Fracture Fixation for Patients with Brain Injuries . . . . 403 Management of Long-Bone Fractures and Pulmonary Injuries . . . . 404 References . . . . 405 33. Complications of Fractures . . . . . . . . 407 Bar Ziv Yaron, Kosashvili Yona, Gelfer Yael, and Halperin Nahum Acute Complications . . . . 408 Late Complications . . . . 414 Summary . . . . 415 References . . . . 415

xiv

Contents

34. Complications of Dislocations . . . . . . 419 Howard Richter and Gregory A. Zych Vascular Injuries Associated with Dislocation . . . . 419 Neural Injuries Associated with Dislocation . . . . 420 Avascular Necrosis After Dislocation . . . . 420 Heterotopic Bone Formation after Dislocation . . . . 421 Posttraumatic Arthritis After Dislocation . . . . 421 Musculotendinous Injury After Dislocation . . . . 421 Instability After Dislocation . . . . 421 Joint Stiffness After Dislocation . . . . 422 References . . . . 422 35. Complications of Amputations . . . . . 423 Yoram Klein and Mauricio Lynn General Considerations . . . . 423 Preoperative Considerations . . . . 423 Intraoperative Considerations . . . . 424 Postoperative Complications . . . . 425 Mortality . . . . 427 Summary . . . . 427 References . . . . 427 36. Complications of Hand Surgery . . . . 429 Charles Eaton Complications of Missed Diagnoses . . . . 429 Complications of Treatment . . . . 432 Complications of Injury . . . . 437 References . . . . 460 PART VII: Neurosurgical Complications 37. Postoperative Pain Management . . . . 463 Edward Lubin, Michael J. Robbins, and Raymond S. Sinatra Patient and Caregiver Variables Influencing Perioperative Analgesic Effect . . . . 463 Anatomy and Pathophysiology . . . . 465 Pain Services and Therapeutic Options for Postoperative Analgesia . . . . 466 Parenterally and Orally Administered Analgesics . . . . 467 Intravenous Patient-Controlled Analgesia . . . . 469

Spinal Opioid Analgesia . . . . 469 Adverse Events and Contraindications of Neuroaxial Analgesic Techniques . . . . 470 Neural Blockade for Acute Pain Management . . . . 471 Posttraumatic Pain . . . . 471 New Ideas in Pain Management . . . . 472 Pain Control and Postsurgical Outcome . . . . 472 Chronic Pain . . . . 473 References . . . . 479 38. Complications After Craniotomy . . . . 483 Andrew Jea and Nizam Razack Mass Lesions . . . . 483 Infection . . . . 486 Infarctions . . . . 487 Metabolic Imbalances . . . . 488 References . . . . 489 39. Spinal Cord Trauma . . . . . . . . . . . . . 491 Michael Y. Wang, Iftikharul Haq, and Barth A. Green Epidemiology . . . . 491 Biomechanics of Injury . . . . 491 Pathophysiology . . . . 492 Clinical Features . . . . 492 Evaluation of Suspected Injury . . . . 493 Acute Management . . . . 494 Pharmacologic Therapy for Spinal Cord Injury . . . . 496 Surgical Management . . . . 496 Prevention of Complications . . . . 496 Spinal Cord Injury in Children . . . . 497 Future Advances . . . . 498 References . . . . 498 40. Complications of Nerve Injury and Repair . . . . . . . . . . . . . . . . . . . . . 501 Patrick W. Owens Diagnosis of Nerve Injuries . . . . 501 Anatomy . . . . 501 Classification of Nerve Injuries . . . . 502 Complications of Nerve Repair . . . . 502 Injuries Associated with Nerve Injuries . . . . 503 Injuries to Nerves from Operations or Other Medical Procedures . . . . 504 References . . . . 507

Contents

41. Psychological and Behavioral Complications of Trauma . . . . . . . . . 511 Thomas Mellman and Gillian Hotz Traumatic Brain Injury . . . . 511 Posttraumatic Stress Disorder . . . . 515 References . . . . 517 PART VIII: Wound and Soft Tissue Complications 42. Complications of Wound Repair . . . . . . . . . . . . . . . . . . . . . . . . 521 Nicole S. Gibran and F. Frank Isik Continuum of Normal Wound Repair Processes . . . . 521 Types of Wounds . . . . 523 Wound Closure and Coverage . . . . 523 Surgical Technical Principles . . . . 523 Risk Factors for Abnormal Wound Repair . . . . 524 Surgical Tissue Injury . . . . 526 Specific Examples of Wound Complications . . . . 528 References . . . . 529 43. Complications of Thermal Injuries . . . . . . . . . . . . . . . . . . . . . . . 531 Mark Cockburn and Nicholas Namias Complications of Burn Injuries . . . . 531 Complications of Electrical Injuries . . . . 535 References . . . . 536 44. Complications of Skin Grafting . . . . 539 Raquel Garcia-Roca, David S. Lasko, and Nicholas Namias Types of Skin Grafts . . . . 539 Graft-Wound Healing . . . . 539 Techniques of Skin Grafting . . . . 540 Complications of Skin Grafting . . . . 540 Donor-Site Complications . . . . 544 Special Issues with Scalp Donor Sites . . . . 544 References . . . . 544 45. Complications of Reconstructive Surgery . . . . . . . . . . . . . . . . . . . . . . . 547 D. Narayan, J. H. Shin, and J. A. Persing Cutaneous Flaps . . . . 547 Mathes–Nahai Classification . . . . 547 Pedicled Flaps . . . . 548

xv

Free Microvascular Flaps . . . . 549 Complications of Reconstructive Surgery: Maxillofacial Surgery . . . . 550 References . . . . 551 Further Readings . . . . 551 PART IX: Genitourinary Surgery Complications 46. Complications in Gynecologic Surgery . . . . . . . . . . . . . . . . . . . . . . . 553 Emery M. Salom and Manuel Penalver Abdominal Gynecologic Surgery . . . . 553 Vaginal Surgery . . . . 559 Hysteroscopy . . . . 564 Laparoscopy . . . . 565 Radical Pelvic Surgery for Gynecologic Malignancy . . . . 566 Summary . . . . 569 References . . . . 569 47. Complications of Bladder and Prostate Surgery . . . . . . . . . . . . . . . . . 573 Adam J. Bell, Josh M. Randall, and Raymond J. Leveillee Bladder Cancer . . . . 573 Bladder Augmentation . . . . 579 Bladder Stones . . . . 581 Vesicoureteral Reflux and Ureteroceles . . . . 582 Urinary Incontinence . . . . 583 Laparoscopy . . . . 584 Bladder Trauma . . . . 585 Conclusions . . . . 587 References . . . . 587 48. Urethral, Scrotal, and Penile Surgery . . . . . . . . . . . . . . . . . . . . . . . 591 Angelo E. Gousse, Robert R. Kester, and Patricia M. Byers Urethra . . . . 591 Scrotum . . . . 597 Testicle . . . . 598 Epididymis and Vas Deferens . . . . 599 Penis . . . . 599 Summary . . . . 601 References . . . . 601 49. Complications of Genitourinary Trauma . . . . . . . . . . . . . . . . . . . . . . . . 605 Yekutiel Sandman and Peter P. Lopez Renal . . . . 605

xvi

Contents

Ureteral . . . . 610 Bladder . . . . 612 Urethra . . . . 616 Male Reproductive Organs . . . . 621 References . . . . 622 50. Surgical Complications of Kidney–Pancreas Transplantation . . . 625 Gaetano Ciancio, Joshua Miller, George W. Burke, and Patricia M. Byers Transplantation . . . . 625 Surgical Complications After Pancreas Transplantation . . . . 626

Summary . . . . 632 References . . . . 632 51. Complications of General Surgery During Pregnancy . . . . . . . . . . . . . . . 637 Raymond P. Compton General Principles of Management . . . . 637 Specific Disorders in Pregnant Women . . . . 639 Summary . . . . 643 References . . . . 643 Index . . . . 645

Contributors

Ha˚kin Ahlman Gothenberg University, Sahlgrenska Universitetsjukhuset, Goteborg, Sweden

University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Fotios M. Andreopoulos Departments of Surgery and Biomedical Engineering, University of Miami School of Medicine, Miami, Florida, U.S.A.

Miguel A. Cobas Division of Trauma Anesthesia and Critical Care, Department of Anesthesiology, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Robert W. Bailey Division of Laparoscopic and Bariatric Surgery, Daughtry Family Department of Surgery, Miller School of Medicine, University of Miami, Miami, Florida, U.S.A. Erik Barquist Ryder Trauma Center, Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A. Adam J. Bell Division of Endourology and Laparoscopy, Department of Urology, University of Miami Miller School of Medicine, Miami, Florida, U.S.A.

Mark Cockburn Department of Surgery, Miller School of Medicine at the University of Miami, Miami, Florida, U.S.A. Raymond P. Compton Paris Surgical Specialists, Paris, Tennessee, U.S.A. Victor Cruz Department of Surgery, Stony Brook University Hospital, Stony Brook, New York, U.S.A.

Edward M. Boyle Division of Cardiothoracic Surgery, University of Washington, Seattle, Washington, U.S.A.

James C. Doherty Division of Trauma Surgery, Advocate Christ Medical Center, Oak Lawn, and Department of Surgery, University of Illinois College of Medicine at Chicago, Chicago, Illinois, U.S.A.

Stephen Brower Department of Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A.

Matthew O. Dolich Division of Trauma and Surgical Critical Care, University of California, Irvine School of Medicine, Irvine, California, U.S.A.

Robert F. Buckman, Jr. Department of Surgery, Temple University School of Medicine, Philadelphia, Pennsylvania, U.S.A.

Charles Eaton U.S.A.

George W. Burke Division of Transplantation, The DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

The Hand Center, Jupiter, Florida,

Akpofure Peter Ekeh Department of Surgery, Wright State University School of Medicine, Dayton, Ohio, U.S.A. Mohamed Fahim Anesthesia Critical Care, Davis Memorial Hospital, Elkins, West Virginia, U.S.A.

Patricia M. Byers Division of Trauma, Burns, and Critical Care, The DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Debra Fertel Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Miami Miller School of Medicine, Miami, Florida, U.S.A.

Denise M. Carneiro-Pla DeWitt Daughtry Family Department of Surgery, Miller School of Medicine, University of Miami, Miami, Florida, U.S.A.

Raquel Garcia-Roca University of Miami School of Medicine/Jackson Memorial Hospital, Miami, Florida, U.S.A.

Gaetano Ciancio Division of Transplantation, The DeWitt Daughtry Family Department of Surgery,

Nicole S. Gibran Department of Surgery, University of Washington, Seattle, Washington, U.S.A.

xviii

Contributors

Enrique Ginzburg Division of Trauma and Surgical Critical Care, DeWitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A. Amy J. Goldberg Department of Surgery, Temple University School of Medicine, Philadelphia, Pennsylvania, U.S.A. Dean Goldberg Department of Surgery, Lee Memorial Hospital, Fort Myers, Florida, U.S.A. Angelo E. Gousse Department of Urology, Miller School of Medicine at the University of Miami, Miami, Florida, U.S.A. Barth A. Green Department of Neurological Surgery & The Miami Project to Cure Paralysis, University of Miami School of Medicine, Miami, Florida, U.S.A. Fahim A. Habib Division of Trauma and Surgical Critical Care, DeWitt Daughtry Family Department of Surgery, University of Miami School of Medicine, Miami, Florida, U.S.A. Allen D. Hamdan Harvard Medical School and Division of Vascular Surgery, Beth Israel Deaconess Medical Center, Boston, Massachusetts, U.S.A. Iftikharul Haq Department of Neurological Surgery & The Miami Project to Cure Paralysis, University of Miami School of Medicine, Miami, Florida, U.S.A. John J. Hong Robert Wood Johnson Medical School, University of Medicine and Dentistry of New Jersey, New Brunswick, New Jersey, U.S.A. Gillian Hotz Department of Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A. Kenji Inaba Department of Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A. George L. Irvin III DeWitt Daughtry Family Department of Surgery, Miller School of Medicine, University of Miami, Miami, Florida, U.S.A. F. Frank Isik Department of Surgery, University of Washington, Seattle, Washington, U.S.A. Michael E. Ivy Hartford Hospital, Hartford, and University of Connecticut School of Medicine, Farmington, Connecticut, U.S.A. Andrew Jea Department of Neurological Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A.

Igor Jeroukhimov Department of Surgery, Assaf Harophe Medical Center, Tel Aviv University, Zerifin, Israel Dory Jewelewicz Department of Anesthesiology, University of Miami/Jackson Memorial Hospital, Miami, Florida, U.S.A. Richard J. Kaplon Cardiac and Thoracic Surgery Medical Group, Sacramento, California, U.S.A. Riyad C. Karmy-Jones Division of Cardiothoracic Surgery, University of Washington, and Thoracic Surgery, Harborview Medical Center, Seattle, Washington, U.S.A. Rosemary F. Kelly Cardiovascular and Thoracic Surgery, University of Minnesota, Minneapolis, Minnesota, U.S.A. Robert R. Kester Department of Urology, Miller School of Medicine at the University of Miami, Miami, Florida, U.S.A. Yoram Klein Division of Trauma and Emergency Surgery, Kaplan Medical Center, Rehovot and Department of Surgery, Hadassah EIN, Kerem Medical Center, Jerusalem, Israel David S. Lasko University of Miami School of Medicine/Jackson Memorial Hospital, Miami, Florida, U.S.A. Raymond J. Leveillee Division of Endourology and Laparoscopy, Department of Urology, University of Miami Miller School of Medicine, Miami, Florida, U.S.A. David M. Levi Department of Surgery, University of Miami, Miami, Florida, U.S.A. Peter P. Lopez Division of Trauma and Surgical Critical Care, DeWitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A. Louis B. Louis IV Division of Cardiothoracic Surgery, DeWitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine, Jackson Memorial Hospital, Miami, Florida, U.S.A. Edward Lubin Department of Anesthesiology, Yale University School of Medicine, New Haven, Connecticut, U.S.A. Mauricio Lynn Division of Trauma and Surgical Critical Care, DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Contributors

xix

Jana B. A. MacLeod Division of Trauma and Critical Care, Department of Surgery, Emory University School of Medicine, Atlanta, Georgia, U.S.A.

James B. Peoples Department of Surgery, Wright State University School of Medicine, Dayton, Ohio, U.S.A.

Sapoora Manshaii St. Louis University School of Medicine, St. Louis, Missouri, U.S.A.

J. Martin Perez Robert Wood Johnson University Hospital, New Brunswick, New Jersey, U.S.A.

Hussein Mazloum Department of Surgery, McLaren Regional Medical Center and the College of Human Medicine–Michigan State University, Flint, Michigan, U.S.A.

J. A. Persing Section of Plastic Surgery, Yale University School of Medicine, New Haven, Connecticut, U.S.A.

Scott McDonald DeWitt Daughtry Family Department of Surgery, University of Miami School of Medicine, Miami, Florida, U.S.A.

Si M. Pham Division of Cardiothoracic Surgery, DeWitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine, Jackson Memorial Hospital, Miami, Florida, U.S.A.

Thomas Mellman Dartmouth Hitchcock Medical Center, Dartmouth Medical School, Lebanon, New Hampshire, U.S.A.

Louis R. Pizano DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Joshua Miller Division of Transplantation, The DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Frank B. Pomposelli, Jr. Harvard Medical School and Division of Vascular Surgery, Beth Israel Deaconess Medical Center, Boston, Massachusetts, U.S.A.

Frederick L. Moffat DeWitt Daughtry Family Department of Surgery, University of Miami School of Medicine, Miami, Florida, U.S.A.

Juan Carlos Puyana Surgical/Trauma Intensive Care Unit, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, U.S.A.

John C. Mullen Department of Cardiac Sciences, University of Alberta, Edmonton, Canada

Xiao-Shi Qi Division of Cardiothoracic Surgery, DeWitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine, Jackson Memorial Hospital, Miami, Florida, U.S.A.

Halperin Nahum Department of Orthopedic Surgery, Tel Aviv University, Sackler Faculty of Medicine, Assaf Harofeh Medical Center, Zeriffin, Israel Nicholas Namias Miller School of Medicine at the University of Miami, Miami, Florida, U.S.A. D. Narayan Section of Plastic Surgery, Yale University School of Medicine, New Haven, Connecticut, U.S.A. Adrian W. Ong Department of Surgery, Allegheny General Hospital, Drexel University College of Medicine, Pittsburgh, Pennsylvania, U.S.A.

Josh M. Randall Division of Endourology and Laparoscopy, Department of Urology, University of Miami Miller School of Medicine, Miami, Florida, U.S.A. Qammar Rashid Department of Surgery, Howard University, Washington, D.C., U.S.A. Nizam Razack Department of Neurological Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A.

Patrick W. Owens Department of Orthopedics and Rehabilitation, University of Miami, Miami, Florida, U.S.A.

Howard Richter Department of Orthopedics and Rehabilitation, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Abhijit S. Pathak Department of Surgery, Temple University School of Medicine, Philadelphia, Pennsylvania, U.S.A.

Michael J. Robbins Department of Anesthesiology, Yale University School of Medicine, New Haven, Connecticut, U.S.A.

Manuel Penalver Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Sylvester Comprehensive Cancer Center, University of Miami/ Jackson Memorial Medical Center, Miami, Florida, U.S.A.

Emery M. Salom Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Sylvester Comprehensive Cancer Center, University of Miami/ Jackson Memorial Medical Center, Miami, Florida, U.S.A.

xx

Contributors

Yekutiel Sandman Department of Urology, Jackson Memorial Medical Center, University of Miami Miller School of Medicine, Miami, Florida, U.S.A. Romualdo J. Segurola, Jr. Cardiovascular and Thoracic Surgery, University of Minnesota, Minneapolis, Minnesota, U.S.A. Gennaro Selvaggi Department of Surgery, University of Miami, Miami, Florida, U.S.A. Christopher K. Senkowski Mercer University School of Medicine, Memorial Health University Medical Center, Savannah, Georgia, U.S.A. Malachi G. Sheahan Division of Vascular Surgery, Louisiana State University School of Medicine, New Orleans, Louisiana, U.S.A. J. H. Shin Section of Plastic Surgery, Yale University School of Medicine, New Haven, Connecticut, U.S.A. Guatam V. Shrikhande Department of Surgery, Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, Massachusetts, U.S.A.

Andreas G. Tzakis Department of Surgery, University of Miami, Miami, Florida, U.S.A. Don H. Van Boerum Section of Trauma Surgery, Department of Surgery, Sutter Roseville Medical Center, Roseville, California, U.S.A. Albert J. Varon Division of Trauma Anesthesia and Critical Care, Department of Anesthesiology, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A. Michael Y. Wang Department of Neurological Surgery & The Miami Project to Cure Paralysis, University of Miami School of Medicine, Miami, Florida, U.S.A. Gelfer Yael Department of Orthopedic Surgery, Tel Aviv University, Sackler Faculty of Medicine, Assaf Harofeh Medical Center, Zeriffin, Israel Bar Ziv Yaron Department of Orthopedic Surgery, Tel Aviv University, Sackler Faculty of Medicine, Assaf Harofeh Medical Center, Zeriffin, Israel

Raymond S. Sinatra Department of Anesthesiology, Yale University School of Medicine, New Haven, Connecticut, U.S.A.

Kosashvili Yona Department of Orthopedic Surgery, Tel Aviv University, Sackler Faculty of Medicine, Assaf Harofeh Medical Center, Zeriffin, Israel

Danny Sleeman DeWitt Daughtry Family Department of Surgery, Miller School of Medicine, University of Miami, Miami, Florida, U.S.A.

Gregory A. Zych Department of Orthopedics and Rehabilitation, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

The Surgical Mortality and Morbidity Review

INTRODUCTION We surgeons establish a delightfully rewarding, gratifying, and unique bond with our patients. For our medical colleagues, the stakes are totally different. Without much trepidation, we, the authors, would permit Helen Keller to take our blood pressure or even palpate our edematous ankles. Similarly, if our health maintenance organization (HMO) relegated our primary care to someone with limited intellectual capacity, we would risk a visit or two before complaining. But when a surgeon suggests that we must go to sleep while he or she clamps our aorta or removes our cancerous tumor, we demand dazzling expertise and total commitment. We may not be able to accurately assess the former (the diplomas on the wall are not sufficient), but we can sense the latter. From our surgeon, we expect knowledge, experience, and technical proficiency; equally important, if things do not go well, we want our surgeon to hurt as much as we do. That is commitment!

chart clearly states that the patient has diabetes or is being treated with drops for glaucoma, but the anesthesiologist misses this statement. That is a system-specific problem. To the conscientious surgeon, any problem is always his or her fault. Whenever we, as surgeons, blame the hospital administrator because the ceiling collapsed or blame ‘‘patient disease’’ because our diabetic patient experienced a wound infection, that is a cop-out. We must always strive to eliminate all trouble for our patients. Every mishap is always avoidable and every problem that does occur is our fault. This philosophy greatly simplifies the adjudication of the mortality and morbidity process. This philosophy also defines hurt. If you want to be God, you must accept responsibility. God controls everything. It follows logically that you receive credit for both the good and the bad. If the bad is your fault, it must hurt. Good surgeons hurt—a lot.

WHO IS RESPONSIBLE? WHAT IS HURT? Physical pain is easy to describe and define. We have all stubbed a toe or bruised an elbow. Calibrating pain is more complex. We have all seen the rancher from Wyoming who brings his traumatically amputated lower limb to the surgeon by bus, and the businessman from New York City who requires morphine for a haircut. Vicarious hurt, on the other hand, may be unique to humans and is more difficult to explain. In its simplest form, the morbidity and mortality conference examines misadventures; dissects the preoperative, intraoperative, and postoperative events related to them; and derives strategies to prevent their recurrence. The varieties of surgical error are comprehensively explored in this book. Some are ‘‘surgeon specific’’: we may commit an error in the patient’s diagnosis, in our surgical technique, or in the perioperative management of the case. Occasionally, these errors are readily apparent: the appendix was normal or the anastomosis leaked. Some problems, however, are ‘‘system specific’’: the Monday morning anesthesiology conference typically runs later than 8:00 A.M. On Mondays, when the rushed anesthesiologists exit their conference, their tardiness prompts an abbreviated preoperative assessment of the patient. The

Ultimately, our goal is to make our patients and their families feel better. If we can accomplish this by virtue of our superior comprehension of some subcellular mechanism of disease or by means of some particular therapy, so much the better. We infuse a phosphodiesterase inhibitor to prevent cyclic-adenosine monophosphate degradation and thus to build the contractile strength of cardiomyocytes. We do this because we know that, with age and congestive failure, patients deplete their cardiac beta-adrenergic receptors, and thus we must resort to a different inotropic strategy. That is great! We can tuck our thumbs into our axillae and strut off to the next lucky patient with the full knowledge that this cardiac cripple is incredibly fortunate to have a surgeon with our degree of omniscience. Conversely, after our patient tiptoes his or her way through the surgical intensive care unit minefield, some well-meaning cardiology fellow or surgical intensive care unit nurse tells the family that Uncle Andy ‘‘almost died’’ and ‘‘really has a bag for a heart.’’ The family (and Uncle Andy) is justifiably delighted to be discharged eventually from the surgical intensive care unit (and the hospital). But, in reality, Uncle Andy arrives home on pins and needles, expecting to die at any moment, for he is now

xxii

The Surgical Mortality and Morbidity Review

obsessed with the knowledge of the horrible condition of his heart. That is a complication! We can have a civil discussion about who is responsible for this travesty of care, but we believe that it is the surgeon. From the moment that a surgical diagnosis is even suggested by the patient, his or her mother-in-law, or the medical consultant, the surgeon must enthusiastically accept responsibility for the pathophysio-psycho-social outcome of his or her surgical endeavors. Transfecting a responsive insulin receptor into an unsuspecting hepatocyte for the purpose of glorifying dysfunctional carbohydrate metabolism is nearly miraculous. Accomplishing this while sacrificing Uncle Andy’s comprehension and adding to the confusion of the family is a travesty of surgical or medical skill. Our ultimate surgical goal is to make the family understand that Uncle Andy really does feel better.

ARE DEATH AND DISEASE INEVITABLE? As conscientious, sensitive, thoughtful, compassionate surgeons, we must not accept the obvious answer to this question of whether death and disease are unavoidable (1). The surgical mortality and morbidity conference is unique in medicine and probably in the civilized world. Short of bank robbery, few events are as independently attributable to their perpetrator as is a surgical procedure. The surgeon meets the patient and family. He or she describes the pathological problem and the proposed surgical solution. If trust and commitment are not immediately emblazoned into this initial interaction, the whole process is fortunate to achieve junior bush league status. If the diagnosis

is omniscient, the surgical technique flawless, the perioperative care masterful, and the patient expeditiously returned to the position of a socially responsible contributor to society, but the patient or the family does not comprehend the process, then the surgeon has failed. Open discussion of this glorious spectrum of therapeutic opportunities is the purpose of a constructively educational surgical mortality and morbidity conference. When we, as surgeons, welcome this amplitude of criticism, we confidently bare our souls in the knowledge that no matter how good we are, we welcome any sacrifice to be even better. And after every patient we are privileged to treat and after every mortality and morbidity conference, we should look ourselves in the mirror, acknowledging that we may not know what profession is second best but that we are incredibly fortunate to be members of the most gratifying, responsible, receptive, critical, rewarding, and fun guild that exists (Polk H, Guandlick S, personal communication, 2002). Julie Heimbach, M.D. Mayo Clinic, Rochester, Minnesota, U.S.A. Jyoti Arya, M.D. University of Colorado, Denver, Colorado, U.S.A. Alden H. Harken, M.D. University of California, San Francisco, and UCSF-East Bay Surgery Program, Oakland, California, U.S.A.

REFERENCE 1. Harken AH. Enough is enough. Arch Surg 1999; 134:1061–1063.

The Surgical Mortality and Morbidity Review: Best Practices and Procedures

The mortality and morbidity conference is an exercise unique to surgeons among all the members of the medical profession. Our colleagues in the medical specialties do not organize such conferences and rarely publicly probe their practices for systematic or individual error. Because surgeons collect data on mortality and morbidity and discuss such events regularly, surgical practice is most frequently analyzed by quality assurance staff. The quality assurance mavens are simply being lazy. They would provide a much greater return, in terms of overall improvement in patient care, if they invested their effort among those, primarily nonsurgeons, who do not engage in regular analysis of potential error. Sad to say, the mortality and morbidity conference in some institutions I have had the privilege to visit sometimes fails to meet the objective of identifying error so that it may be corrected. Failure most frequently is due to lack of any real discussion in depth about the details of complications and deaths. Cases of senior or powerful attending physicians are not brought up for discussion, or details of errors in such cases are glossed over. No one utters critical remarks; moderators shirk their responsibility to probe. The conference is a sham. Even worse, in some institutions, the mortality and morbidity conference is converted into a presentation of ‘‘interesting’’ cases—an exercise better reserved for Grand Rounds—and investigation of the potential for prevention of deaths and complications does not occur. If a mortality and morbidity conference is to achieve its desired goal, all cases resulting in any deviation, however minor, from the desired and expected outcome must be identified and be eligible for discussion. There can be no exceptions. Additionally, the conference must be appropriately structured; attendance without exception must be required of all staff, residents, and students assigned to the service or department. The conference must be led by moderators who understand and accept that it is their duty to probe for errors and for the causes of untoward outcomes. The conference should be open to nurses and other members of the hospital staff interested in the presentations and discussions. To enhance effective discussion and sharing of experience, both the number of cases eligible for review and the size of the audience participating in

the discussion should be neither too large nor too small. Between 30 and 50 is about the right number for both parameters. If necessary, because of the great workload of the service or department, multiple mortality and morbidity conferences should be conducted by subunits. The conference should be conducted weekly, be limited to one hour in duration so that the attention of the audience does not wander, start on time, and end on time. The room in which the mortality and morbidity conference is held should be large enough to accommodate the audience comfortably. There should be facilities for displaying X rays—about eight large films simultaneously—and equipment to project occasional pertinent slide or computer illustrations also should be available. It is not necessary, in my view, to have a pathologist or radiologist in regular attendance. These specialists tend to spend too much time on demonstration of details that are of great interest within their medical niche but do not quickly advance the point of the mortality and morbidity conference: determination of the accuracy and effectiveness of the diagnosis or treatment in the case under discussion. An agenda listing pertinent statistical and case information should be distributed at the entrance to the meeting room. No information identifiable with a specific patient should be included in the agenda. Material presented at mortality and morbidity conferences is usually protected by quality assurance privileges and regulations. Nonetheless, all copies of the agenda should be collected and destroyed at the end of the conference. The agenda should include all material for a fixed time period; from eight o’clock Sunday morning to the same time on the following Sunday morning is a convenient interval. The agenda should list the following statistics: numbers of admissions, discharges, open operations, closed (e.g., laparoscopic) procedures, other procedures, complications, and deaths. This statistical information should be followed by a morbidity summary that separately lists each instance of an unexpected or untoward outcome in outline fashion (e.g., wound infection; 64F, colectomy, drained sixth pod). The sketchy information serves simply to remind the audience of the patient involved. The morbidity data are followed by a mortality summary in which each death is individually listed, also in shorthand

xxiv

The Surgical Mortality and Morbidity Review: Best Practices and Procedures

fashion (e.g., 64F, colectomy; stroke fourth pod; pneumonia; died 10th pod; cause: MOF; no autopsy). The assignment as moderator should be rotated among several of the most junior members of the staff. These persons are more likely to be up to date in their knowledge base. The moderator chooses from among the cases listed on the agenda those for presentation and discussion, emphasizing cases with unexpected outcomes, teaching value, rarity, etc. The moderator must maintain control of the conference, insisting that presentations be succinct, discussions be pertinent, and no witch-hunting be practiced by the audience. Knowledgeable members of the staff should be called upon for comment, especially if it is likely they will dissent. Residents should be liable to be called upon at any time—a device that helps to keep them awake. The moderator should generate any required minutes or reports soon after the conference has been completed. Presentations, as requested by the moderator, are made by the resident team involved. All members of the team should stand at the front of the room. Usually a student or intern quickly summarizes the clinical course up to the time of operation or other salient event. These presentations should be rehearsed so that they are short and succinct, no more than a minute or two in length, even for a complicated case. Because all cases are eligible for discussion, all will have to be prepared. Pertinent X rays should be put up during the initial presentation for viewing by the audience; the X rays must be sorted in advance because there is no time within an efficiently conducted mortality and morbidity conference for rummaging through an X-ray file. The presentation is next taken up and completed, beginning with the operation and continuing through the postoperative course, by the most senior resident involved in the procedure. I advise residents to read Forgive and Remember by Bosk (1) and to identify and follow the 15 rules of successful resident behavior identified therein. If they do so, they will acknowledge their errors during their presentation, thus anticipating reaction from the audience and controlling the discussion at the mortality and morbidity conference. If they do not, their experience at mortality and morbidity will sometimes be unnecessarily unpleasant. I made it a rule that if residents had not disagreed at the time with the attending physician about

any decision or other matter, then they carried responsibility for the outcome and had to conduct the presentation and answer questions from the audience. On the other hand, if they had clearly established their position of dissent, all they had to do was say so; the attending physician then had to come forward to complete the presentation and handle the discussion. After the completed presentation, the moderator may ask the members of the audience whether they have questions that might clarify details of the presentation. Then the moderator initiates discussion of alternatives and possible error. If their presentation has been conducted as it should be, the residents will have left no question unanswered and the conference will simply move on to the next case. The mortality and morbidity conference should be the best teaching exercise conducted by a surgical service or department. The goal is to have an open, thorough, detailed discussion of untoward outcomes so that all may learn from these events with the aim of improving the excellence of patient care. Properly conducted, the mortality and morbidity conference will, at least sometimes, be an uncomfortable exercise for some participants because no one, save an intellectual masochist, relishes having his or her potential lapses discussed in public. But, if properly conducted in a spirit of open intellectual inquiry, the mortality and morbidity conference will serve to enhance the professional knowledge and conduct of all participants and will help to avoid repetitive error. It is, therefore, ‘‘worth the price’’ of occasional embarrassment. Robert E. Condon, M.D., M.Sc, FACS Department of Surgery, The Medical College of Wisconsin, Milwaukee, Wisconsin, and The University of Washington Medical School, Seattle, Washington, U.S.A.

REFERENCE 1. Bosk CL. Forgive and Remember: Managing Medical Failure. 2nd ed. Chicago: University of Chicago Press, 2003.

PART I Complications in the Intensive Care Unit

1 Complications of Anesthesia Miguel A. Cobas and Albert J. Varon Division of Trauma Anesthesia and Critical Care, Department of Anesthesiology, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

The study of complications that follow the administration of anesthesia is challenging because their occurrence is infrequent and they are influenced by a number of different factors attributable to the patient, the surgical procedure, or the method or agent of anesthesia. Historically, there have been two common methods of studying complications in anesthesia. The first method is evaluation of cases, preferably prospectively, albeit many studies have been done retrospectively with a large cohort of cases. However, because the incidence of major complications in modern anesthesia is very low, an enormous amount of data set would be required to provide the study with sufficient power to detect significant differences. The second method is to study a single complication in an attempt to detect a pattern associated with the occurrence of that complication. Many authors have tried to elucidate the real risk of death for anesthetized patients. In the 1950s, Beecher and Todd’s (1) landmark study of more than 500,000 anesthetics concluded that anesthesia was a contributory factor in mortality in 1:560 cases, and was the primary cause of death in 1:2680 cases. The determination that an important contribution to the mortality rate was residual paralysis by curare was noteworthy. Since this study by Beecher and Todd, many others have attempted to determine the mortality rate associated with anesthesia. Anesthesia was found to be the primary cause of death in 1 of every 852 cases (2) to 1 in every 14,075 cases (3). A significant number of fatalities occurred in the patient’s hospital room, emphasizing the need for postoperative monitoring. In fact, in most of the studies from the 1960s, the main safety recommendation was the creation of a postanesthesia-care unit. Other anesthesia complications commonly cited as important causes of mortality were aspiration, errors in airway management, hypovolemia, and overdose of anesthetic agents. These studies also showed that the highest mortality rates occurred among elderly patients, those with preexisting diseases, and those who required emergency surgery. In the 1980s, the landscape of anesthesia started to change across the world. At that time the work force of anesthesia in most developed countries consisted of a trained anesthetist, in most cases a physician. In the United States,

the concept of the anesthesia-care team, in which a physician supervises a Certified Nurse Anesthetist, was consolidated. The advent of newer and safer anesthetic drugs and improvements in patient monitoring, such as the widespread use of pulse oxymetry, and a few years later, capnography, lead to a dramatic reduction in mortality rates (3). Holland studied over one million anesthetics spanning from 1960 to 1985. According to this study, the incidence of death attributed to anesthesia decreased from 1:550 in 1960 to 1:10,250 in 1970, and to 1:26,000 in 1984 (4). The shortcomings of this article were its retrospective design and the fact that the authors only reviewed mortality within the first 24 hours after anesthesia. Perhaps the most cited study in anesthesia mortality is the Confidential Enquiry into Perioperative Death (CEPOD) (5). The authors reported the results of over 500,000 cases in three areas of the United Kingdom over a period of one year. The report included deaths that occurred within 30 days after the anesthetic was administered, and found that anesthesia was solely responsible for death in three patients, or in about one death for every 185,000 cases. It is worth noting that the criteria for adjudicating a death exclusively to anesthesia were stricter in the CEPOD study than in other studies, including those performed by the same authors (6,7). In 1984, the American Society of Anesthesiologists (ASA) embarked on a nationwide project to study all the closed claims related to anesthesia. The goal of the Closed Claims Project was to discover unappreciated patterns of anesthesia care that may contribute to patient injury and subsequent litigation (8). This database represents the largest resource for the study of adverse outcomes related to anesthesia care and provides a reliable assessment of common complications, their consequences, and the importance that the patient gives to them. According to this database, death, although extremely uncommon, remains the most common cause of lawsuits against anesthesiologists, but its percentual weight has declined from 56% in the 1970s to 32% in the 1990s. Emerging as an important liability cause for anesthesiologists is peripheral nerve injury, currently accounting for 16% of all claims.

2

Cobas and Varon

COMPLICATIONS OF AIRWAY MANAGEMENT Major Complications Respiratory events constitute the largest group of claims against anesthesiologists, and they are responsible for the largest proportion of deaths and brain injury settlements. There are three major mechanisms of injury that account for such devastating outcomes: inadequate ventilation, esophageal intubation, and difficult tracheal intubation. Esophageal intubation is neither uncommon nor disastrous when the patient’s airway is difficult to access, but failure to recognize this complication is lethal. It is interesting to note that in 48% of the cases of unrecognized esophageal intubation, breath sounds were auscultated and documented (9). A sustained end-tidal CO2 (ETCO2) is essential to confirming tracheal intubation because brief CO2 signals can sometimes be detected when the endotracheal tube is placed in the esophagus. Although the best way to definitely secure the airway is by means of endotracheal intubation, sometimes the rush to intubate makes the clinician forget that with a self-inflating bag and proper technique, it is possible to provide adequate oxygenation and ventilation in most occasions. Even when intubation is the ultimate goal, preoxygenation is extremely valuable. By denitrogenating the lungs and filling the functional residual capacity (FRC) with a high concentration of oxygen, the time the patient can tolerate apnea is greatly prolonged. It is important to develop a plan for securing the airway in the first attempt and a backup plan if the first option fails. The ASA has developed a

comprehensive algorithm for dealing with a potential difficult airway (Fig. 1). In this algorithm, great emphasis is placed on anticipation and preparation. A suspected difficult airway should ideally be secured with the patient awake and breathing spontaneously because taking away the native respiratory drive with the use of anesthetics and muscle relaxants could prove lethal. The most feared scenario occurs when the trachea cannot be intubated and mask ventilation is not effective. In this ‘‘can’t intubate, can’t ventilate’’ situation, the practitioner has four options: laryngeal mask airway (LMA), Combitube, transtracheal jet ventilation (TTJV), and cricothyroidotomy. The LMA and the Combitube are devices that are blindly inserted into the oropharynx and are designed to provide supraglottic oxygenation and ventilation, but do not protect against aspiration. Although their use is limited in the presence of laryngospasm or upper airway obstruction, they are extremely useful in rescue situations, while other options for definitively securing the airway are being contemplated or the patient is allowed to resume spontaneous breathing. The two other techniques recommended for the above scenario are infraglottic and require more skill and time. TTJV requires the availability of a highpressure oxygen source and tubing to deliver oxygen into the trachea through a large gauge (i.e., 14 or 15) catheter inserted in the cricoidthyroid membrane. Cricothyroidotomy is the preferred surgical airway in an emergency situation because the anatomic structures are easier to locate and dissect when compared with the traditional tracheostomy.

Difficult Airway Recognized

un

Proper Preparation

Unrecognized co o pa pe tie rat nt ive

General Anesthesia +/_ Paralysis

No Mask Ventilation emergency pathway Yes

LMA Combitube, TTJV Surgical Airway

Awaken

non-emergency pathway

Awake Intubation Choices

Succeed

Fail Surgical Airway

Regional Anesthesia Cancel Case, Regroup

Intubation Choices

Intubation Choices Succeed Fail

Confirm Surgical Airway Anesthesia w/ Mask Ventilation

Awaken

Extubate Over Jet Stylet

Figure 1 American Society of Anesthesiologists’ algorithm for the difficult airway. Abbreviations: LMA, laryngeal mask airway; TTJV, transtracheal jet ventilation. Source: From American Society of Anesthesiologists on management of the difficult airway.

Chapter 1: Complications of Anesthesia

Airway Trauma A recent review of airway trauma during anesthesia revealed that claims for this complication rank fourth, after death, brain damage, and nerve damage malpractice claims (10). Interestingly enough, this complication is more likely to occur among female patients than among male patients and among children than adults, and these differences are statistically significant. Difficult intubation played a significant role (38%) in all claims for airway trauma. When considering airway trauma related to perioperative care, the sites most commonly affected are the larynx, pharynx, esophagus, and trachea. Laryngeal injuries comprise about a third of all airway trauma complications reported in the Closed Claims Project database (8). This type of injury was not associated with difficult intubation. Most of the laryngeal injuries were found in short-term intubations. One-third of all laryngeal injuries resulted in vocal cord paralysis, 16% in granulomas, and 8% in arytenoid dislocation. Tracheal injuries represented 14% of all the airway complications. Most of these were associated with the creation of a tracheostomy tract in the setting of a lost airway. Forty percent of tracheal injuries involve lacerations that occur during intubation. A rare but potentially lethal complication of intubation is tracheobronchial disruption. A review of the literature reveals mostly case reports and small series. Risk factors implicated in tracheal rupture include old age, chronic obstructive pulmonary disease, and corticosteroid therapy. Factors related to the intubation may include the use of a stylet, especially if it protrudes at the tip of the endotracheal tube, and forceful entry of the tube into the trachea after a difficult intubation. The disruption usually occurs in the posterior membranous part of the trachea, and physical signs include respiratory distress, stridor, and subcutaneous emphysema. Radiographic confirmation usually reveals a balloon that has been overinflated in an attempt to reduce the air leak, and pneumomediastinum with or without pneumopericardium. Fiberoptic bronchoscopy is the best method to diagnose tracheobronchial disruption. There is considerable controversy regarding the appropriate treatment of this injury. While some authors advocate emergent repair, two recent case reports of conservative treatment in elderly patients suggest that this could be a feasible option (11–13). Injuries to the esophagus and the oropharynx tend to be more severe because pharyngeal and esophageal tears often result in abscesses or mediastinitis. These entities carry a high mortality rate when there is a delay in the diagnosis. Patients who have had difficult tracheal intubation should be observed carefully for signs of mediastinitis or retropharyngeal abscess.

Aspiration Aspiration is defined as the passage of stomach contents through the glottic opening, causing varying degrees of lung damage and potential death.

3

Risk factors for aspiration include emergency surgery, surgery in pregnancy as early as 10 weeks, small bowel obstruction, bleeding ulcers, and autonomic neuropathy delaying gastric emptying. The classic example of aspiration pneumonitis was described in pregnant patients undergoing general anesthesia for cesarean section. Later studies postulated that an amount of at least 0.4 mL/kg and a pH of less than 2.5 were major determinants of the severity of aspiration pneumonitis. An ounce of prevention is worth a pound of cure when presented with patients at risk. The most common strategies used for the prevention of aspiration include the administration of drugs designed to increase the pH of the aspirate. H2 blockers are an effective option; however, they must be given at least two hours before surgery for their effect to take place. When premedicating just before surgery, it is more effective to use a nonparticulate antacid such as sodium citrate. Metoclopramide has also been advocated to promote gastric emptying, but is not devoid of side effects. Cricoid pressure has been advocated as the single most effective measure for preventing aspiration during intubation. It is important to remember that the operator applying cricoid pressure should not discontinue the maneuver until verification of appropriate endotracheal intubation and cuff inflation. If a patient is deemed to be at risk of aspiration, it is recommended to delay extubation until the patient is awake and has protective airway reflexes. If aspiration does occur, treatment is usually supportive. Prompt bronchoscopy is only indicated for retrieval of aspirated particulate matter. Most of the time, the aspiration has no major consequences, and it will manifest itself as a mild hypoxemia, which can be treated with supplemental oxygen. The absence of symptoms in the two first hours after the event usually correlates with a benign course (14). In more severe cases, the patient can present signs of clinical pneumonitis, with wheezing, rales, and rhonchi, but this clinical picture, as well as its X-ray changes, can be delayed up to 12 to 24 hours. In some cases, the syndrome is so severe that it requires mechanical ventilation and positive end-expiratory pressure. Antibiotics are usually not recommended because most of the time the aspirated material is sterile. If there is worsening of the patient status three to five days after the event, bacterial contamination should be suspected, and therapy should be guided by culture data.

Laryngospasm Laryngospasm is defined as an involuntary contraction of the vocal cords resulting in airway obstruction. It is typically a phenomenon related to airway instrumentation and is described more frequently in the pediatric population. Environmental factors may play an important role in the development of laryngospasm. Lakshmipathy et al. studied

4

Cobas and Varon

310 healthy children and found a strong association between household smoking and development of laryngospasm under general anesthesia (15). One of the most common causes of laryngospasm is the contact of secretions, blood, or vomitus against the vocal cords, causing reflex closure. Clinical signs of laryngospasm include stridor and an increased respiratory effort with visible tugging at the suprasternal notch. Laryngospasm is usually short-lived and selflimited, and sometimes there is a valve-like effect of the vocal cords allowing some expiratory flow. The first line of treatment for laryngospasm is positivepressure ventilation delivered gently by face mask. In children, where short periods of apnea are associated with rapid desaturation, administration of a small dose of succinylcholine (0.25 mg/kg) will relax the vocal cords and allow ventilation. It is worth noting that if pharmacologic paralysis is not feasible, laryngospasm will eventually recede due to hypoxia and hypercarbia.

Minor Complications Temporomandibular Joint Injuries Injuries to the temporomandibular joint have a unique set of characteristics. Almost all of these injuries occur in previously healthy young females, and are not associated with prolonged or difficult intubation. The reason for this demographic profile is perhaps the fact that temporomandibular joint disease is more prevalent in young females (16). Dislocation of the mandible occurs when there is subluxation of the condylar process over the articular eminence of the temporomandibular joint and, characteristically, the patients cannot bring their teeth together. This complication may not be apparent during anesthesia, but will manifest with severe pain and consequent muscle spasm over the temporomandibular joint region. The treatment for this condition consists of restoring the mandibular condyle back into its socket. Most of the time, this can be accomplished with the patient awake, after administration of a benzodiazepine to improve the muscle spasm. An operator then places the thumbs inside the mouth of the patient, over both the inferior molar surfaces, and then pushes downward and backward to snap the joint back into place. It is obvious that the operator must take great care to avoid damage to his or her thumbs because the patient’s teeth may close forcefully.

Dental Injury Dental damage is the most common type of injury during general anesthesia (17,18), with an incidence ranging from 0.01% to 0.1% (19,20). The upper central incisors are the teeth having the highest risk for dental injury, and damage is much more common if there is a

preexistent dental disease. Decidual teeth have shallow roots, which makes them prompt to dislodgment. Therefore, the best strategy to prevent this type of injury is a thorough dental evaluation before surgery. If there is extensive tooth decay, protruding teeth, loose teeth, or prosthesis, the anesthetist should inform the patient that there is a possibility of teeth being dislodged, and some practitioners would advocate the removal of very loose teeth while the patient is awake. If a tooth becomes dislodged during laryngoscopy, immediate retrieval from the oropharynx is mandated. Sometimes, X-rays of the neck, chest, and abdomen are necessary to locate the piece. If the tooth is located in the lung, a bronchoscopy should be performed promptly. The tooth should be reimplanted immediately when feasible, applying pressure over the piece; this will minimize the risk of root resorption. A dental consult should be requested for as soon as possible.

Sore Throat The reported incidence of sore throat after anesthesia varies widely, ranging from 6% to 40% (21). This transient minor complaint is not an airway injury, but should be regarded as an important outcome to be avoided for patient satisfaction. A sore throat is one of the most common and least desirable events for patients after anesthesia (22) and appears to be more frequent in women than in men (23). A study reported that sore throat and coughing on the first postoperative day were significantly less frequent in patients in the low pressure–high volume cuff group. This finding suggests that the pressure exerted to the tracheal mucosa is more important than the total contact area of the cuff. The authors, however, found no differences when comparing voice changes or difficulty in swallowing between the groups (24). Another factor implicated in the genesis of sore throat after anesthesia is the use of succinylcholine; however, this has not been substantiated (25,26). Inflating the endotracheal tube cuff with lidocaine or lubricating the outer surface of the endotracheal tube with lidocaine gel has been used to prevent sore throat after anesthesia. The utility of these techniques is controversial and, therefore, cannot be advocated as a standard of care (27,28).

COMPLICATIONS OF THE RESPIRATORY SYSTEM Bronchospasm Bronchospasm is a reflex constriction of the smooth bronchial muscle leading to increased airway resistance, expiratory flow obstruction, and air entrapment in a fashion similar to that of an acute asthma attack. Perhaps the main difference between a bronchospastic episode under anesthesia and asthma is the presence of inflammation in the latter. Bronchospasm under anesthesia has been recognized as a serious problem;

Chapter 1: Complications of Anesthesia

it is more likely to occur in individuals with asthma or reactive airway disease, smokers, patients with chronic obstructive pulmonary disease, and those with an acute viral respiratory infection; however, it also can occur in healthy individuals. The underlying mechanism for bronchospasm during anesthesia seems to be related to an increase in bronchomotor tone associated with intubation and laryngoscopy, as demonstrated by Gal and Suratt in healthy volunteers (29). In predisposed individuals, premature manipulation of the airway will result in an exaggerated response. The preoperative assessment of a patient at risk for bronchospasm should include a detailed history, with emphasis on the severity of symptoms, use of bronchodilators and steroids, and previous intubation. If a patient scheduled for elective surgery is actively wheezing during the preoperative visit, all efforts should be made to optimize the patient’s condition; otherwise, surgery should be postponed. Bronchospasm that develops intraoperatively typically occurs after induction of anesthesia, usually as a result of manipulation of the airway before adequate depth of anesthesia has been established. Minimizing the response to intubation and laryngoscopy is paramount. Several agents have been used with success for this purpose, the most common being b2 agonists and lidocaine (30). Regarding specific induction agents, it appears that propofol has a slightly more favorable profile than thiopental with respect to airway resistance (31), but this effect seems to be limited to the nongeneric formulation (32). Ketamine is the only induction agent consistently shown to induce bronchodilation, and may be considered as an option when patients need to undergo surgery while actively wheezing. Due to its undesirable side effects, however, this agent cannot be recommended routinely. The physical signs of bronchospasm that can be recognized during surgery include difficult ventilation (‘‘stiff bag’’), wheezing, and elevated peak airway pressures. It is important to remember that wheezing is a nonspecific sign, and can be associated with other conditions, including mechanical obstruction, main stem intubation, pneumothorax, or heart failure. The treatment includes deepening of anesthesia or sedation, confirmation of the proper position of the endotracheal tube (an endotracheal tube touching the carina is extremely irritating to the airway), continuing positive-pressure ventilation, and aggressive use of b2 agonists. The use of neuromuscular blocking agents helps decrease chest wall rigidity, but not bronchomotor tone. The amount of b2 agent that actually reaches the alveoli when given through an endotracheal tube is highly variable because a significant percentage of the dose delivered adheres to the polyvinylchloride of the tube. Therefore treatment must be guided by clinical response, rather than by a fixed number of puffs.

5

Atelectasis Arterial oxygenation is impaired during general anesthesia with either spontaneous or controlled ventilation. This impairment in oxygenation is largely due to ventilation–perfusion mismatch and shunt, with the magnitude of the shunt correlating closely with the degree of the atelectasis. Many factors influence the development of atelectasis during surgery. Induction of general anesthesia is consistently associated with a significant decrease in FRC and the formation of compression atelectasis. These changes occur within minutes of the anesthetic induction. The reduction in FRC correlates well with an increase in the venous admixture and alveoloarterial oxygen gradient. This is a very common cause of decreased arterial oxygen saturation in the recovery room. Another important factor in the development of intraoperative atelectasis is the type of surgery. Upper abdominal and thoracic procedures predispose the patient to this complication. The mechanical factors inherent to the procedure are compounded by the inability to cough and mobilize secretions. Adequate pain control in the postoperative period is essential to minimize the risk of atelectasis in these patients. Epidural anesthesia is an effective technique for pain control in patients whose respiratory function may be compromised and who require upper abdominal or thoracotomy procedures.

COMPLICATIONS OF THE CARDIOVASCULAR SYSTEM Perioperative Myocardial Ischemia or Infarction Coronary artery disease is the most frequent cause of perioperative cardiac morbidity and mortality in noncardiac surgery. Clinicians often need to identify patients at risk for developing cardiac complications. An extensive effort has been done to classify patients according to their risk factors. A history of prior myocardial infarction is an important predictor of postoperative reinfarction, and its incidence is inversely proportional to the time elapsed since the event. In the 1970s, reinfarction was reported in 30% of patients with recent (less than three months) myocardial infarction (33), but that number decreased to less than 5% in the 1990s (34), presumably due to improved hemodynamic monitoring and a prompt therapy of hemodynamic disturbances. What seemed clear was that the rate of reinfarction did not stabilize until approximately six months after the original myocardial infarction; thus previous recommendations to postpone all elective surgery until that period had passed. Even then, patients with a previous myocardial infarction will have an increased risk of reinfarction when compared to the general population.

6

Cobas and Varon

In 1996, the American College of Cardiology and the American Heart Association published guidelines for perioperative cardiovascular evaluation for noncardiac surgery (35). These guidelines provided a framework for considering cardiac risk in a variety of patients and surgical conditions. The major clinical predictors of increased perioperative cardiovascular risk are unstable coronary syndromes, defined as recent myocardial infarction, unstable or severe angina, decompensated congestive heart failure, significant arrhythmias, and severe valvular disease. Factors considered to be of intermediate significance are mild angina, old myocardial infarction, compensated congestive heart failure, and diabetes. Other factors include advanced age, uncontrolled hypertension, history of stroke, or abnormal nonspecific electrocardiogram findings such as ST segment changes, left ventricular hypertrophy, or left bundle branch block. When considering the probability of postoperative cardiac ischemic events, it is important to stratify according to the type of surgery. The cardiac risk is high—combined incidence of cardiac death and myocardial infarction more than 5%—in emergent major operation, aortic or peripheral vascular surgery, or any operation with major blood loss. Intermediate surgical risk—cardiac risk less than 5%—is represented by carotid endarterectomy, intraperitoneal and intrathoracic operations, and orthopedic, prostate, and head and neck surgery. Those patients deemed to be at risk based on either their functional status or the type of surgery should undergo further preoperative evaluation. From all the interventions attempted to reduce the incidence of postoperative myocardial infarction, the only one that has shown to be effective is the perioperative administration of b blockers (36,37). These agents are extremely useful, yet widely underutilized. If myocardial infarction does occur, it usually presents in the first 48 to 72 hours postoperatively. Most of these infarctions are subendocardial (non–Q wave); however, the diagnostic criteria are not uniformly defined. Furthermore, therapy with thrombolytic agents or anticoagulation in the setting of recent surgery is often contraindicated. Therefore, the main therapeutic approach in these cases is to maximize oxygen supply and reduce demand. In this setting, b blockers have also proven to be very useful.

Intraoperative Arrhythmias Intraoperative arrhythmias are very common during anesthesia, occurring in approximately 20% of patients, but only a very small number are considered serious. Not surprisingly, arrhythmias tend to occur more frequently in the elderly, in patients with preexisting heart disease or receiving digitalis, and at times of hemodynamic disturbances such as induction and emergence.

There are important anesthetic interactions that could make the heart more susceptible to arrhythmias. The discussion of specific conduction disturbances is beyond the scope of this chapter; however, it is important to know that various clinical situations can potentiate the arrhythmogenic effects of anesthetics. Although halogenated inhalational agents per se are unlikely to cause a clinically significant arrhythmia, in the presence of hypercarbia, hypoxia, and/or increased catecholamines, there is a higher risk for the development of ventricular ectopy. The dose of epinephrine required to induce arrhythmias is almost half in the presence of halothane than in the presence of isoflurane, suggesting that halothane reduces the threshold for arrhythmias (38). Most of the intravenous agents commonly used in anesthetic practice do not produce clinically significant arrhythmias. Perhaps an exception to this is ketamine, which has strong sympathomimetic properties. Local anesthetics block sodium channels, and thus have antiarrhythmic properties, but at toxic dosages, they become arrhythmogenic. Bupivacaine is especially notorious in this regard, with many reports of induction of ventricular tachycardia, ventricular fibrillation, and cardiac arrest. The proposed mechanism for this cardiac toxicity is that the drug is removed slowly from the sodium channel blocked, causing a delay in repolarization, leading to an increasingly prolonged QT interval and torsade de pointes. Cardiac arrests induced by bupivacaine are considered very difficult to treat, and there is support for using bretylium in this scenario (39).

COMPLICATIONS OF THE RENAL SYSTEM Renal Toxicity of General Anesthetics All inhaled anesthetics depress renal function as manifested by a decreased glomerular filtration rate (GFR), a decreased renal blood flow, and electrolyte disturbances. Free fluoride ions produced by the metabolism of these compounds can be directly nephrotoxic. Metoxiflurane, enflurane, and sevoflurane produce free fluoride ions when metabolized. The classic example of fluoride-induced nephrotoxicity was described in 1966 with metoxiflurane. Its toxic effect was manifested by decreased concentrating ability. Later reports confirmed that prolonged exposure to enflurane was also a risk factor for this complication (40). Halothane, isoflurane, and desflurane do not result in the production of fluoride in concentrations large enough to cause any kidney damage (40–42). Methoxiflurane is no longer in use in current human anesthetic practice. Enflurane’s use is also very limited due to the widespread acceptance of newer agents. Sevoflurane is a fluorinated anesthetic used for many years in Japan, which has progressively gained acceptance in the United States. It has a low blood gas

Chapter 1: Complications of Anesthesia

partition coefficient, which provides for a rapid onset of action, and it is not an irritant to the airways, making it an attractive choice for inhalation anesthesia. Serum fluoride concentration from the metabolism of this gas sometimes peaks above what is considered a safe threshold; however, because of its rapid elimination, these ions do not, stay around long enough to produce any toxicity (43). Sevoflurane does produce another metabolite, the compound A, after it reacts with carbon dioxide absorbers. Although at low gas flows this metabolite has been shown to be directly nephrotoxic in rats (44), the effect in humans remains unclear (45–48). Currently, the Food and Drug Administration recommends the use of fresh gas flow of at least 2 L/min when using sevoflurane.

Postoperative Renal Dysfunction The development of acute renal failure results in a significant increase in morbidity and mortality, as well as in health-care costs because renal replacement therapy is expensive and not free of complications. Development of renal failure in the perioperative period is almost always the result of a combination of factors and almost never the result of an anesthetic agent alone. Whether the cause of renal failure is aortic crossclamping, cardiopulmonary bypass, radiological contrast agent exposure, or protracted hypotension, the primary mechanism of injury is the ischemia of the outer renal medulla. This area represents a vascular watershed and is the first to manifest injury after a hypoxemic event (49). One of the challenges facing physicians is to characterize patients at risk for the development of renal failure. These patients include those who undergo procedures requiring cardiopulmonary bypass, with a reported incidence between 1% and 7% (50,51). Vascular surgery patients represent another important risk group for the development of renal failure, where it has traditionally been associated with poor prognosis (52,53). As expected, the incidence of acute renal failure after infrarenal aortic surgery is less than when aortic clamping is above the renal vessels. If the surgery is performed on an emergency basis, the incidence of this complication increases. Emergency operations for ruptured or leaking aneurysms are associated with renal failure in up to 30% of cases (54). Acute renal failure following trauma is rare. Morris retrospectively reviewed over 70,000 patients admitted to nine trauma centers during a five-year span. These authors found that the incidence of patients who required dialysis was only 0.01%. In the majority of the cases of renal failure presenting after trauma, the signs developed late and were associated with multiple organ failure (55). Preexistent renal failure has been consistently shown to be an important predictor of postoperative

7

renal failure. Therefore, it is essential to maintain proper renal function throughout the stress of surgery by providing the patient with adequate fluid volume, and using invasive hemodynamic monitoring when necessary. Adequate preoperative hydration of the patient not only ensures an adequate oxygen delivery to the renal tubules, but also diminishes demand by reducing the stimulus for sodium reabsorption. It is appropriate to set euvolemia as a goal while avoiding hypervolemia, in order to optimize oxygen delivery to the outer medulla without increasing the demand. A theoretical approach to minimize damage to the kidney would be to effectively reduce renal medullary oxygen demand. It is known that inhibitors of the adenosine triphosphatase (ATPase) located in the medullary thick ascending limb, such as furosemide, can dramatically reduce the metabolic demands of this region. In a study of radiocontrast exposure, however, pretreatment with furosemide failed to show a beneficial effect (56). Prophylactic oral administration of n-acetylcysteine, along with hydration, has been reported to prevent reduction in the renal function by low-osmolality contrast agents in patients with chronic renal insufficiency (57). The most controversial agent proposed to offer renal protection is DA at a low dose (‘‘renal dose’’). DA has been utilized extensively in doses less than 3 mg/kg/min to provide renal protection by improving renal blood flow. The conventional wisdom has been that at this dose range, only dopaminergic receptors will be stimulated, promoting splanchnic and renal vasodilation, as well as an increase in urine output and sodium excretion. These findings have been corroborated in animals and humans, but do not necessarily translate into an improved clinical outcome. In 1998, a comprehensive review of the literature found no evidence of a renal protective effect of DA in the perioperative period (58). More recently, a multicenter, randomized, double-blind, placebo-controlled study failed to show any benefit in critically ill patients at risk for renal dysfunction (59). A meta-analysis seems to confirm these findings (60). In fact, attempting to ‘‘improve’’ the renal function intraoperatively may actually predispose the kidney to ischemic injury by increasing oxygen requirements, and may result in undesirable cardiovascular side effects, i.e., tachycardia or arrhythmias. The notion of isolated dopaminergic receptor stimulation has been challenged because a study of normal volunteers given weight-based infusions of DA reported an enormous intersubject variability in plasma concentrations (61). Presently, there is insufficient evidence to support the use of DA as a renal protecting agent. Dopexamine is a pure dopaminergic agonist that has gained some attention because it may improve renal perfusion without the risks of adrenergic stimulation. However, large and well-designed trials are necessary to assess its usefulness.

8

Cobas and Varon

Fenoldopam is a specific DA 1–receptor agonist. It has been shown to reduce blood pressure in a dosedependent manner, while preserving renal perfusion and GFR (62). At a low dose, fenoldopam has been reported to improve renal blood flow without changes in systemic blood pressure. However, there is currently insufficient data to support the use of fenoldopam in critically ill patients.

COMPLICATIONS OF THE NEUROLOGIC SYSTEM Perioperative Stroke The incidence of perioperative strokes after surgery other than cardiac or neurologic is approximately 0.05% (63). As expected, the incidence is greater in the elderly and in patients with atherosclerosis. The pathophysiology of intraoperative stroke is for the most part thromboembolic. A study of the mechanisms of perioperative cerebral infarction reported that atrial fibrillation was present in 33% of patients (64). On the other hand, the presence of a carotid bruit in an otherwise asymptomatic patient does not seem to be a risk factor (65). A surgical procedure with a higher risk for perioperative stroke is carotid endarterectomy, where the incidence is reported to be about 5%. It seems reasonable that a general approach to prevent perioperative stroke is to maintain an adequate balance between cerebral oxygen supply and demand. This may be achieved by various interventions including maintaining blood pressure within 20% of baseline values. Because patients with longstanding hypertension have their autoregulatory cerebral perfusion curve shifted toward higher pressures, what could be considered as normotension for most patients might represent hypotension in this population. If a stroke is suspected, an immediate computed tomography (CT) scan of the brain with contrast should be obtained, along with neurology consultation. Thrombolytic therapy is usually contraindicated in the immediate postoperative period, and instituting anticoagulation must proceed with caution.

Postoperative Cognitive Dysfunction More than 150 years have passed since the first anesthetic was demonstrated, and the mechanism of action of general anesthetics is still elusive. Although all anesthetic drugs create a temporary dysfunction in the brain, the exact duration of this effect is unknown. Fine motor coordination is impaired for up to five hours after only a few minutes of exposure to halothane (66), and prolonged exposure to isoflurane may affect behavior up to 48 hours after the event (67). However, these effects are believed to be shortlived and self-limited, especially in a young healthy population. There is special concern about the possibility of long-lasting cognitive impairment in the elderly.

Although the results in this area of investigation are still controversial, a recent, prospective, randomized trial showed that up to 6% of patients undergoing either general or epidural anesthesia with sedation had some degree of cognitive impairment that was evident even six months after the operation (68). Another study evaluated 261 patients for five years to determine a long-term cognitive impairment after coronary artery bypass graft surgery using cardiopulmonary bypass. The investigators found cognitive dysfunction in 53% of patients at discharge from the hospital and in 42% at five years, with early cognitive impairment being a strong predictor of long-term impairment. Although there was no control group in this study, the fact that there was such a pronounced cognitive decline immediately after surgery suggests that this type of procedure may hasten cognitive impairment (69). These results, disturbing as they are, need to be confirmed by other trials evaluating different patient populations, anesthetic regimens, and types of surgery. If that is the case, permanent cognitive impairment will become an important perioperative complication to be considered in the geriatric population.

Perioperative Visual Loss Perioperative visual loss is rare but one of the most devastating complications that can occur in the immediate postoperative period. This complication has been reported after cardiopulmonary bypass surgery in the prone position, prolonged hypotension, and direct compression of the globe. The most common diagnosis of this condition is ischemic optic neuropathy (ION), which is a wellknown entity affecting patients with risk factors such as hypertension, atherosclerosis, and diabetes. The ASA established a database for cases of perioperative visual loss. In 23 cases of perioperative visual loss reported up to 1999, ION was present in 20 of them. In all the cases, the anesthetic was longer than five hours, and blood loss was considerable, averaging 2.2 L. Patients were in the prone position in 50% of the cases, and hypotension (systolic blood pressure or mean arterial pressure 40% below baseline) was found in 52% of the cases (70). The diagnosis of ION should be suspected in any patient with complaints of visual loss immediately after surgery, especially if it is painless. Suspicion of this entity should be followed by early ophthalmologic consultation. Unfortunately, this disorder carries a poor prognosis (70,71). Due to the multifactorial etiology of this complication, there is not a single specific measure to prevent this complication. However, obvious precautions should include intraoperative monitoring of ocular areas, avoiding external compression on the eyes, and aggressive treatment of hypotension.

Chapter 1: Complications of Anesthesia

Awareness Awareness during anesthesia is defined as the spontaneous recall of intraoperative events. It is a form of explicit memory because patients usually are able to tell exactly what was being said or done during a specific time of the procedure. Awareness has been described as one of the most horrible and traumatic experiences a patient can suffer, especially if they are under the effect of muscle relaxants. Patients have described this experience as the feeling of being buried alive. The incidence of awareness for noncardiac, nonobstetric surgery has been reported to be 0.2% (72). In cardiac surgery, it can reach up to 1.5%, and has been reported in as high as 11% of emergency trauma cases. In obstetrics, the incidence ranges from 0.4% to 7% (73). Frequent scenarios for the development of awareness are those in which administration of anesthetics is limited due to hemodynamic instability. Other causes include equipment malfunction or drug errors. A common time for awareness to occur is between induction of general anesthesia and skin incision. This is a period of relative quietness, when there is little stimulation to the patient, and therefore blood pressure tends to drift down. The usual response of anesthesia practitioners in this circumstance is to lower the concentration of anesthetics, increasing the risk of awareness. It is important to note that the patient can be suffering awareness, and yet nothing seems wrong even when the anesthetic chart is carefully reviewed. Increases in heart rate and blood pressure could represent indirect signs of awareness, but the most reliable indicator of the patient being awake and aware of the surroundings is motion. Therefore, one must carefully consider the use of muscle relaxants, especially when the surgical procedure does not require them. Patients who experienced awareness will sometimes complain of sleep disturbances, dreams and nightmares, flashbacks, and daytime anxiety. A small percentage of patients will go on to develop posttraumatic stress disorders (74). According to the ASA closed claims database, awareness represents almost 2% of 4183 total claims. The profile of the patient who suffered awareness was that of a healthy (ASA I or II) woman undergoing elective surgery. Awake paralysis, the most severe form of awareness, was caused by errors in administration or labeling of anesthetic drugs in 94% of cases. Other risk factors for the development of awareness were the intraoperative use of muscle relaxants and techniques that used little or no potent inhaled anesthetic (75). A monitor designed to measure the sedative and amnesic effects of anesthetics in the central nervous system has been developed. The Bispectral Index (BIS) monitor is a processed electroencephalogram

9

(EEG) with advanced algorithms derived from a database of thousands of patients undergoing multiple regimens of general anesthesia. The result is displayed on a scale of 0 (complete EEG inactivity) to 100 (awake), with 60 or lower representing unconsciousness associated with no recall (76). The BIS monitor has been reported to decrease the total dose of anesthetics used over time (77), and to increase the number of patients who are able to bypass the first phase of the postanesthesia-care unit (78). However, the BIS monitor is not 100% foolproof, and awareness even with numbers below 60 has been reported (79).

COMPLICATIONS OF REGIONAL ANESTHESIA Post–Dural Puncture Headache Post–dural puncture headache (PDPH) is a known complication of spinal anesthesia, where the duramater is intentionally penetrated, and can also occur in epidural anesthesia, when the duramater is unintentionally perforated. Studies using small-gauge noncutting needles for spinal anesthesia have shown that the incidence of this complication is about 1%. When using a large 17-gauge epidural needle, the incidence can be as high as 75% (80). The pathophysiology of PDPH is related to the loss of cerebrospinal fluid through the dural puncture, resulting in a downward traction of the meninges and intracranial vessels. Pain is characteristically described as frontal, radiating to the occiput, but can also extend to the posterior cervical region; with an increase in severity, it becomes circumferential and can be associated with blurred vision, diplopia, and tinnitus. It is worsened by sitting upright and improved by bed rest. PDPH usually occurs 12 to 24 hours postoperatively, but it can be immediate if the leak is significant. Risk factors for developing PDPH can be mechanical or patient-related. Mechanical factors include needle size, with the incidence being directly proportional to the diameter of the needle, and design of the needle tip. Pencil-tipped (Whitacre) needles have a dramatically reduced incidence of causing PDPH compared with cutting-tipped (Quincke) needles of equal diameter. Pencil-tipped needles divide rather than cut fibers as they traverse the dura. Patients at risk of developing PDPH are young females. Old age seems to be protective. Conservative management for PDPH, including bed rest, encouragement of oral or intravenous hydration, and use of caffeine-containing beverages or caffeine benzoide infusion is effective in 50% of the cases. A more invasive form of treatment is the performing of an epidural blood patch. This consists of an injection of 10 to 20 mL of the patient’s own blood into the epidural space with the purpose of sealing the dural hole and preventing further cerebrospinal fluid leakage. This method is effective in over 90% of the

10

Cobas and Varon

cases (81). The choice of conservative measures or epidural blood patch largely depends on the severity of the headache, the risk and/or potential difficulty for performing the procedure, and the patient’s wishes. At our institution, epidural blood patch is offered prophylactically to all parturients who have suffered a dural puncture with an epidural needle (‘‘wet tap’’). This approach has been reported by others to decrease the incidence of PDPH by 58% (82).

Spinal Hematoma This is an extremely rare, but potentially devastating complication that may develop from the accumulation of blood in the subarachnoid, subdural, or epidural space. Hematoma can occur after spinal or epidural anesthesia has been attempted or performed. Neurologic damage results from cord compression. The epidural space is rich in venous plexuses and is the most common site for blood accumulation. The incidence of this complication is so small that it would require an enormous number of patients to define it with certainty; however, two large series reported an estimate of the incidence in 1:200,000 procedures (83,84). The main risk factor for the development of epidural hematoma is the presence of coagulopathy, either native or acquired. Special care must be taken to ascertain whether there is a patient or family history of bleeding problems or use of anticoagulant drugs (85). The initial diagnosis of an epidural hematoma is clinical and includes the presence of severe back pain at the site of the injection in association with a block that is not receding. Late signs include bowel and bladder dysfunction. Usually, signs and symptoms occur within 24 hours of the blockade. Because time is of the essence, as soon as an epidural hematoma is suspected, a neurosurgical consultation should be obtained. Once diagnosis is confirmed, the treatment of choice is a surgical laminectomy; complete recovery is enhanced when the spinal cord is decompressed within eight hours of diagnosis (83).

COMPLICATIONS OF SPECIFIC NERVE BLOCKS When performing a nerve block for any surgical procedure, the benefits of the technique must outweigh the risks inherent to the procedure. In the case of nerve blocks, advantages clearly relate to the fact that all the complications associated with airway manipulation are avoided and there is decreased interference with pulmonary and circulatory responses. However, nerve blocks also carry complications that occur as the result of four main mechanisms: local toxicity of the anesthetic, systemic toxicity of the local anesthetic, mechanical or direct nerve damage, and damage or injection to adjacent structures.

Local anesthetics are cytotoxic to neural tissue, an effect that is directly related to the concentration used. The effects of a concentrated solution of local anesthetic in close proximity to large trunks or nerves in the spinal cord has recently stirred a controversy regarding the safety of the most widely used local anesthetic, lidocaine (86,87). Mechanical damage by direct trauma to the nerve is possible when using a technique that deliberately elicits paresthesias for identifying the location of a nerve. The risk of nerve damage is greater when using a cutting-edge needle than a blunt-tipped one. When the paresthesia technique was compared to the transarterial technique for upper arm blockades, the incidence of neuropathy postblock was not very different (88). Common sense, however, dictates that procedures should be stopped and the needle repositioned if paresthesia or pain increases during injection. In sufficient concentrations, all local anesthetics have systemic toxic effects. The neurologic and cardiovascular effects are of particular concern. Toxic reactions correlate with the concentration of the anesthetic at the end organ, which usually correlates with the blood concentration of the drug, except in those cases where the anesthetic is injected almost directly into the affected organ, such as in carotid injection, where less than 1 mL of lidocaine 2% can produce seizures. Intercostal blocks are associated with the highest plasma concentration of local anesthetics because these blocks are usually performed several times, and the area is richly vascularized. If the same amount of the local anesthetic is deposited in a deep sheath such as the axillary or the femoral sheath, the risks of toxicity are less because the absorption is considerably less from these areas. The addition of epinephrine to the solution of a local anesthetic significantly reduces its systemic absorption, allowing the use of higher doses for any given block. This effect of epinephrine is not seen with bupivacaine, presumably because of its own vasocontrictive properties. Blocks performed in the neck area, such as in the stellate ganglion or the cervical plexus, or in the laryngeal or interscalene approach to the brachial plexus, have a risk of intravascular (carotid, vertebral, or jugular) injection. They can also be complicated with intrathecal injection, producing total spinal anesthesia. There are reports of spinal cord damage when using long needles after interscalene blocks were performed in anesthetized patients for postoperative pain (89). Ipsilateral phrenic nerve paralysis is a common temporary side effect of interscalene and stellate ganglion blocks, but it can produce ventilatory complications in patients with decreased respiratory reserve. The performance of bilateral interscalene blocks is contraindicated even in healthy individuals. Compartment syndrome with nerve ischemia can occur if local anesthesia is injected into a

Chapter 1:

nondistensible space, such as the ulnar groove at the level of the elbow.

Perioperative Nerve Injury Although patient positioning and lack of adequate vigilance is blamed frequently for the development of perioperative nerve injury, the mechanism remains one of the most difficult to explain. Peripheral nerve injuries are a significant source of claims against anesthesiologists. Settlements are frequent in these lawsuits, despite the fact that in a large proportion of these cases the standard of care was met. Ulnar nerve and brachial plexus injuries comprise approximately half of all the claims filed for nerve damage, followed by lumbosacral nerve root and spinal cord injuries. The final common pathway to nerve injury seems to be ischemia, resulting from either compression or stretching. The presence of comorbidities seems to play an important role in the development of peripheral nerve injuries. Injuries to the ulnar nerve represent 28% of all the cases studied in the ASA Closed Claims Project (90). The characteristics of this neuropathy are particularly difficult to explain because the vast majority of cases (75%) occur in males; it has a particularly late onset, with a median delay of three days; and it has occurred even when extra padding has been applied to the humeral epycondile. Anatomic considerations that could explain this neuropathy are the shallow location of the nerve at the elbow and the fact that it is surrounded by a taut aponeurosis. The ulnar nerve can be compressed against the lateral surface of the operating room table when the arms are ‘‘tucked in,’’ or when the arm is abducted and in the prone rather than supine position. Many characteristics of ulnar neuropathy defy simplistic explanation. Although one would expect that an awake or lightly sedated patient would complain of compression of the ulnar nerve, the reported incidence of neuropathy is no different when comparing general versus regional anesthesia (91). Age is also an important factor, with a median age for this complication being 50 years and there being a total absence of it in the pediatric population (90). Based on the above, it is clear that the positioning and lack of padding cannot always explain the occurrence of ulnar neuropathy. Currently, there are no evidencebased strategies to prevent it, and the cause of the problem is most likely beyond the control of the anesthesiologist. Furthermore, electromyography studies have shown evidence of subclinical neuropathy in the contralateral side of the injury (92). Brachial plexus injuries can be more easily traced to positioning problems, usually the result of overstretching the plexus, with exaggerated abduction of the arm, or sustained neck extension. The use of shoulder braces and the head-down position have also been implicated.

Complications of Anesthesia

11

Spinal cord and lumbosacral root injuries are the third most common perioperative nerve injuries, and they are usually related to the performance of a regional anesthetic or pain management procedure. Epidural injection in the presence of anticoagulants poses a special risk (see above). Sciatic nerve injuries can result from compression during hip surgery because the nerve passes between the ischial tuberosity and the greater trochanter of the femur. In the lithotomy position, the nerve can be overstretched, which leads to injury. The superficial branch of the common peroneal nerve is susceptible to injury when compressed against the support brace during lithotomy position because it superficially wraps the fibular neck. The clinical presentation of the peripheral nerve damage varies widely, ranging from abrupt postoperative loss of sensation or motor function to slowly progressive deficits over the course of days, or even weeks. The presence of pain is common. When this complication occurs, it is useful to consult a neurologist. Spinal cord injuries mandate an immediate CT or magnetic resonance imaging scan. An electromyogram should be ordered to differentiate acute from chronic damage. Denervation changes usually take approximately three weeks to develop, so, if present, the diagnosis of preexisting neuropathy is made. Electromyography studies also help delineate the exact site of injury. With the probable exception of ulnar nerve injuries, most of these complications can be prevented by the combined effort of the anesthesiologist, the surgeon, and the operating room personnel. Careful positioning, padding, and frequent reassessing of pressure points at regular intervals throughout the procedure represent simple and inexpensive measures that can reduce morbidity in the postoperative period.

SYSTEMIC AND METABOLIC COMPLICATIONS Malignant Hyperthermia Malignant hyperthermia (MH) is probably, of all the complications discussed in this chapter, the only one attributable directly and exclusively to anesthetic agents. It is defined as an acute, potentially fatal disorder in which skeletal muscles unexpectedly increase their metabolic rate during general anesthesia (93,94). This hypermetabolic state increases the oxygen consumption significantly, sometimes above oxygen delivery, and causes increased CO2 production, lactate and heat production, respiratory and metabolic acidosis, muscle rigidity, sympathetic stimulation, and increased cellular permeability. The incidence of MH is about 1:75,000 in the adult population, whereas in children it is reported to be approximately 1:4,000. A review of the calls to the MH Hotline, a dedicated 24-hour response line, showed that during 1990 to 1994, there were 978 clinical cases: 25% occurred during ENT surgery, 25%

12

Cobas and Varon

during orthopedic surgery, and 25% during emergency surgery. It is important to note that 43% of cases in the ENT group had previously undergone anesthesia without developing this complication (95). When MH was first described in 1960, the mortality rate was close to 90%. The rate declined slowly over the next decade and a half, but it was not until dantrolene sodium was introduced in 1975 that the mortality rate decreased substantially to 7%. In 1995, data show a mortality rate below 2%. MH is a genetic disease, and its pattern of transmission is in the mode of autosomal dominance with reduced penetrance and variable expression. Reduced penetrance is implied when fewer offspring than one would predict are affected, whereas variable expression accounts for different susceptibilities among family members. One of the most common chromosome defects observed is located in chromosome 19, associated with the ryanodine receptor. This receptor is involved with the pathway of calcium release from the sarcoplasmic reticulum to the cytoplasm (96). The pathogenesis of MH lies in the inability of the sarcoplasmic reticulum to store calcium, with high levels of myoplasmic calcium even in the resting stage, and as much as 17 times normal after the triggering of a MH episode. The elevated myoplasmic calcium causes the activation of ATPase, thus accelerating the hydrolysis of ATP to adenosine diphosphate. It also inhibits troponin, eventually leading to muscle contraction, and activates phosphorylase kinase, with the production of ATP and heat. The decrease in venous blood oxygen saturation and the overloading of carbon dioxide and lactic acid are the accumulative results of the cellular demands of hypermetabolism. All potent inhalation anesthetics have been implicated in the genesis of MH, including the newer agents sevoflurane and desflurane. Nitrous oxide is a safe agent to use in patients with a history of MH. Among the muscle relaxants, only succinylcholine has been implicated. All other nondepolarizing muscle relaxants have failed to trigger an episode of MH. Induction agents, including thiopental, propofol, and ketamine, as well as opioids and benzodiazepines, are nontriggering. Local anesthetics also appear to be safe. It is important to differentiate this syndrome from other causes of fever and hypercarbia in the operating room, such as sepsis, neurologic injury, thyroid storm, pheochromocytoma, and acute cocaine intoxication. Of all the standard monitors required during general anesthesia, ETCO2 is the most sensitive and useful for the diagnosis of a hypermetabolic event; the ETCO2 concentration is elevated before any evident change in pulse or respiration, and usually does not respond to vigorous hyperventilation. The rising temperature is a late sign of this syndrome. Therapy for a MH episode should not be delayed, and it is aimed at discontinuing triggering agents and decreasing the hypermetabolic state while

treating the hyperthermia. The patient should receive 100% oxygen, surgery should conclude as soon as possible, and hyperventilation and cooling measures should be instituted. A Foley catheter to monitor the urinary output is essential. An arterial line to monitor blood pressure and to follow the acid–base status is also highly desirable. Dantrolene sodium should be started at doses of 2.5 to 3.0 mg/kg, followed in 45 minutes by a bolus dose of 10 mg/kg if all signs and symptoms of the syndrome have not resolved. The response to dantrolene takes 6 to 20 minutes, and by 45 minutes, the patient’s condition should return to normal. In about 10% of patients, the syndrome redevelops after four to eight hours, a phenomenon that has been called recrudescence. It has been suggested that dantrolene prevents the release of calcium from the sarcoplasmic reticulum or antagonizes its effect at the myofibril level, or both. At the usual dosages, it has significant muscle relaxant properties, but does not depress respiration even at doses of up to 30 mg/kg. Prophylactic treatment with dantrolene sodium for patients susceptible to MH is no longer recommended. However, avoidance of MH-triggering agents is essential when caring for these patients. Acute mortality from MH is usually a result of malignant arrhythmias; after a few hours, it is more likely the result of pulmonary edema, coagulopathy, or acid–base imbalance. Late complications include hemolysis and hemoglobinuria, leading to renal failure. In summary, MH is a serious disease, but with adequate vigilance, proper monitoring, and aggressive therapy, it should be a less threatening and curable entity.

Latex Allergy Latex allergy is a common problem among a specific group of patients in whom there is production of an antibody of the immunoglobulin E class in response to the protein component of sap of the rubber tree, Hevea brasilienses. Patients at risk for this complication include all those who are frequently exposed to the product. These comprise children with a history of spina bifida or those who require chronic urinary catheterization. Health-care workers, because of their constant exposure to latex products, also represent a risk group. Latex allergy should be suspected if unexpected anaphylaxis occurs after the start of the surgical procedure, without any temporal relationship to medication or blood product administration. Any history or symptoms suggestive of latex allergy (itching, hives, or wheezing) after contact with latex products should raise a red flag in the mind of all the personnel taking care of the patient. The only effective way to prevent an episode of latex allergy is to avoid exposure to the antigen. It is now common to have a ‘‘latex allergy’’

Chapter 1:

cart, where all the products that are safe to use in this population are stored and labeled. It is generally recommended that these cases be posted first in the daily schedule in order to avoid cross-contamination with the equipment used in nonsusceptible patients. When a case of anaphylaxis to latex exposure does occur, treatment is no different than that administered with other causes of life-threatening allergic reactions. Contact with the antigen must be stopped as soon as possible, and the airway should be secured if an endotracheal tube is not already in place. Anaphylactic reactions promote capillary leakage with massive redistribution of fluids to the interstitial space. Therefore, a rapid administration of isotonic crystalloids is necessary to maintain an effective perfusion. The mainstay of pharmacologic treatment for an anaphylactic reaction is the use of intravenous epinephrine. Rather than a fixed dose, the drug should be titrated in 100-mg increments, until the blood pressure is restituted. An infusion may be needed in cases of protracted hypotension. Antihistamines and glucocorticoids may also be useful to stabilize the allergic response after the initial resuscitation has taken place.

CONCLUSIONS In this chapter, we have discussed complications that are related to the administration of anesthesia. Many of these problems cannot be exclusively attributed to the anesthetic agent or the technique that is chosen, but rather, they are a result of the compounding effects of surgery, previous medical conditions, and the overall stress response that is imposed on the body by the surgical procedure. Although we may never know with certainty the true incidence of many of these complications, it is clear that the identification of patients at risk and advanced preparation play a major role in preventing most of them. Anesthesia is perhaps the safest medical specialty of all, when parameters of appropriate care and vigilance are met. Anesthesia care has evolved in great leaps since its first public demonstration in 1846. A more profound understanding of the disease processes, better patient selection, and intraoperative monitoring and advances in surgical techniques have made possible the realization of increasingly complex procedures on sicker patients. In today’s world, the risk of dying in a car accident is greater than that of dying while under anesthesia (97).

REFERENCES 1. Beecher HK, Todd DP. A study of deaths associated with anesthesia and surgery. Ann Surg 1954; 140:2. 2. Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia in surgical mortality. JAMA 1961; 178:261.

Complications of Anesthesia

13

3. Bodlander FMS. Deaths associated with anesthesia. Br J Anaesth 1975; 47:36. 4. Holland R. Anaesthetic mortality in New South Wales. Br J Anaesth 1987; 59(7):834–841. 5. Buck N, Devlin HB, Lunn JL. Report on the Confidential Enquiry into Perioperative Death. London: Nuffield Provincial Hospitals Trust, The Kings Fund Publishing House, 1987. 6. Lunn JN. The study of anesthetic-related mortality. Anaesthesia 1982; 37:856. 7. Lunn JN, Hunter AR, Scott DB. Anaesthesia-related surgical mortality. Anaesthesia 1983; 38:1090. 8. Caplan RA. The closed claims project: looking back, looking forward. ASA Newslett 1999; 63(6):7–9. 9. Caplan RA, Posner KL, Ward RJ, Cheney FW. Adverse respiratory events in anesthesia: a closed claims analysis. Anesthesiology 1990; 72(5):828–833. 10. Domino KB. Closed malpractice claims for airway trauma during anesthesia. ASA Newslett 1998; 62(6):10–11. 11. Martian CH, Picard E, Jonquet O, et al. Membranous tracheal rupture after intubation. Ann Thorac Surg 1995; 60:1367–1371. 12. Marquette CH, Bocquillon N, Roumilhac D, Neviere R, Mathieu D, Ramon P. Conservative treatment of tracheal rupture. J Thorac Cardiovasc Surg 1999; 117(2):399–401. 13. Zettl R, Waydhas C, Biberthaler P, et al. Nonsurgical treatment of a severe tracheal rupture after endotracheal intubation. Crit Care Med 1999; 27(3):661–663. 14. Warner MA, Warner ME, Weber JG. Clinical significance of pulmonary aspiration during the perioperative period. Anesthesiology 1993; 78(1):56–62. 15. Lakshmipathy N, Bokesch PM, Cowen DE, Lisman SR, Schmid CH. Environmental tobacco smoke: a risk factor for pediatric laryngospasm. Anesth Analg 1996; 82(4):724–727. 16. LeResche L. Epidemiology of temporo-mandibular disorder: implication for the investigation of etiologic factors. Crit Rev Oral Biol Med 1997; 8:291–305. 17. Risk management foundation. Anesthesia claims analysis shows frequency loss, losses high. Risk Manage Found Forum 1983; 4:1–2. 18. Aitkenhead AR. The pattern of litigations against anaesthetists. Br J Anaesth 1994; 73:10–21. 19. Gaiser RR, Castro AD. The level of anesthesia resident training does not affect the risk of dental injury. Anesth Analg 1998; 87:255–257. 20. Lockhart PB, Feldbau EV, Gabel RA, et al. Dental complications during and after tracheal intubation. J Am Dent Assoc 1986; 112:480–483. 21. Cronin M, Redfern PA, Utting JE. Psychometry and postoperative complaints in surgical patients. Br J Anaesth 1973; 45:879–886. 22. Macario A, Weinger M, Carney S, Kim A. Which clinical anesthesia outcomes are important to avoid? The perspective of patients. Anesth Analg 1999; 89(3): 652–658. 23. Mylis PS, Hunt JO, Moloney JT. Postoperative ‘‘minor’’ complications: comparison between men and women. Anaesthesia 1997; 52(4):300–306. 24. Lipp M, Brandt L, Daublander M, Peter R, Barz L. Frequency and severity of throat complaints following general anesthesia with the insertion of various endotracheal tubes. Anaesthesist 1988; 37(12):758–766.

14

Cobas and Varon

25. Joorgensen LN, Weber M, Pedersen A, Munster M. No increased incidence of postoperative sore throat after administration of suxamethonium in endotracheal anaesthesia. Acta Anaesthesiol Scand 1987; 31(8):768–770. 26. Capan LM, Bruce DL, Patel KP, Turndorf H. Succinylcholine-induced postoperative sore throat. Anesthesiology 1983; 59(3):202–206. 27. Porter NE, Sidou V, Husson J. Postoperative sore throat: incidence and severity after the use of lidocaine, saline, or air to inflate the endotracheal tube cuff. AANA J 1999; 67(1):49–52. 28. Navarro RM, Baughman VL. Lidocaine in the endotracheal tube cuff reduces postoperative sore throat. J Clin Anesth 1997; 9(5):394. 29. Gal TJ, Suratt PM. Resistance to breathing in healthy subjects following endotracheal intubation under topical anesthesia. Anesth Analg 1980; 59(4):270–274. 30. Tam S, Chung F, Campbell M. Intravenous lidocaine: optimal time of injection before tracheal intubation. Anesth Analg 1987; 66(10):1036–1038. 31. Eames WO, Rooke GA, Wu RS, Bishop MJ. Comparison of the effects of etomidate, propofol, and thiopental on respiratory resistance after tracheal intubation. Anesthesiology 1996; 84(6):1307–1311. 32. Brown RH, Greenber RS, Wagner EM. Efficacy of propofol to prevent bronchoconstriction. Anesthesiology 2001; 94:851–855. 33. Tarhan S, Moffitt EA, Taylor WF, Giuliani ER. Myocardial infarction after general anesthesia. JAMA 1972; 220(11):1451–1454. 34. Shah KB, Kleinman BS, Sami H, Patel J, Rao TL. Reevaluation of perioperative myocardial infarction in patients with prior myocardial infarction undergoing noncardiac operations. Anesth Analg 1990; 71(3): 231–235. 35. Eagle KA, Brundage BH, Chaitman BR, et al. Guidelines for perioperative cardiovascular evaluation for noncardiac surgery: an abridged version of the report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Mayo Clin Proc 1997; 72(6):524–531. 36. Badner NH, Knill RL, Brown JE, Novick TV, Gelb AW. Myocardial infarction after noncardiac surgery. Anesthesiology 1998; 88(3):572–578. 37. Poldermans D, Boersma E, Bax J, et al. The effect of bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. The Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography Study Group. N Engl J Med 1999; 341:1789–1794. 38. Johnston RR, Eger EI II, Wilson C. A comparative interaction of epinephrine with enflurane, isoflurane, and halothane in man. Anesth Analg 1976; 55(5):709–712. 39. Kasten GW, Martin ST. Bupivacaine cardiovascular toxicity: comparison of treatment with bretylium and lidocaine. Anesth Analg 1985; 64(9):911–916. 40. Mazze RI, Calverley RK, Smith NT. Inorganic fluoride nephrotoxicity: prolonged enflurane and halothane anesthesia in volunteers. Anesthesiology 1977; 46(4): 265–271. 41. Mazze RI, Cousins MJ, Barr GA. Renal effects and metabolism of isoflurane in man. Anesthesiology 1974; 40(6):536–542. 42. Sutton TS, Koblin DD, Gruenke LD, et al. Fluoride metabolites after prolonged exposure of volunteers

43.

44.

45. 46.

47.

48.

49. 50. 51.

52. 53. 54. 55. 56.

57.

58. 59.

and patients to desflurane. Anesth Analg 1991; 73(2): 180–185. Frink EJ Jr., Ghantous H, Malan TP, et al. Plasma inorganic fluoride with sevoflurane anesthesia: correlation with indices of hepatic and renal function. Anesth Analg 1992; 74(2):231–235. Gonsowski CT, Laster MJ, Eger EI II, Ferrell LD, Kerschmann RL. Toxicity of compound A in rats. Effect of a 3-hour administration. Anesthesiology 1994; 80(3):556–565. Eger El II, Koblin DD, Bowland T, et al. Nephrotoxicity of sevoflurane versus desflurane anesthesia in volunteers. Anesth Analg 1997; 84:160–168. Campbell JAH, Corrigall AV, Guy A, Kirsch RE. Immunohistologic localization of alpha, mu, and pi class glutathione-s-transferase in human tissues. Cancer 1991; 67:1608–1613. Kharasch ED, Frink EJ Jr., Zager R, Bowdle TA, Artu A, Nogami WM. Assessment of low-flow sevoflurane and isoflurane effects on renal function using sensitive markers of tubular toxicity. Anesthesiology 1997; 86: 1238–1254. Bito H, Ikeuchi Y, Ikeda K. Effects of low-flow sevoflurane anesthesia on renal function: comparison with high-flow sevoflurane anesthesia and low-flow isoflurane anesthesia. Anesthesiology 1997; 86:1231–1237. Brezis M, Rosen S. Hypoxia of the renal medulla—its implications for disease. N Engl J Med 1995; 332(10): 647–655. Koning HM, Koning AJ, Leusink JA. Serious acute renal failure following open heart surgery. Thorac Cardiovasc Surgeon 1985; 33(5):283–287. Mangano CM, Diamondstone LS, Ramsay JG, Aggarwal A, Herskowitz A, Mangano DT. Renal dysfunction after myocardial revascularization: risk factors, adverse outcomes, and hospital resource utilization. The Multicenter Study of Perioperative Ischemia Research Group. Ann Intern Med 1998; 128(3):194–203. Kazmers A, Jacobs L, Perkins A. The impact of complications after vascular surgery in Veterans Affairs Medical Centers. J Surg Res 1997; 67(1):62–66. Gornick CC Jr., Kjellstrand CM. Acute renal failure complicating aortic aneurysm surgery. Nephron 1983; 35(3):145–157. Chawla SK, Najafi H, Ing TS, et al. Acute renal failure complicating ruptured abdominal aortic aneurysm. Arch Surg 1975; 110(5):521–526. Morris JA Jr., Mucha P Jr., Ross SE, et al. Acute posttraumatic renal failure: a multicenter perspective. J Trauma-Injury Infect Crit Care 1991; 31(12):1584–1590. Solomon R, Werner C, Mann D, D’Elia J, Silva P. Effects of saline, mannitol, and furosemide to prevent acute decreases in renal function induced by radiocontrast agents. N Engl J Med 1994; 331(21):1416–1420. Tepel M, van der Giet M, Schwarzfeld C, Laufer U, Liermann D, Zidek W. Prevention of radiographiccontrast-agent-induced reductions in renal function by acetylcysteine. N Engl J Med 2000; 343(3):210–212. Perdue PW, Balser JR, Lipsett PA, Breslow MJ. ‘‘Renal dose’’ dopamine in surgical patients: dogma or science. Ann Surg 1998; 227(4):470–473. Bellomo R, Chapman M, Finfer S, Hickling K, Myburgh J. Low-dose dopamine in patients with early renal dysfunction: a placebo-controlled randomised trial. Australian and New Zealand Intensive Care Society

Chapter 1:

60. 61.

62.

63. 64. 65. 66.

67. 68.

69.

70. 71. 72. 73. 74. 75. 76.

77.

78.

(ANZICS) Clinical Trials Group. Lancet 2000; 356(9248): 2139–2143. Kellum JA, Decker JM. Use of dopamine in acute renal failure: a meta-analysis. Crit Care Med 2001; 29(8): 1526–1531. MacGregor DA, Smith TE, Prielipp RC, Butterworth JF, James RL, Scuderi PE. Pharmacokinetics of dopamine in healthy male subjects. Anesthesiology 2000; 92(2): 338–346. Mathur VS, Swan SK, Lambrecht LJ, et al. The effects of fenoldopam, a selective dopamine receptor agonist, on systemic and renal hemodynamics in normotensive subjects. Crit Care Med 1999; 27(9):1832–1837. Warner MA, Shields SE, Chute CG. Major morbidity and mortality within 1 month of ambulatory surgery and anesthesia. JAMA 1993; 270(12):1437–1441. Hart R, Hindman B. Mechanisms of perioperative cerebral infarction. Stroke 1982; 13(6):766–773. Ropper AH, Wechsler LR, Wilson LS. Carotid bruit and the risk of stroke in elective surgery. N Engl J Med 1982; 307(22):1388–1390. Korttila K, Tammisto T, Ertama P, Pfaffli P, Blomgren E, Hakkinen S. Recovery, psychomotor skills, and simulated driving after brief inhalational anesthesia with halothane or enflurane combined with nitrous oxide and oxygen. Anesthesiology 1977; 46(1):20–27. Davison LA, Steinhelber JC, Eger EI II, Stevens WC. Psychological effects of halothane and isoflurane anesthesia. Anesthesiology 1975; 43(3):313–324. Williams-Russo P, Sharrock NE, Mattis S, Szatrowski TP, Charlson ME. Cognitive effects after epidural versus general anesthesia in older adults. A randomized trial. JAMA 1995; 274(1):44–50. Newman MF, Kirchner JL, Phillips-Bute B, et al. Longitudinal assessment of neurocognitive function after coronary-artery bypass surgery. N Engl J Med 2001; 344(6):395–402. Lee LA. Post-operative visual loss data gathered and analyzed. ASA Newslett 2000; 64(9):25–27. Williams EL, Hart WM, Tempelhoff R. Postoperative ischemic optic neuropathy. Anesth Analg 1995; 80(5): 1018–1029. Liu WH, Thorp TA, Graham SG, Aitkenhead AR. Incidence of awareness with recall during general anaesthesia. Anaesthesia 1991; 46(6):435–437. Juul J, Lie B, Friberg Nielsen S. Epidural analgesia versus general anesthesia for cesarean section. Acta Obstet Gynecol Scand 1988; 67(3):203–206. Moerman N, Bonke B, Oosting J. Awareness and recall during general anesthesia. Facts and feelings. Anesthesiology 1993; 79(3):454–464. Domino KB, Postner KL, Caplan RA, Cheney FW. Awareness during anesthesia: a closed claims analysis. Anesthesiology 1999; 90:1053–1061. Glass PS, Bloom M, Kearse L, Rosow C, Sebel P, Manberg P. Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology 1997; 86(4):836–847. Song D, Joshi GP, White PF. Titration of volatile anesthetics using bispectral index facilitates recovery after ambulatory anesthesia. Anesthesiology 1997; 87(4): 842–848. Sebel PS, Payne FB, Gan TJ, Rosow CE, Greenwald S. Bispectral analysis (BIS) monitoring improves PACU

79. 80.

81.

82. 83. 84. 85. 86. 87.

88. 89.

90. 91.

92. 93. 94. 95. 96.

97.

Complications of Anesthesia

15

recovery from propofol/alfentanil/nitrous oxide anesthesia. Anesthesiology 1996; 85:A468. Mychaskiw G II, Horowitz M, Sachdev V, Heath BJ. Explicit intraoperative recall at a Bispectral Index of 47. Anesth Analg 2001; 92(4):808–809. Lambert DH, Hurley RJ, Hertwig L, Datta S. Role of needle gauge and tip configuration in the production of lumbar puncture headache. Region Anesth 1997; 22(1):66–72. Molnar R. Spinal, Epidural and Caudal Anesthesia: Clinical Anesthesia Procedures of the Massachusetts General Hospital. 5th ed. Boston: Little, Brown and Company, 1993:206–225. Colonna-Romano P, Shapiro B. Prophylactic epidural blood patch in obstetrics. Anesthesiology 1988; 69:A665. Vandermeulen EP, Van Aken H, Vermylen J. Anticoagulants and spinal-epidural anesthesia. Anesth Analg 1994; 79(6):1165–1177. Tryba M. Epidural regional anesthesia and low molecular heparin. Anasthesiologie, Intensivmedizin, Notfallmedizin, Schmerztherapie 1993; 28(3):179–181. Cobas M. Preoperative assessment of coagulation disorders. Int Anesthesiol Clin 2001; 39(1):1–15. Rigler ML, Drasner K, Krejcie TC, et al. Cauda equina syndrome after continuous spinal anesthesia. Anesth Analg 1991; 72(3):275–281. Schell RM, Brauer FS, Cole DJ, Applegate RL II. Persistent sacral nerve root deficits after continuous spinal anaesthesia. Canadian J Anaesth 1991; 38(7): 908–911. Selander D, Edshage S, Wolff T. Paresthesiae or no paresthesiae? Nerve lesions after axillary blocks. Acta Anaesthesiol Scand 1979; 23(1):27–33. Benumof JL. Permanent loss of cervical spinal cord function associated with interscalene block performed under general anesthesia. Anesthesiology 2000; 93:1541–1544. Cheney FW, Domino KB, Caplan RA, Posner KL. Nerve injury associated with anesthesia: a closed claims analysis. Anesthesiology 1999; 90(4):1062–1069. Warner MA, Warner ME, Martin JT. Ulnar neuropathy. Incidence, outcome, and risk factors in sedated or anesthetized patients. Anesthesiology 1994; 81(6): 1332–1340. Alvine FG, Schurrer ME. Postoperative ulnar-nerve palsy. Are there predisposing factors? J Bone Joint Surg—Am Vol 1987; 69(2):255–259. Britt BA, Kalow W. Malignant hyperthermia: a statistical review. Can Anaesth Soc J 1970; 17(4):293–315. Britt BA. Etiology and pathophysiology of malignant hyperthermia. Fed Proc 1979; 38(1):44–48. Greenberg C, Hall S, Karan S. Malignant hyperthermia (MH) during outpatient pediatric ENT surgery—an anesthesia concern. Anesthesiology 1995; 83:A1004. Ohnishi ST, Taylor S, Gronert GA. Calcium-induced CA2þ release from sarcoplasmic reticulum of pigs susceptible to malignant hyperthermia. The effects of halothane and dantrolene. FEBS Lett 1983; 161(1):103–107. National Highway Traffic Safety Administration. National Center for Statistics and Analysis U.S. Department of Transportation. Traffic Safety Facts 1999: A Compilation of Motor Vehicle Crash Data from the Fatality Analysis Reporting System and the General Estimates System, December 2000:88.

2 Complications of Acute Fluid Loss and Replacement Juan Carlos Puyana Surgical/Trauma Intensive Care Unit, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, U.S.A.

During the past 100 years, fluids have been given intravenously for the management of fluid deficits. In 1883, Sidney Ringer discovered that calcium-containing tap water was better than distilled water for resuscitation. The understanding of the circulatory system and the importance of maintaining the circulatory volume were realized long ago. Furthermore, many years ago, researchers discovered the desired elements and their approximate concentrations in fluids that serve as intravenous plasma substitutes. However, the search for the optimal resuscitation fluid has been uneventful for a notable period of time. The first known intravenous infusions occurred in 1492. Blood from three youngsters was given to the dying pope by a vein-to-vein anastomosis in a desperate attempt to save him. The patient and the three donors died as a result of this transfusion. As early as 1667, the first known successful animal-to-animal transfusion was performed. In 1818, Dr. James Blundell performed the first successful transfusion of fluid into a human: the blood was given to a patient who was hemorrhaging during childbirth. In 1830, the gold-plated steel needle for intravenous use was invented. As Cosnett (1) reports, in 1831, a paper published by O’Shaughnessy described the effective use of salts and water to treat patients with cholera—an idea that was put into practice by Thomas Latta soon thereafter. During the 1930s, Baxter and Abbot produced the first commercial saline solutions. Two decades later, plastic intravenous tubing replaced rubber tubing, and soon thereafter, the central venous approach was described by a French military surgeon. This approach was a breakthrough in the estimation of the state of hydration [central venous pressure (CVP) measurements] and the application of volume support. Perhaps the most serious complications of intravenous fluid therapy are those related to bloodstream infections and septicemia (discussed in a later chapter). Blalock’s fundamental study of shock clearly showed that injury precipitates obligatory local and regional fluid losses, the effects of which can be ameliorated by vigorous restoration of intravascular volume. This concept became central to the understanding of the pathophysiology of shock, and provided new insight into the theory of shock

and a fundamental rationale for fluid-based therapy for hemorrhage and hypovolemia. As a result of noteworthy contributions made by surgeons during World War I and World War II, the introduction of blood transfusions dramatically changed the outcome of patients experiencing severe hemorrhage and traumatic shock. During the Korean Conflict, fluid overload became a common and lethal side effect because little was known about how infusates are dispersed and eliminated during trauma. In the period between the Korean Conflict and the Vietnam War, researchers discovered that there are tremendous fluid shifts into cells after severe hemorrhagic shock. As a consequence, the treatment of patients with shock was altered during the Vietnam War; these changes resulted in better outcomes and fewer cases of renal failure. Improved prehospital care, trauma system development, and emergency room management of shock have resulted in new issues related to the consequences of fluid loss, fluid replacement, and resuscitation. The understanding of shock today extends to a wide series of events that result from impaired cellular perfusion and compromised oxygen delivery with associated inflammatory and contrainflammatory responses, all of which ultimately result in severe organ dysfunction and failure if hypoperfusion is not promptly recognized and treated. This chapter will review some basic concepts of fluid physiology and theoretical concepts of fluid therapy in several clinical conditions associated with fluid loss and fluid deficit. After a brief review of the literature on resuscitation, a summary of the current concepts about the adequacy of resuscitation and end points of fluid replacement therapy will be presented.

THEORETICAL BASIS OF FLUID DISTRIBUTION Physicians’ comprehension of the effectiveness of resuscitative fluid therapy can be greatly enhanced if they understand several basic principles of fluid dynamics, body compartments, and membrane behavior. These concepts not only are necessary for an understanding of the effects of volume expanders on

18

Puyana

the circulation, but also are required for an appropriate interpretation of the conflicting results frequently seen in studies of fluid replacement therapy. Recognition of these principles and their application to circulation can result in accurate predictions in real-life situations of what has been observed theoretically and also in anticipating the physiologic effects of a specific fluid-based therapeutic intervention in the trauma room, the operating room, the intensive care unit (ICU), or the surgical ward. Severe fluid loss, fluid deficit, or both may result from a variety of clinical scenarios. Furthermore, a myriad of surgical conditions that are characterized by circulatory failure may trigger a complex inflammatory response that has been associated with the start of multiple organ dysfunction and death. Therefore, a fundamental knowledge of the physiologic basis of fluid therapy is necessary for preventing and minimizing the consequences of severe fluid losses and shock.

Body Fluid Compartments There are three body fluid compartments: the intravascular or plasma volume, the interstitial volume, and the intracellular volume. Under normal conditions, the interstitial volume is three times greater than the intravascular volume, and the intracellular volume is about two and one-half to three times greater than the interstitial volume. Thus, the intracellular volume is seven to nine times greater than the intravascular volume (Fig. 1). The intravascular volume is extraordinarily well defended by the body. Significant changes in the intravascular volume are not well tolerated. Loss of 30% to 40% of the intravascular volume will lead to severe hypovolemia and profound hypotension. Cardiac arrest usually occurs after 50% to 60% of the blood volume has been lost. In contrast, a 20% to 30% increase in volume leads to pulmonary edema (2). Sophisticated homeostatic mechanisms are therefore in charge of maintaining the intravascular volume.

Figure 1 Body composition.

All reflexive defense mechanisms in the body that maintain intravascular volume do so very efficiently: specifically, the renovascular response is designed to use the kidney to conserve fluid, and the neurovascular responses, the chemoreceptor response, the baroreceptor response, and the angiotensin response are aimed at maintaining a constant intravascular volume.

Fluid Maintenance and Regulation The interstitial volume is in continuous equilibrium with the intravascular volume, and, in fact, the interstitial volume acts like a large electrical capacitor that either absorbs fluid from the intravascular space when it is overhydrated, or fills the intravascular space when it is underhydrated. This constant equilibrium provides the flexibility necessary to withstand acute volume changes in the intravascular compartment. The interstitial volume has two very interesting properties. First, it is not extensively maintained by any of the reflexive mechanisms that exist to maintain the intravascular volume; therefore, the interstitial volume fluctuates widely during a particular disease course. Second, the interstitial volume has an impressive ability to expand. The compliance of the interstitial compartment is extraordinarily high, and until the interstitial compartment is filled to approximately three times its normal volume, the interstitial pressure remains low (3). Understanding the properties of the membranes that separate the body fluid compartments is crucial to predicting the effects of a specific volume expander. These membranes are quite different, and the events that regulate fluid exchange also differ in each compartment. The intravascular space and the interstitial space are separated by the capillary endothelium; therefore, all properties of the capillary endothelium are relevant. The capillary endothelium, however, functions differently in different organs of the body. For example, the capillary endothelium is much more permeable in the lung and the liver than it is in peripheral tissues such as muscle, skin, and subcutaneous fat (4). This difference in permeability gives rise to dramatically different effects in the response of organs to hemodilution. The greater the permeability is in a capillary bed, the less the capillary bed is affected by hemodilution because the interstitial concentration of protein is already higher. This characteristic is intrinsic to the lung (5). The membrane that separates the interstitial space from the cells is the cell surface membrane (Fig. 2), which functions differently than the capillary endothelium. The permeability of the capillary endothelium allows small molecules to pass through essentially unhindered. Water and bicarbonate molecules and sodium, potassium, and chloride ions all move freely through the capillary endothelium. Larger molecules, however, are restricted at the capillary endothelium. This restriction is apparent when the effect of albumin is analyzed. Albumin is a relatively large

Chapter 2: Complications of Acute Fluid Loss and Replacement

Figure 2 Membrane characteristics of body water compartments.

molecule that cannot readily pass through the pores of the capillary endothelium. However, substantial albumin leakage occurs through other membranes of the body, and this leakage can vary substantially depending on the endothelial characteristics of a specific organ. For example, albumin leakage from the lungs under normal circumstances is so high that the interstitial concentration is approximately 70% to 80% of the serum concentration. Therefore, the relative albumin concentration gradient across the pulmonary capillary membrane is relatively minor. The permeability of the liver endothelium to albumin is only slightly less than that of the lung; the interstitial concentration is approximately 60% of the serum concentration. In contrast, much less albumin leaks through peripheral tissues, specifically muscle and fat, than through the liver or the lung; therefore, the interstitial concentration is approximately 20% to 30% of the serum concentration. Unlike the capillary endothelium and other membranes of the body, the cell surface membrane is impermeable to proteins. The entry of proteins into cells is accomplished by active transport; therefore, oncotic pressure has a minimal effect at the cell surface membrane. The sodium pump is the active mechanism that operates at the cell surface, ejecting sodium from the cells while potassium is exchanged or diffuses passively to maintain charge equilibrium on both sides of the cell membrane. Bicarbonate molecules and chloride ions cross the cell membrane relatively easily, and because the membrane is permeable to water, this fluid crosses readily to maintain osmotic equilibrium. There are two crucial differences between the capillary endothelium and the cell surface membrane. First, the capillary endothelium is a passive membrane that does not require energy: it functions for an extended period of time independently of the delivery of oxygen and adenosine triphosphate (ATP). In

19

contrast, the cell surface membrane requires energy. As soon as the production of ATP is impaired, as it is, for example, during severe shock, the sodium pump stops working. At that point, the passive diffusion of sodium into the cells increases their osmotic pressure, and water immediately follows. This influx of water causes increased cellular edema. This phenomenon was described by Shires et al. (6) in animal models of severe irreversible shock. During severe shock, the amount of energy delivery to the cell membrane decreases, the cell surface membrane is immediately affected, and the cellular exchange of fluid is immediately impaired. The capillary endothelium is much more resilient to these changes than is the cell membrane. The second important difference between the capillary endothelium and the cell surface membrane is that large molecules that dictate the oncotic pressure or the differential gradient across the capillary endothelium do not play a crucial role at the cell membrane. The small molecules that pass through the cell surface membrane produce the total gradient. Osmotic pressure is the gradient that causes water to move across the cell membrane. Oncotic pressure and hydrostatic pressure across the capillary endothelium influence water movement.

VOLUME EXPANSION Theoretical Models of Fluid Replacement On the basis of the principles explained above, a theoretical model aimed at predicting the distribution of volume expanders can be constructed. These predictions result not from the findings of animal studies, but rather from theoretical calculations based on the rules that dictate membrane behavior and fluid exchange across biological membranes.

Resuscitation with Crystalloid Solutions Multiple effects on the intravascular, interstitial, and intracellular volumes can be observed when 2 L of a balanced salt solution (such as Ringer’s lactate solution) is infused. Thirty minutes after infusion, this crystalloid solution has equilibrated into the intravascular and interstitial space. Because all components of a balanced salt solution freely cross the capillary endothelium, the capillary endothelium in no way restricts the movement of a balanced salt solution. Therefore, when this salt solution is administered, in effect, it acts as though there is no boundary between the intravascular space and the interstitial space. The solution immediately crosses from one space to the other and is distributed between the two compartments in exact proportion to their starting volumes (Fig. 3). For example, if the starting volume of the interstitial space is three times that of the intravascular volume, as is normally the case in a healthy person, then the balanced salt solution will be distributed in

20

Puyana

Figure 3 Infusion of 2 L of Ringer’s lactate solution.

Figure 4 Infusion of 2 L of 5% albumin.

the same 1:3 ratio (Fig. 1). If 2 L of solution is given, then 500 mL will remain in the intravascular volume, and 1500 mL will move into the interstitial volume. Because a balanced salt solution is isoosmotic, no gradient in osmolarity is produced; therefore, there is no net movement across the cell surface membrane, which responds only to osmotic pressures. The volume of the intracellular compartment is therefore unchanged. This effect is crucial, but it has never been adequately emphasized in the literature.

the vascular space and three-fourths of the colloid solution (1500 mL) would remain in the intravascular space. Therefore, the ratio of intravascular filling with a colloid solution to intervascular filling with a crystalloid solution is 3:1. Almost every study in which the effects of crystalloid and colloid solutions have been compared has found that this ratio of intravascular expansion is consistent at approximately 3:1. This ratio fits the proposed theoretical model; it is what is measured hemodynamically, and it is predictable. Unfortunately, analyzing the results of published studies is difficult because most studies of crystalloid and colloid solutions have been based on protocols that did not consider this physiologic reality. Achieving the same effect on intravascular volume expansion requires the administration of three times as much crystalloid solution as colloid solution because only one-fourth of the balanced crystalloid solution will remain in the vascular space (the rest of the crystalloid solution will end up in the interstitial space). Administering a crystalloid solution therefore may induce significant interstitial edema. The concept of leakage or intravascular retention of a crystalloid solution is based on studies of shock in animal models using incomplete or inadequate resuscitation. When a volume deficit is created and then replaced with inadequate volume, even though the volume is similar to what has been lost, one may conclude that fluid has leaked. In a stable replacement model that includes the theoretical considerations explained above, the volume of distribution will be considered. Once full equilibration has been achieved, the relative volumes of distribution should remain the same. Depending on the volume used and the deficit replaced, the effects of volume resuscitation can therefore be predicted and anticipated on the basis of these three basic axioms: (i) one-fourth of the amount of balanced salt solutions administered remains in the vascular space; (ii) three-fourths of the amount of colloid solutions administered remains

Resuscitation with Colloid Solutions If a colloid solution such as 5% albumin is introduced into the intravascular volume, the relative leakage of the albumin solution will be proportional to the net albumin leakage in the entire body, i.e., approximately 25% to 35% depending on permeability. There is a reason that the net albumin leakage is much closer to the leakage of muscle, skin, and fat. The organs that are highly permeable to albumin (e.g., the liver and the lungs) make up a relatively small fraction of the body mass. The tissues that are not as permeable to albumin (e.g., muscle, skin, and fat) make up most of the body mass. As a result of this effect, an iso-oncotic solution administered into the intravascular compartment will leak in rough proportion to the total leakage of albumin in the body, i.e., approximately 25% to 35%. For example, if 2 L of a 5% albumin solution (colloid) is administered, the volume would be distributed as follows: 500 mL (25%) would leak into the interstitium, and 1500 mL would be retained in the intravascular volume. Because the albumin solution is iso-osmotic, there would be no net gradient into cells. Therefore, the cellular volume would not change (Fig. 4). When the net effect of volume expansion resulting from the administration of 2 L of Ringer’s lactate is compared with that resulting from the administration of 2 L of albumin, the results show that one-fourth of the balanced solution (500 mL) would remain in

Chapter 2: Complications of Acute Fluid Loss and Replacement

in the intravascular space; and (iii) the ratio of intravascular filling with crystalloid solutions to that with colloid solutions is 3:1.

Resuscitation with Hypertonic Saline Solution Hypertonic saline solution has been the topic of considerable discussion in recent years because of its potential for use as a prehospital fluid. When a 7.5% saline solution is administered, its hypertonic effect immediately exerts eight times the normal osmotic pressure of the body. Therefore, infusing this solution increases the osmotic pressure in the intravascular space, and this increase in pressure immediately pulls water from the intracellular space. Again because all of the ions that produce the osmotic gradient move freely across the capillary endothelium, there are no osmotic gradients across the capillary endothelium. Only the cell membrane restricts them; therefore, the gradient is acutely generated only across the cell membrane (Fig. 5). The osmotic pressures created by 7.5% saline infusions can be hundreds of millimeters of mercury, even when relatively small volumes of 7.5% saline solution are infused. As a result, fluid is pulled to the intravascular space with extraordinary rapidity. Whereas colloid equilibrium typically develops within 10 to 30 minutes of infusion, development of hypertonic saline equilibrium requires less than 60 seconds. In fact, this latter type of equilibration occurs so rapidly that it cannot be measured by any available technique. The best evidence indicates that the equilibration occurs within approximately three to five seconds after administration of the solution. Therefore, when a hypertonic solution is given, the pull of fluid into the vascular space is effectively instantaneously. The net effect, however, has a direct repercussion on the intravascular volume. Administering hypertonic solution forces equilibrium between

Figure 5 Infusion of 250 mL of 7.5% hypertonic saline.

21

the intracellular osmotic pressure and the intravascular osmotic pressure; because the saline solution is eight times more concentrated than the normal osmotic pressure, the solution pulls seven times its volume and dilutes itself by a factor of eight in the vascular space. For example, an intravenous infusion of 250 mL of a 7.5% saline solution pulls 1750 mL from the cellular space, thus resulting in an initial net increase in the intravascular volume of 2 L. Therefore, the volume reexpansion achieved by the administration of 250 mL of 7.5% saline solution is equivalent to that achieved by the administration of 2 L of an isotonic salt solution. Once the 7.5% saline solution has forced 1750 mL of fluid into the vascular space (from the intracellular space), redistribution occurs between the intravascular space and the interstitial space. Again, the now balanced salt solution is distributed in proportion to the sizes of the spaces. Therefore, of the 2 L of fluid pulled from the cellular space, 500 mL ends up in the intravascular space and 1500 mL ends up in the interstitial space. The only difference is that the net deficit occurs at the expense of the cells; therefore, the cellular compartment now contains 1750 mL less fluid. In contrast, this compartment is unaffected by the administration of Ringer’s lactate solution.

VOLUME REPLACEMENT AFTER ACUTE BLOOD LOSS The objective of fluid replacement therapy is to restore isovolemia. However, in the case of hemorrhage, the following question should be asked: ‘‘What amount of fluid do we have to administer to compensate for a given volume of blood loss?’’ The answer can be calculated on the basis of the previous observations for each particular type of fluid. Because albumin leakage is equal to approximately one-fourth of the administered volume of colloid solution, we would have to administer 133 mL of colloid solution to achieve effective volume reexpansion after the loss of 1 L of blood (Fig. 6). To compensate for the same amount of blood loss, we would have to administer 4 L of a balanced salt solution or 500 mL of a hypertonic salt solution (Fig. 6). Unfortunately, patients would not be able to tolerate the administration of that volume of hypertonic saline because it would induce severe hypernatremia and could cause seizures. The maximum volume of hypertonic saline that can be safely administered is approximately 250 mL. In 1967, Shires and his group (7) demonstrated the need for volume reexpansion in a classic model of hemorrhagic shock in dogs. The lowest survival rate occurred among animals whose lost fluid was replaced with blood alone; the survival rate was higher for the group treated with blood and plasma. The survival rate was highest for the group resuscitated with blood plasma and crystalloid solution (lactated Ringer’s): these dogs remained alive for as long as 14 days after the onset of hemorrhagic shock.

22

Puyana

Figure 6 (A-C) Examples of volume replacement after acute blood loss.

In their original report, Shires and his colleagues (7) reported that the extracellular fluid volume (ECV) was severely contracted after the hemorrhagic insult. A 10% blood loss was well tolerated, with no evidence of hypotension. When 25% of the total blood volume was lost, the animals experienced severe hypotension associated with a reduction of 18% to 26% in the functional ECV. Further losses of ECV paralleled increases in the amount of blood lost. Shires and colleagues proposed that intracellular swelling occurs as a result of shock (7,8). It is likely that the cellular swelling is a phenomenon that occurs during the shock period and is most probably related to a failure of the sodium pump. The finding that there is an expansion of the intracellular volume, which in some patients is believed to last for several days, has not been observed in milder or reversible models of shock. Indeed, the findings of Shires and coworkers showed

that the cellular volume returned to normal in subjects whose period of shock did not last for more than two hours (8). Cellular edema probably results from damage to the energetic machine of the membrane. In some patients, there may be an unrecognized, subclinical, ongoing energy deficit despite normal or near-normal values for clinical measures such as blood pressure, heart rate, and urinary output. These circumstances may occur more often than expected and may eventually generate a more protracted form of cellular dysfunction that ultimately will manifest itself as multiple organ failure. In clinical practice, we often find that patients who require massive volume resuscitation will ultimately exhibit the effects of massive expansion of interstitial space. In these patients, the interstitial space may have increased by a factor of two or three

Chapter 2: Complications of Acute Fluid Loss and Replacement

on the second or third day after surgery; thus, the normal 3:1 ratio explained above is no longer applicable. These patients may exhibit a ratio of 4:1, 5:1, or even 6:1. At this point, any additional crystalloid solution administered to the patient will be distributed in the interstitial space in proportion to these size ratios; therefore, the intravascular filling will become progressively less effective. The efficiency of volume expansion by crystalloid solution is reduced; instead of retaining one-fourth of the volume in the intravascular space, the patient may retain no more than onefifth or one-sixth of the volume administered (Fig. 7). There is a point at which the efficiency of volume expansion becomes so low that administering a colloid solution rather than a crystalloid solution may be a better option because the colloid solution causes more intravascular filling.

GOALS OF FLUID REPLACEMENT AND END POINTS OF RESUSCITATION Porter and Ivatury (9) recently reviewed the available findings regarding end points for the resuscitation of patients with traumatic injury. Although their study focused only on the acute resuscitation of victims of trauma, their conclusions are probably also applicable to critically ill and high-risk patients in general. Most clinicians would agree that heart rate, systemic arterial blood pressure, skin temperature, and urine flow (i.e., the primary end points of resuscitation used by clinicians before the era of invasive hemodynamic monitoring) provide relatively little information about the adequacy of oxygen delivery to tissues. Accordingly, reliance on these simple indices of perfusion may result in failure to recognize ongoing anaerobiosis (cryptic shock). With the introduction of central venous and Swan–Ganz catheterization, clinicians sought to titrate

Figure 7 Effect of infusing 2L of Ringer’s lactate solution in a patient with severe edema.

23

resuscitation therapy to achieve ‘‘adequate’’ indices of ventricular preload, cardiac output, and systemic oxygen delivery. On the basis of extensive analyses of the hemodynamic profiles of survivors and nonsurvivors of critical illness, Shoemaker et al. (10) proposed that patients suffering from trauma and shock develop an oxygen ‘‘debt’’ and therefore require supranormal levels of oxygen delivery to reestablish homeostasis. Tuchschmidt et al. (11) later reported similar findings from a study of patients with sepsis. Subsequently, in three prospective, randomized trials (12–14), Shoemaker and coworkers obtained evidence that survival is improved by titrating resuscitative measures to achieve the target values established in earlier observational studies (specifically, a cardiac index greater than 4.5 L/min/m2 of body area, a systemic oxygen delivery index greater than 600 mL/min/m2, and systemic oxygen consumption greater than 170 mL/min/m2). In another trial (15), a significant improvement in survival rates was achieved when high-risk surgical patients were treated with dopexamine, an inotrope and vasodilator that increases cardiac output during the perioperative period. No significant differences in systemic oxygen consumption or blood lactate concentration were found between patients who were and were not treated with dopexamine; this finding suggests that dopexamine has beneficial effects, independent of its hemodynamic actions. To complicate the picture, several subsequent clinical studies (16–19) have failed to demonstrate that survival rates improve when resuscitation therapy is titrated to achieve supranormal values for oxygen delivery or cardiac output. Resuscitation therapy can also be adjusted to achieve certain biochemical end points, such as arterial base deficit or blood lactate concentration. These end points can be used because tissue hypoperfusion leads to increased anaerobic metabolism. During anaerobic metabolism, large quantities of pyruvate are converted to lactate, and thus do not enter the tricarboxylic acid cycle. Meanwhile because of the stoichiometry of substrate-level (rather than oxidative) phosphorylation of adenosine diphosphate to ATP, there is a net accumulation of protons (20). Accordingly, increases in arterial base deficit, blood lactate concentration, or both are evidence of an increase in the rate of anaerobic metabolism. Numerous studies (21–24) have documented that high concentrations of blood lactate portend an unfavorable outcome for patients with shock, but it has not been proven that survival is improved when therapy is titrated by using blood lactate concentration as an end point. Base deficit is the amount of base (in millimoles) required to titrate 1 L of whole blood to a pH of 7.40 while the sample is maintained at 37
C, fully saturated with oxygen, and equilibrated with an atmosphere containing carbon dioxide at a PCO2 of 40 mmHg. Base deficit is calculated by arterial blood gas analyzers that use a nomogram developed by Astrup et al. (25). Base deficit is more quickly and

24

Puyana

easily measured than is lactate concentration, and it has prognostic value for patients with shock (26–29). Although titrating therapy to a base deficit end point is intuitively reasonable, whether it improves survival rates remains unproven. Until recently, a randomized clinical trial of gastric tonometry (30), a form of tissue capnometry, was the only source of published findings demonstrating that the use of a monitoring tool to guide resuscitation can improve the outcome of critically ill patients. However, Rivers et al. (31) recently published the results of a partially blinded, randomized trial of goal-directed therapy initiated in the emergency ward for patients with septic shock. An algorithm was developed to adjust CVP to 8 to 12 mmHg, mean arterial pressure to 65 to 90 mmHg, and central venous oxygen saturation to more than 70%. A central venous oximetry catheter was used to titrate resuscitative therapy in an attempt to balance systemic oxygen supply with oxygen demand. Unlike similar studies carried out in an ICU setting, this study initiated goal-directed therapy at an earlier point after injury. The findings showed that early institution of goal-directed hemodynamic support prevented cardiovascular collapse in high-risk patients, and reduced hospital mortality rates from 46.5% to 30.5% (p ¼ 0.009) (31,32). Perhaps the most rational way to titrate resuscitative therapy is to use a measure of the adequacy of regional tissue perfusion. Several highly complex approaches can achieve this goal, such as using near-infrared spectroscopy to assess the redox state of cytochrome a, a3 (the terminal enzyme complex in the mitochondrial respiratory chain). However, tissue capnometry offers the promise of being inexpensive, reliable, and minimally invasive.

COMPLICATIONS OF FLUID LOSSES IN SURGICAL PATIENTS Many surgical complications are associated with fluid losses and electrolyte disturbances that result from conditions other than hemorrhage. These conditions may have a broader but less acute impact on all of the body fluid compartments, and may be associated with severe electrolyte derangements. Therefore, therapy must be directed at restoring homeostasis and minimizing iatrogenic complications. Because sodium and water are the primary determinants of the adequacy of volume status, the surgeon must have a clear understanding of the interactions of these two important components of the internal milieu. External losses or internal shifts of fluids are commonly associated with surgical patients and may initially be indicated by signs of inadequate volume and perfusion, without marked changes in plasma sodium concentrations. The most common cause of hyponatremia is inappropriate therapy, and the most common conditions associated with hypernatremia are excessive diuresis and unrecognized or miscalculated

losses of free water. The plasma sodium concentration is an index of the relative proportions of sodium and water in the extracellular fluid (ECF). The constant redistribution of fluid across all body fluid compartments is ruled by the laws of tonicity. The combined depletion of sodium and water is a common occurrence among patients with volume depletion resulting from excessive fluid losses. In a similar fashion, the loss of gastrointestinal fluid, which may occur as the result of vomiting, diarrhea, fistulas, or prolonged nasogastric suction, is also typically characterized by combined deficits of water and sodium. Third spacing or sequestration of fluid in patients with severe peritonitis, pancreatitis, or other local inflammatory conditions in the abdominal cavity is also associated with combined losses of sodium and water.

Hyponatremia Hyponatremia occurs when there is an excess of total body water relative to total body sodium content. Hyponatremia may be associated with decreased, increased, or near-normal amounts of total body sodium. In general, hyponatremia occurs as a disorder of the kidneys’ ability to dilute urine. The approach to therapy for the hyponatremic patient can be simplified by establishing first whether the patient’s ECV is reduced (depletion) or increased (edema). Renal losses, which include diuretic excess, mineralocorticoid deficiency, salt-losing nephritis, renal tubular acidosis, and osmotic diuresis, are characterized by volume depletion and a urinary sodium concentration greater than 20 mmol/L. When the urinary sodium concentration is less than 10 mmol/L, the hyponatremic state is usually the result of extrarenal losses, a common condition among surgical patients. Those with extrarenal losses usually have a history of vomiting or nasogastric suction, third spacing of fluid, pancreatitis, burns, or soft-tissue trauma. Both types of hyponatremic conditions (renal and extrarenal losses) respond to fluid replacement with isotonic solution (33). Patients with normal or mildly diminished ECV, abnormally low sodium concentration, and no edema are not commonly seen by surgery services. Such patients often have a glucocorticoid deficiency or inappropriate antidiuretic hormone secretion. These conditions usually respond to water restriction. Finally, hyponatremia can occur in patients with edema and enhanced ECF; these patients typically have conditions associated with impaired renal perfusion, such as congestive heart failure, cirrhosis, or nephrotic syndrome. In these patients, a urinary sodium concentration of less than 10 mmol/L is common. If the urinary sodium concentration is greater than 20 mmol/L in patients with edema, then a component of acute or chronic renal failure is also present.

Hypernatremia The renal concentrating mechanism is the first defense against water depletion and hyperosmolality. When

Chapter 2: Complications of Acute Fluid Loss and Replacement

this mechanism is impaired, thirst becomes a very effective mechanism for preventing further increases in serum sodium concentration. Unfortunately, most clinical conditions experienced by surgical patients are also associated with an impairment in water intake. The most practical approach to treating the patient with an elevated serum sodium concentration relies on a basic assessment to determine whether the patient is experiencing hypernatremia with sodium and water losses, hypernatremia with mostly water losses, or hypernatremia with mostly increased sodium intake. Again, identifying the site of sodium or water losses is crucial; for example, patients receiving diuretic therapy (osmotic or loop diuretics) or having postobstruction or intrinsic renal disease will produce isotonic or hypotonic urine, their urinary sodium concentration will be greater than 20 mEq/L, and their total body sodium concentration will be low (33). If the urinary sodium concentration is less than 10 mEq/L, the most likely cause of hypernatremia will be extrarenal losses (sweating, heat exposure, burns, diarrhea, or fistula). If most of the losses are free water, then the total body sodium concentration must be close to normal values. These patients’ urinary sodium concentration will vary, and their clinical symptoms will resemble those of patients with diabetes insipidus syndromes, or those with insensible water losses that are purely respiratory and dermal. If hypernatremia is present and an increase in total body sodium concentration is suspected, the patient usually has primary hyperaldosteronism, Cushing syndrome, or hypertonic dialysis; alternatively, the patient may chronically ingest large amounts of sodium bicarbonate or sodium chloride tablets. The urinary sodium concentration usually exceeds 20 mEq/L. Management consists of the replacement of free water and the initiation of diuretic therapy.

INHERENT COMPLICATIONS OF COMMONLY USED FLUID REPLACEMENT SOLUTIONS During the past 10 years, we have seen the publication of a plethora of reports presenting more specific findings about the many physiological and biochemical effects of several solutions used for fluid replacement therapy. Furthermore, new and intriguing findings concerning such solutions have been obtained from recent animal model studies, evidence-based medicine studies, meta-analyses, and clinical trials. The following is a summary of these recent findings.

Complications Associated with the Use of Albumin Solutions Recently, a Cochrane report on the use of albumin as treatment for critically ill patients was published (34). The authors carried out a systematic review of 30 randomized controlled trials comparing the effects of administering albumin or plasma protein fraction

25

with the effects of not administering this protein fraction or administering crystalloid solution to 1419 critically ill patients with hypovolemia, burns, or hypoalbuminemia. For each patient category, the risk of death in the albumin-treated group was higher than that in the comparison group. The relative risk of death after albumin administration was 1.46 (95% confidence interval, 0.97–2.22) for hypovolemic patients, 2.40 (1.11–5.19) for patients with burns, and 1.69 (1.07–2.67) for patients with hypoalbuminemia. However because this review was based on relatively small trials in which the number of deaths was small, these results must be interpreted with caution. The analyses suggest that the use of human albumin to treat critically ill patients should be reconsidered. Interestingly, a recent abstract submitted to the Society of Critical Care Medicine reported the results of a study of the effectiveness and safety of plasbumin-5 in resuscitating adults with shock. In this open-label, randomized, multicenter, controlled trial, 19 patients were treated with albumin, and 23 were given Ringer’s lactate solution. Baseline multiple organ dysfunction scores for the two groups were equivalent. There were no statistically significant differences between groups with respect to days on mechanical ventilation, oxygenation failure, length of stay in the ICU, or 28-day mortality rates. The incidence of bacteremia was significantly lower in the group treated with albumin (p ¼ 0.023). The authors requested access to the Cochrane database, and added the results of their trial to the meta-analyses of the subgroup of critically ill patients with burns (34,35). When the data from the recent trial were added to the data considered in the Cochrane review, there was no longer a statistically significant difference in the relative risk of death between burn patients who received albumin and those who received crystalloid solution (35). In summary, the benefit of administering albumin to critically ill patients is unproven. Epidemiologic evidence suggests that using human albumin solution to treat patients with burns, hypoalbuminemia, and hypotension is associated with an increased risk of death. In the face of critical illness, hypoalbuminemia results from transcapillary leak, decreased synthesis, large-volume body fluid losses, and dilution caused by fluid resuscitation. When treating patients with hypoalbuminemia, physicians must focus their efforts on correcting the underlying disorder rather than on reversing the hypoalbuminemia (36).

Complications Associated with the Use of Ringer’s Lactate Solution It is somewhat surprising that although Ringer’s lactate solution has been used as a volume expander and resuscitation fluid for nearly 100 years, only recently have we begun to elucidate the immunologic and proinflammatory effects of the solution on neutrophils and other cells involved in host defense mechanisms. In one study using a swine model of

26

Puyana

shock, Rhee et al. (37) compared the effects of different methods of fluid resuscitation in three groups of patients: group I received Ringer’s lactate solution; group II, shed blood; and group III, 7.5% hypertonic saline solution. Neutrophil activation in whole blood was measured by flow cytometry, which detected intracellular superoxide burst activity. Neutrophil activation increased significantly immediately after hemorrhage in all three groups, but the increase was greatest after resuscitation with Ringer’s lactate. Neutrophil activity in animals that received shed blood or 7.5% hypertonic saline solution returned to the baseline state after resuscitation (37). Another study by the same research group (38) found that different resuscitative fluids may immediately affect the degree of apoptosis after hemorrhagic shock in rats. Fluid resuscitation with Ringer’s lactate solution significantly increased apoptosis of cells in the small intestine and liver. Administering Ringer’s lactate solution to rats in a sham hemorrhage group increased apoptosis of cells in the intestinal mucosa and muscularis externa. Animals that underwent sham hemorrhage, resuscitation with blood or hypertonic saline solution, or sham resuscitation experienced no increase in apoptosis in either the liver or the small intestine.

CONCLUSIONS The use of intravenous fluids is one of the main pillars of resuscitative therapy for surgical patients. Many conditions, such as acute hemorrhage, burn injuries, and intra-abdominal inflammatory catastrophes, require aggressive fluid resuscitation. The clinician must be very familiar with the type and proper dosage of the many available solutions. Clear objectives and end points of resuscitation strategies must be determined in advance, and the possible side effects and complications need to be predicted and identified early in the course of treatment if the best possible outcome is to be achieved. Unfortunately, fluid therapy is not always seen as a pharmacologic intervention, but we must realize that fluids, like any other drug, may be indicated or contraindicated in specific situations. Finally, we should note that, in contrast to the pace of new drug development, the pace of innovation in fluid therapy has been remarkably slow. For example, most studies of antisepsis drugs date only to the early 1980s, and new trials are reported every two to three months. The use of mechanical ventilation devices has been widespread since the 1960s, and a new model of ventilator appears every 22 months. Antibiotics have been in use since the 1940s, and a new agent is approved every six months. In contrast, intravenous fluid therapy has been used for resuscitation since the 1800s, and we still use these fluids today. After a century of Ringer’s lactate use, the development of a new product is overdue (Kellum JA. Personal communication, 2003).

REFERENCES 1. Cosnett JE. The origins of intravenous fluid therapy. Lancet 1989; 333:768–771. 2. Tranbaugh RF, Lewis FR. Mechanisms and etiologic factors of pulmonary edema. Surg Gynecol Obstet 1984; 158:193–206. 3. Manning RD Jr, Guyton AC. Dynamics of fluid distribution between the blood and interstitium during overhydration. Am J Physiol 1980; 238:H645–H651. 4. Baldwin AL, Thurston G. Mechanics of endothelial cell architecture and vascular permeability. Crit Rev Biomed Eng 2001; 29:247–278. 5. Tranbaugh RF, Elings VB, Christensen J, Lewis FR. Determinants of pulmonary interstitial fluid accumulation after trauma. J Trauma 1982; 22:820–826. 6. Shires GT, Cunningham JN, Backer CR, et al. Alterations in cellular membrane function during hemorrhagic shock in primates. Ann Surg 1972; 176:288–295. 7. McClelland RN, Shires GT, Baxter CR, Coln CD, Carrico J. Balanced salt solution in the treatment of hemorrhagic shock. Studies in dogs. JAMA 1967; 199:830–834. 8. Illner HP, Cunningham JN Jr, Shires GT. Red blood cell sodium content and permeability changes in hemorrhagic shock. Am J Surg 1982; 143:349–355. 9. Porter JM, Ivatury RR. In search of the optimal end points of resuscitation in trauma patients: a review. J Trauma 1998; 44:908–914. 10. Shoemaker WC, Montgomery ES, Kaplan E, Elwyn DH. Physiologic patterns in surviving and nonsurviving shock patients. Use of sequential cardiorespiratory variables in defining criteria for therapeutic goals and early warning of death. Arch Surg 1973; 106:630–636. 11. Tuchschmidt J, Fried J, Swinney R, Sharma OP. Early hemodynamic correlates of survival in patients with septic shock. Crit Care Med 1989; 17:719–723. 12. Fleming A, Bishop M, Shoemaker W, et al. Prospective trial of supranormal values as goals of resuscitation in severe trauma. Arch Surg 1992; 127:1175–1179; discussion 1179–1181. 13. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest 1988; 94:1176–1186. 14. Bishop MH, Shoemaker WC, Appel PL, et al. Prospective, randomized trial of survivor values of cardiac index, oxygen delivery, and oxygen consumption as resuscitation end points in severe trauma. J Trauma 1995; 38:780–787. 15. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA 1993; 270:2699–2707. 16. Hayes MA, Timmins AC, Yau EH, Palazzo M, Hinds CJ, Watson D. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med 1994; 330:1717–1722. 17. Tuchschmidt J, Fried J, Astiz M, Rackow E. Elevation of cardiac output and oxygen delivery improves outcome in septic shock. Chest 1992; 102:216–220. 18. Yu M, Levy MM, Smith P, Takiguchi SA, Miyasaki A, Myers SA. Effect of maximizing oxygen delivery on morbidity and mortality rates in critically ill patients: a prospective, randomized, controlled study. Crit Care Med 1993; 21:830–838.

Chapter 2: Complications of Acute Fluid Loss and Replacement

19. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goaloriented hemodynamic therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med 1995; 333: 1025–1032. 20. Hochachka PW, Mommsen TP. Protons and anaerobiosis. Science 1983; 219:1391–1397. 21. Marecaux G, Pinsky MR, Dupont E, Kahn RJ, Vincent JL. Blood lactate levels are better prognostic indicators than TNF and IL-6 levels in patients with septic shock. Intensive Care Med 1996; 22:404–408. 22. Bernardin G, Pradier C, Tiger F, Deloffre P, Mattei M. Blood pressure and arterial lactate level are early indicators of short-term survival in human septic shock. Intensive Care Med 1996; 22:17–25. 23. Broder G, Weil MH. Excess lactate: an index of reversibility of shock in human patients. Science 1964; 143:1457–1459. 24. Bakker J, Gris P, Coffernils M, Kahn RJ, Vincent JL. Serial blood lactate levels can predict the development of multiple organ failure following septic shock. Am J Surg 1996; 171:221–226. 25. Astrup P, Engel K, Jorgensen K, Siggaard-Andersen O. Definitions and terminology in blood acid-base chemistry. Ann N Y Acad Sci 1966; 133:59–65. 26. Siegel JH, Rivkind AI, Dalal S, Goodarzi S. Early physiologic predictors of injury severity and death in blunt multiple trauma. Arch Surg 1990; 125:498–508. 27. Davis JW, Parks SN, Kaups KL, Gladen HE, O’Donnell-Nicol S. Admission base deficit predicts transfusion requirements and risk of complications. J Trauma 1996; 41:769–774. 28. Davis JW, Kaups KL, Parks SN. Base deficit is superior to pH in evaluating clearance of acidosis after traumatic shock. J Trauma 1998; 44:114–118.

27

29. Rutherford EJ, Morris JA Jr, Reed GW, Hall KS. Base deficit stratifies mortality and determines therapy. J Trauma 1992; 33:417–423. 30. Gutierrez G, Palizas F, Doglio G, et al. Gastric intramucosal pH as a therapeutic index of tissue oxygenation in critically ill patients. Lancet 1992; 339:195–199. 31. Rivers E, Nguyen B, Havstad S, et al. Early GoalDirected Therapy Collaborative Group. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345:1368–1377. 32. Rivers EP, Ander DS, Powell D. Central venous oxygen saturation monitoring in the critically ill patient. Curr Opin Crit Care 2001; 7:204–211. 33. Berl T, Anderson RJ, McDonald KM, Schrier RW. Clinical disorders of water metabolism. Kidney Int 1976; 10:117–132. 34. Alderson P, Bunn F, Lefebvre C, et al. Human albumin solution for resuscitation and volume expansion in critically ill patients. Cochrane Database Syst Rev 2004; (4):CD001208. 35. Cooper A. Efficacy and safety of Plasbumin-5 for adult burn shock resuscitation. Crit Care Med, Suppl 2003; 31(2) (abstract 71). 36. Pulimood TB, Park GR. Debate: albumin administration should be avoided in the critically ill. Crit Care 2000; 4:151–155. 37. Rhee P, Burris D, Kaufmann C, et al. Lactated Ringer’s solution resuscitation causes neutrophil activation after hemorrhagic shock. J Trauma 1998; 44:313–319. 38. Deb S, Martin B, Sun L, et al. Resuscitation with lactated Ringer’s solution in rats with hemorrhagic shock induces immediate apoptosis. J Trauma 1999; 46:582–588; discussion 588–589.

3 Complications of Antibiotic Therapy Mohamed Fahim Anesthesia Critical Care, Davis Memorial Hospital, Elkins, West Virginia, U.S.A. Nicholas Namias Miller School of Medicine at the University of Miami, Miami, Florida, U.S.A.

Antibiotics are frequently used as an adjunct to the surgical therapy of infections. All antibiotics are potentially harmful, and various benefit-to-risk factors must be considered whenever they are used (1). Antimicrobial chemotherapy used in association with surgery may be complicated by failure of therapy or unwanted side effects. Most antibioticrelated adverse reactions are predictable and are often dosedependent. Unpredictable reactions occur independently of the dose and route of administration and are due to drug intolerance, allergy, and other idiosyncratic responses (2). Other reactions occur rarely and are unique to the compound administered; one such example is toxic epidermal necrolysis (Stevens–Johnson Syndrome) induced by sulfonamides (1). The problems encountered in the use of antimicrobial chemotherapeutic agents can be conveniently divided into general complications (e.g., those associated with the route of administration, hypersensitivity reactions, failure of therapy, induction of resistance, antagonism, and effects on immune response) and specific complications related to each individual antimicrobial agent.

GENERAL COMPLICATIONS Complications Associated with the Route of Administration Oral administration of an antimicrobial agent may cause complications involving the gastrointestinal tract, such as nausea, vomiting, diarrhea, gastritis, and pseudomembranous colitis (PMC). Parenteral administration of antimicrobial agents can result in reactions at the injection site, such as pain, inflammation, abscess, necrosis, edema, hemorrhage, cellulitis, hypersensitivity, atrophy, ecchymosis, and skin ulcer. Parenteral administration of antibiotics may also result in neurovascular reactions, including warmth, vasospasm, pallor, mottling, gangrene, numbness of the extremities, cyanosis of the extremities, and neurovascular damage. Intravenous administration of

antibiotics may cause thrombophlebitis. It is well known that intramuscular injection of antibiotics such as penicillin G may result in sciatic nerve injury (3).

Hypersensitivity Reactions Anaphylaxis Anaphylaxis is the most severe reaction experienced by patients treated with antibiotics. It is most frequently encountered after parenteral injection of penicillin or one of its synthetic analogs (4). Anaphylaxis occurs when certain pharmacologically active mediators are rapidly released in response to interactions between an antigen (the antimicrobial compound) and immunoglobulin E (IgE) antibody. These mediators, which are released from basophils and mast cells, include histamine, a slowly reacting substance of anaphylaxis, and, perhaps, serotonin and bradykinin (5). Clinically, the reaction may develop within minutes to hours of drug administration. Patients may experience primary vascular collapse with hypotension, as well as bronchospasm and laryngeal edema. Dermal manifestations include eruption and hives. Angioedema may also occur. The drug of choice for the treatment of anaphylaxis is epinephrine, diluted 1:1000, given intramuscularly in a volume of 0.3 mL. However, when anaphylaxis is very severe, intravenous administration of epinephrine may be required if a clinical response is to be to achieved. Progression of the laryngeal edema may even require immediate tracheostomy or emergent cricothyroidotomy because endotracheal intubation may not be possible. Persistent hypotension that does not respond to epinephrine may be treated with volume expanders and vasoconstrictors (6). Prevention of anaphylaxis depends on eliciting a thorough drug history before antibiotics are administered. The most reliable way to assess a patient’s risk for a type I IgE-mediated reaction is to administer a skin test to measure the response to the ‘‘major’’ and ‘‘minor’’ skin determinants. Unfortunately, only the major skin-testing determinant is commercially available (7).

30

Fahim and Namias

Cutaneous Eruptions Cutaneous (dermal) eruptions, the most common manifestation of hypersensitivity to antibiotics, are due to delayed hypersensitivity mediated by the cellular immune system (8). Such eruptions may be morbilliform, petechial, maculopapular, or bullous. Stevens–Johnson syndrome, exfoliative dermatitis, and erythema nodosum may also occur. When a dermal reaction appears, the best course of action is to discontinue the administration of the drug (4). Taking a careful drug history and avoiding drugs to which the patient has a history of hypersensitivity can prevent a reaction. An agent can be selected from another class of antibiotics that would be unlikely to cross-react with the drug to which the patient is allergic. Penicillins and cephalosporins both possess the betalactam ring, and cross-reactivity may occur, but this event probably occurs in no more than 5% to 10% of patients allergic to one of these types of antibiotics (9). Cephalosporins are usually safe for patients allergic to penicillins, but not for those who have had an immediate or anaphylactic reaction to the penicillins (4).

Drug Fever Drug fever is presumably a hypersensitivity reaction, and may be the most difficult complication to diagnose because of its similarity to fever due to the infection. Also, other drugs administered simultaneously must be considered as possible causes of fever. Drug fever may be associated with the use of any antimicrobial agent (4) and may be accompanied by eosinophilia or cutaneous eruption. Diagnosis of drug fever should always be considered when fever is sustained or recurs despite the apparent effectiveness of antimicrobial therapy (10). Drug fever usually cannot be prevented; it is treated with prompt withdrawal of the antibiotic suspected of being responsible for causing it.

Failure of Antimicrobial Therapy Antimicrobial therapy alone is invariably ineffective in treating abscesses because antibiotics cannot penetrate the abscess and thus cannot reach effective concentrations in the purulent exudates (4). Also, some antibiotics appear to be inactivated by constituents of the exudate in the abscess (11).

Induction of Resistance to Antimicrobial Agents One of the most difficult problems in dealing with infectious diseases is microbial resistance to antimicrobial agents (12–14). In response to the specific and well-defined needs of the clinician, a number of procedures have been developed to guide the selection of antibiotic agents and their dosages for the treatment and eradication of a given bacterial infection. Among the procedures available, the most useful have been those that yield the qualitative antibiotic susceptibility profile and the quantitative minimum

inhibitory concentration (MIC) of the clinically indicated antibiotics (15). The MIC is the concentration of the antibiotic that inhibits the growth of a standardized concentration of bacteria. In contrast, the minimum bactericidal concentration (MBC) is the concentration of a drug that kills 99.9% of a bacterial population after exposure to the antibiotic for 24 hours. Therefore, the MBC is often used to treat life-threatening infections such as endocarditis. The MBC, with a few exceptions, is greater than the MIC. Comparing the MIC with the MBC provides an estimate of the drug’s potency: a small difference between the two indicates that the antibiotic is potent; a large difference indicates that the bacterium is tolerant of the antimicrobial compound (15). Acquired resistance arises from the microbe’s acquisition of genetic material or its mutation. Genetic exchange can arise from conjugation, transformation, phage transduction, or mutations within the genome of the bacterium. Plasmids and transposons can carry determinants that confer resistance in many ways; for example, enzymatic inactivation, decreased uptake, cell surface alterations, and efflux properties have all been reported to be transferable by plasmids. Mutation can also result in the loss of outer membrane proteins, or the alteration of target site (16,17). Worldwide, many strains of Staphylococcus aureus are already resistant to all antibiotics except vancomycin. Methicillin-resistant Staphylococcus aureus (MRSA) was first detected in England (18), and constituted 46.7% of all S. aureus isolates collected in 1998 by the National Nosocomial Infections Surveillance Program of the Centers for Disease Control and Prevention; this percentage represented a 31% increase over the previous reported value for the years 1993 to 1997 (19). Beta-lactamase production is also an important mechanism of resistance in gram-negative organisms. There is now an extended spectrum of organisms whose beta-lactamase will hydrolyze the beta-lactam ring of multiple beta-lactam drugs (20). Vancomycinintermediate and vancomycin-resistant strains of S. aureus have been isolated, although they have not yet become widespread (21). Resistance to vancomycin has also become common in the enterococci (19), and there have been case reports of patients infected with vancomycin-dependent enterococci (22).

Antagonism ‘‘Antibiotic synergy’’ connotes a greater-than-additive treatment effect that occurs when more than one antibiotic is used. Strictly defined, ‘‘antagonism’’ means a less-than-additive effect, but it is more often taken to mean that the effect produced by a group of antibiotics is less than that caused by the most active drug in the group (23). Synergy and antagonism can be categorized into five major groups: direct effects on the bacteria (e.g., interaction between the mechanisms responsible for antibiotics’ specific antibacterial activity); indirect effects on the bacteria (e.g., drug inactivation, emergence of resistance, or environmental

Chapter 3:

changes that modify antibiotic action); pharmacological effects (e.g., inactivation, bioconversion, or both; absorption, elimination, or both; and diffusion, binding, or both); modification of host defenses that enhance or adversely affect the antibiotic action (e.g., phagocytosis or immunologic response); and effects that cause toxicity (24). The synergy or antagonism between antimicrobial drugs is a direct consequence of combining the agents used. There are at least five major reasons for using combinations of antimicrobial agents (25). First, such combinations may be used as initial therapy for serious infections. Empiric therapy for severely ill patients or for compromised hosts should involve a combination of antimicrobial drugs. No single drug can be reasonably expected to provide sufficient coverage against gram-positive and gram-negative bacteria that cause serious infections. Second, combinations of antibiotics may be used to treat infections caused by multiple organisms. Mixed infections are a reasonable justification for the use of antimicrobial combinations because not all the pathogens may be susceptible to a single agent. The best examples of infections requiring treatment with multiple antibiotics are the mixed aerobic–anaerobic infections, which arise from the gastrointestinal or genital tracts. Third, combinations of antibiotics may be used to decrease the emergence of antibiotic-resistant microbes. This rationale has been the basis for the use of combinations of antituberculous drugs. Fourth, the use of antibiotic combinations may lessen the dose-related toxicity of treatment. This reason for using combinations is rarely advocated; most clinicians prescribe the maximum tolerated dose of a single drug and do not rely on synergy. Finally, combinations of antibiotics may be used to eradicate infections that cannot be successfully treated with a single active drug (synergism). The concept of synergy and its clinical relevance is the most controversial aspect of the use of antimicrobial combinations. In some situations, antimicrobial agents may act synergistically on a given microorganism through their different binding to receptors in the cell membrane. Another mechanism is the sequential blockade of successive steps in a metabolic pathway in a given organism, such as that accomplished through the synergism achieved by the combined use of the combination of sulfamethoxazole and trimethoprim. Finally, synergism can result from the combined use of an antimicrobial agent that inhibits the synthesis of the bacterial cell wall and another agent that is otherwise unable to penetrate cells with an intact wall. For example, penicillins and various aminoglycosides exhibit this synergism against infections caused by enterococci and other microorganisms. Many mechanisms may explain antagonism. Drugs that are capable of impairing cell division may antagonize the activity of antibiotics that alter the cell wall, such as the penicillins. However, the mechanisms of antagonism resulting from combination of drugs that act within the bacterial cell appear to be more complex.

Complications of Antibiotic Therapy

31

Effects of Antimicrobial Agents on the Immune Response Life-threatening bacterial, fungal, and opportunistic infections that occur among patients whose immune response is altered has led to the prolonged use of intensive prophylactic and therapeutic antimicrobial regimens, which often are administered concomitantly with corticosteroids, cytotoxic chemotherapy, or both. Because some antimicrobial agents at therapeutic concentrations can affect humoral or cellular immune responses, there is a need for a better understanding of the possible beneficial and deleterious effects of antibiotic therapy on the immune response, especially in the immunocompromised patient. The impetus to evaluate the effect of antimicrobials on the immune response stems largely from observations of the toxic or allergic reactions to commonly used antimicrobial agents (26–31); the similarity between the structures of certain antibiotics and chemical structure of various antimetabolites and cytotoxic drugs (32,33); and the ability of certain antimicrobials to enter mammalian cells, especially phagocytes, thus raising the possibility of untoward effects on cellular function (34–41). Various detailed reviews of the literature on this subject have generally found that a large number of antimicrobial agents can affect various immune functions (42–47). However, the meaning, clinical relevance, and significance of the observed changes are unclear: the results of different studies have been conflicting, the effects of antibiotics in vitro have been incorrectly assumed to occur in vivo, the mechanisms by which antibiotics affect immune responses remain unknown, and few controlled clinical studies have been carried out to determine the effect of antimicrobials on the immune response in humans. Because of these limitations in interpreting the currently available data, we selectively review some of the more commonly used antimicrobial agents, specifically focusing on their effects on cell-mediated and humoral immunity, as well as on polymorphonuclear leukocyte function as determined by in vitro and in vivo investigations.

Lymphocyte Transformation Lymphocyte transformation in vitro is the metabolic activation of lymphocytes with antigens or mitogens. Thus, lymphocyte transformation reflects the ability of lymphocytes to proliferate after exposure to such antigens or mitogens (48). This effect occurs in response to treatment with various antibiotics. For example, acyclovir appears to delay the development of and diminish the peak of in vitro lymphocyte transformation responses to inactivated herpes simplex virus antigens in patients with genital herpes (49).

Delayed-Type Hypersensitivity Delayed-type hypersensitivity (DTH) is a cell-mediated immune reaction that can be elicited by intracutaneous injection of antigen; the subsequent cellular infiltrate

32

Fahim and Namias

and edema reach their peak between 24 and 48 hours after antigenic challenge (50). DTH can be thought of as an in vivo model of cell-mediated immune responses that are modulated by specifically sensitized thymusderived lymphocytes (T-lymphocytes). In attempts to evaluate the effect of antibiotics on cell-mediated immune function in vivo, several investigators have studied DTH skin tests with a variety of antigens. A significant reduction of DTH occurred in groups of mice treated with doxycycline and tetracycline. The fact that suppressive effects were more pronounced on the manifestation phase than the induction phase of DTH suggests that these antibiotics may have untoward effects on sensitized lymphocytes and on macrophages. This hypothesis is supported by the finding that the drugs inhibit mitogen-induced lymphocyte transformation in vivo.

Antibody Production Several studies of humans have evaluated the effect of rifampin on the antibody response to vaccination. Seroconversion following vaccinia vaccination was suppressed in normal volunteers whose vaccination site was treated with a cream containing 15% rifampin (51).

Polymorphonuclear Leukocyte Function Chemotaxis is the process by which phagocytes are attracted to the vicinity of pathogenic microorganisms by such factors as bacterial products, tissue proteases, and complement. A number of pharmacologic agents including tetracyclines and amphotericin B inhibit chemotaxis in vitro. The inhibitory effect of tetracyclines may be related to their ability to chelate calcium (52).

Miscellaneous Effects Natural killer (NK) cells compose a lymphoid subpopulation that is cytotoxic for a variety of targeted tumor cells (53). Hauser et al. (54) demonstrated that murine NK cell cytotoxicity for the murine lymphoma YAC-1 is significantly augmented during acute toxoplasma infection. Amphotericin B inhibited toxoplasmainduced murine NK cell activity against YAC-1; this antibiotic may adversely affect NK cell activity in other hosts as well (55).

SPECIFIC ANTIBIOTICS AND THEIR ASSOCIATED COMPLICATIONS Penicillins Natural Penicillins Penicillin G (Procaine Penicillin G and Benzathine Penicillin G)

Hypersensitivity to penicillin G is common; symptoms include skin eruptions that range from maculopapular to exfoliative dermatitis, urticaria, laryngeal edema, fever, and eosinophilia. Other serum sickness–like reactions (including chills, fever, edema, arthralgia, and prostration) may be controlled with antihistamines

and, if necessary, systemic corticosteroid therapy. Anaphylaxis including shock and death may also occur. Whenever hypersensitivity occurs, penicillin G should be discontinued. Anaphylactic reactions require immediate emergency treatment with epinephrine. Steroids should be administered intravenously, oxygen should be given, and airway management, including intubation, should also be conducted when indicated. Hemolytic anemia is an uncommon complication, and pancytopenia, though rare, can occur. Nephropathy in the form of interstitial nephritis can complicate intravenous administration of large doses of penicillin G. Most patients recover from this complication when administration of the drug is stopped. An overdose of penicillin can cause neuromuscular hyperirritability or convulsive seizures. Penicillin neurotoxicity can be prevented by determining the proper dose for patients with impaired renal function. The Jarisch–Herxheimer reaction has been observed among patients treated for syphilis. The reaction is the possible result of exacerbation of an existing syphilitic lesion. It is apparent that it is mediated by the action of cytokines released into the circulation (56), causing malaise, chills, fever, sore throat, myalgia, headache, and tachycardia. This reaction usually occurs six to eight hours after penicillin G is administered; it subsides within 24 hours (57). Aminopenicillin

Ampicillin, a semisynthetic antibiotic, and amoxicillin, an analog of ampicillin, have been associated with hypersensitivity reactions such as urticaria and anaphylaxis. Rashes associated with ampicillin are not always urticarial (58). Macular rashes appear to be ampicillin specific and do not indicate true hypersensitivity to penicillin (59). Orally or, less commonly, parenterally administered ampicillin therapy can cause nausea and diarrhea (60).

Penicillins with Beta-Lactamase Inhibitor Amoxicillin–Clavulanate

Amoxicillin–clavulanate is an orally administered antibacterial combination consisting of the semisynthetic antibiotic amoxicillin and the beta-lactamase inhibitor clavulanate potassium. Clavulanic acid results from the fermentation of Streptomyces clavuligerus (61). Structurally related to the penicillins, this beta-lactam can inactivate a wide variety of beta-lactamases (62) by blocking the active sites of these enzymes. Side effects associated with the use of amoxicillin–clavulanate include gastrointestinal problems such as diarrhea, nausea, vomiting, indigestion, gastritis, stomatitis, glossitis, black ‘‘hairy’’ tongue, mucocutaneous candidiasis, enterocolitis, and PMC. Ticarcillin–Clavulanate

Ticarcillin–clavulanate is a combination of the semisynthetic ticarcillin disodium and clavulanate potassium.

Chapter 3:

The administration of clavulanic acid has been associated with positive results on the direct Coomb’s test, but hemolysis has not been observed (63). Ampicillin–Sulbactam

Ampicillin–sulbactam consists of the semisynthetic antibiotic ampicillin sodium and the beta-lactamase inhibitor sulbactam sodium. Intramuscular administration of sulbactam causes pain at the site of injection, but no unexpected side effects have been seen after intravenous administration of ampicillin alone (64). Piperacillin–Tazobactam

Piperacillin–tazobactam is a combination consisting of the semisynthetic antibiotic piperacillin sodium and the beta-lactamase inhibitor tazobactam sodium. Diarrhea was the only side effect reported more often after treatment with piperacillin–tazobactam than with piperacillin alone (65).

Antipseudomonal Penicillins Antipseudomonal penicillins such as carbenicillin and ticarcillin can adversely affect platelets. Although almost all penicillins can cause platelet dysfunction (66), the effect is most severe with carbenicillin and ticarcillin. The penicillins disturb platelet membrane function by interfering with adenosine diphosphate receptors and leaving them unavailable to agonists that induce aggregation (67). The newer antipseudomonal penicillins (3) such as mezlocillin and piperacillin can disturb platelet function, but not as severely as carbenicillin and ticarcillin at an equivalent dosage (68).

Cephalosporins Some patients who are allergic to penicillin are also allergic to cephalosporins (69). Therefore, it is best to avoid the use of cephalosporins to treat patients with a history of anaphylaxis to penicillin (70). Despite the possibility of this adverse cross-reactivity, it appears that 93% to 97% of patients with a history of penicillin allergy do not react to cephalosporins (71,72).

First-Generation Cephalosporins (Cefazolin) Eosinophilia commonly occurs with cefazolin therapy (73). Nephrotoxicity is rare, mild, and reversible (74). Moreover, a transient rise in aspartate aminotransferase (AST) and alkaline phosphatase concentrations has been observed without clinical evidence of hepatic impairment (73).

Complications of Antibiotic Therapy

33

and may cause diarrhea (77). Reversible rises in AST concentrations may occur (78). Also, cefuroxime may exert an immunosuppressive effect that has no known clinical significance (79). The cephamycins, such as cefoxitin and cefotetan, are frequently included in discussions of second-generation cephalosporins. Frequently used in surgery, the cephamycins act against a spectrum of bacteria that include the facultative aerobic Enterobacteriaceae spp. and anaerobes. Cefotetan has been associated with hypoprothrombinemia and bleeding in patients with preexisting impaired coagulation, or in patients receiving anticoagulant therapy (80).

Third-Generation Cephalosporins Third-generation cephalosporins are extended-spectrum compounds that are stable to the presence of betalactamases produced by gram-negative bacteria and are highly potent against most Enterobacteriaceae spp. (81). These drugs can be separated into groups with poor antipseudomonal activity (e.g., ceftizoxime), and those with good antipseudomonal activity (e.g., cefoperazone). Their side effects are similar to those of other cephalosporins.

Monobactams (Aztreonam) The antibacterial spectrum of aztreonam somewhat resembles that of aminoglycosides and is not as wide as that of the third-generation cephalosporins (82). In patients with impaired hepatic or renal function, appropriate monitoring is recommended during therapy (83,84). Cross-reactivity with penicillins and cephalosporins seems to be rare.

Carbapenems (Imipenem–Cilastatin and Meropenem) Imipenem has high activity against aerobic and anaerobic bacteria. Adverse effects on the central nervous system (CNS), such as confusion, myoclonic activity, and seizures, have been observed during treatment with imipenem, especially when recommended dosages were exceeded. These experiences have occurred most commonly in patients with CNS disorders (e.g., brain lesions or history of seizures), compromised renal function, or both (85). Meropenem, which is similar to imipenem, but is relatively stable to human renal dehydropeptidase-1, may be less likely to induce convulsions (86). The most common side effects observed in one study were diarrhea and elevated liver enzymes (87).

Second-Generation Cephalosporins (Cefuroxime)

Aminoglycosides (Amikacin, Gentamicin, Tobramycin, Netilmicin, and Streptomycin)

Second-generation cephalosporins are active against many gram-negative organisms that are resistant to the first-generation cephalosporins (75). Cefuroxime is associated with relatively few side effects (3), although high doses may interfere with platelet function (76)

All aminoglycosides have the potential to induce auditory, vestibular, and renal toxicity, and neuromuscular blockade. These events are most common among patients with a history of renal impairment, those receiving other ototoxic or nephrotoxic drugs,

34

Fahim and Namias

and those treated for periods longer than recommended or given doses higher than recommended (85). Neurotoxicity, appearing as vestibular and permanent bilateral auditory ototoxicity, may be a side effect of treatment with aminoglycosides. Vertigo may be an indication of vestibular injury. Damage to the vestibular system occurs more frequently with gentamicin administration, and cochlear damage is more common with amikacin therapy (88,89). Prolonged neuromuscular blockade and respiratory paralysis have been reported with the use of aminoglycosides, especially in patients receiving anesthetics and neuromuscular blocking agents (85,90). Other effects of neurotoxicity may include numbness, skin tingling, muscle twitching, and convulsions. Patients who suffer cochlear damage during therapy may not experience symptoms to warn them of developing cranial nerve–eight toxicity. Total or partial irreversible bilateral deafness may occur after the drug has been discontinued. Aminoglycosideinduced ototoxicity is usually irreversible (85). The most common clinical manifestation of gentamicin nephrotoxicity is nonoliguric renal failure with proteinuria and increased concentrations of serum creatinine and blood urea. Renal function changes are usually reversible when administration of the drug is discontinued (3). Although less common, an acute oliguric renal failure and a subsequent diuretic phase may occur (91). Avoiding this and other forms of toxicity requires careful monitoring of serum concentrations of aminoglycosides, when feasible, to ensure that drug concentrations are adequate, but are not at potentially toxic levels (85).

Macrolides Erythromycin Erythromycin is produced by a strain of Streptomyces erythraeus and belongs to the macrolide group of antibiotics. The most frequent side effects of orally administered erythromycin preparations are gastrointestinal and dose related. They include nausea, vomiting, abdominal pain, diarrhea, and anorexia (85). Intravenous administration of the drug can also cause these side effects (92).

Clarithromycin The most frequently reported side effects experienced by adults taking clarithromycin (Biaxin1 tablets) were diarrhea, nausea, abnormal taste, dyspepsia, and abdominal pain or discomfort (85).

Azithromycin Overall, the most common side effects of azithromycin involved the gastrointestinal system. One study found that these side effects were more common in patients receiving a single-dose regimen of 1 g of azithromycin than among those receiving the multiple-dose regimen (85). Among patients with AIDS,

diarrhea is the principal side effect of azithromycin, but treatment cessation is not usually necessary (93).

Tetracyclines (Doxycycline, Oxytetracycline, and Minocycline) The use of drugs of the tetracycline class during tooth development (last half of pregnancy, infancy, and childhood to the age of eight years) may cause permanent discoloration of the teeth (yellow, gray, or brown) (85). Nausea, heartburn, epigastric pain, vomiting, and diarrhea are more commonly associated with tetracyclines than with most other orally administered antibiotics (3).

Fluoroquinolones Some patients treated with ciprofloxacin have experienced nausea, vomiting, diarrhea, abnormalities of the hepatic enzymes, and eosinophilia (3). Headache, restlessness, and rash were also observed among more than 1% of patients treated with the most common doses of ciprofloxacin (85). Moreover, ciprofloxacin may also cause CNS events (94), including dizziness, confusion, tremors, hallucinations, depression, and, rarely, suicidal thoughts or acts. Convulsions, increased intracranial pressure, and toxic psychosis have been observed among patients receiving quinolones, including ciprofloxacin (85). Like ciprofloxacin, ofloxacin is associated with multiple side effects: nausea, insomnia, headache, dizziness, diarrhea, vomiting, rash, and pruritus (95). Some quinolones have been associated with prolongation of the QT interval, as revealed by electrocardiography, and, infrequently, with cases of arrhythmia. Rare cases of torsades de pointes have reportedly occurred in patients taking levofloxacin (85). Temafloxacin was withdrawn from the market after five months of clinical use because of the ‘‘temafloxacin syndrome,’’ which is characterized by fever, chills, hemolysis, and jaundice, and is frequently associated with renal failure, hepatic dysfunction, and coagulopathy (96). Because trovafloxacin was found to be associated with death due to hepatic failure, it is now used only to treat critically ill patients.

Antifolate Agents The most serious type of adverse reaction to sulfonamides (trimethoprim and sulfamethoxazole) is the Steven–Johnson syndrome (97). This syndrome consists of erythema multiforme and ulceration of the mucous membranes of the mouth, eyes, and urethra, and it can sometimes be fatal. Acute agranulocytosis can occur, although it is more commonly associated with the use of older sulfonamides (3).

Miscellaneous Vancomycin Vancomycin, when administered rapidly (i.e., over several minutes) as a bolus, may be associated with exaggerated hypotension (98) and, rarely, cardiac

Chapter 3:

Complications of Antibiotic Therapy

35

arrest (99). During or soon after rapid infusion of vancomycin, patients sometimes experience anaphylactoid reactions such as hypotension, wheezing, dyspnea, urticaria, or pruritus (85). Rapid infusion may also cause flushing of the upper body (‘‘Red Man syndrome’’) (100) or pain and spasms in the muscles of the chest and back. Although such events usually resolve within 20 minutes, they are infrequent if vancomycin is infused slowly over a 60-minute period (85). Ototoxicity has been reported in association with vancomycin. One study found that most patients who experienced ototoxicity had kidney dysfunction or were receiving an ototoxic drug such as aminoglycoside (101). Nephrotoxicity has occurred among patients who were given vancomycin and aminoglycosides concomitantly, or who had preexisting kidney dysfunction (85).

vancomycin-resistant Enterococcus faecium bacteremia. Quinpristin–dalfopristin is bacteriostatic against E. faecium and bactericidal against strains of methicillinsusceptible and methicillin-resistant Staphylococcus spp. The most common adverse reactions that are thought to be related to quinpristin–dalfopristin use are myalgia and arthralgia. Quinpristin–dalfopristin significantly inhibits cytochrome P450 3A4 metabolism of cyclosporine A, midazolam, nifedipine, and terfenadine. The concomitant administration of quinpristin– dalfopristin and other drugs primarily metabolized by the cytochrome P450 3A4 enzyme system may result in increased plasma concentrations of these drugs, which could increase or prolong their therapeutic effect, increase the number or severity of adverse reactions, or both (85).

Clindamycin

Linezolid

Because clindamycin therapy has been associated with severe or even fatal colitis, the use of this agent should be reserved for serious infections for which less-toxic antimicrobial agents are inappropriate. Treatment with antibacterial agents alters the normal flora of the colon and may permit overgrowth of Clostridium spp. Studies have indicated that a toxin produced by Clostridium difficile is one primary cause of ‘‘antibiotic-associated colitis’’ and can be detected by tissue culture assay. Among some patients, diarrhea, colitis, and PMC occurred several weeks after the cessation of clindamycin therapy (85). The most serious side effect of clindamycin therapy is PMC. The clinical manifestations of PMC include diarrhea (sometimes watery), cramping, abdominal pain, and fever. Manifestations that occur less frequently include abdominal tenderness with rebound and leukocytosis. The risk of PMC is apparently greater among patients who receive the drug orally than among those who receive it parenterally (4). PMC is diagnosed by proctoscopy, which reveals raised yellow-whitish plaques on mucosa that is often erythematous or edematous and sometimes friable. The most important treatment for PMC is prompt withdrawal of clindamycin (or other inciting antibiotic). Fluid replacement is important. Among patients with prolonged diarrhea after discontinuation of antimicrobial therapy, several studies conducted (102,103) indicate that vancomycin or metronidazole given by mouth is effective in eliminating C. difficile from the colon and subsequently relieving diarrhea. For adults, vancomycin is given orally at a dosage of 500 mg every six hours for 7 to 14 days. There is no effective way to prevent PMC other than the judicious use of antimicrobial agents (4).

Linezolid is a member of the oxazolidinone class of antibiotics. It is effective against Enterococcus faecalis and E. faecium infections and also against MRSA infections. Because myelosuppression has been reported with the use of linezolid, complete blood count should be monitored weekly while patients are taking this drug (104).

Quinpristin–Dalfopristin One approved indication for the use of quinpristin– dalfopristin is the treatment of patients with serious or life-threatening infections associated with

Metronidazole Metronidazole has been shown to be carcinogenic in rodents (105). Convulsive seizures (106) and peripheral neuropathy (107) have been observed among patients treated with metronidazole. Metronidazole (intravenously administered Flagyl1) should be used with care in treating patients with evidence of or history of blood dyscrasia (85) because neutropenia, which is reversible, has been observed during its administration (108).

Chloramphenicol Hematologic toxicity is the most important adverse effect that complicates the administration of chloramphenicol (3). Adults may experience two types of toxicity involving the hematopoietic system. The first type, which is reversible and dose dependent (3,4), is maturation arrest of bone marrow. This toxicity manifests itself as anemia, leukopenia, thrombocytopenia, and reticulocytopenia. Chloramphenicol is metabolized in the liver, and the inactive metabolites are not toxic. No change in dosage is needed in patients with severe renal failure (4). The second type of hematologic toxicity experienced by adults is bone marrow aplasia, which is very rare and usually irreversible (4). This form of toxicity may appear during the first two weeks of therapy or after a latent period of weeks or months (109). Bone marrow transplantation is the only effective treatment (110). This toxicity is best prevented by limiting the use of chloramphenicol to serious infections that could not be more effectively treated with other drugs. Chloramphenicol

36

Fahim and Namias

should not be administered to patients who have or may have preexisting marrow damage. Gray baby syndrome is a type of circulatory collapse that can occur among premature and newborn infants who were given chloramphenicol (111).

Antifungal Agents Azoles Fluconazole, a synthetic triazole antifungal agent, can cause hepatic reactions that range from mild transient elevations in transaminases to clinical hepatitis, cholestasis and fulminant hepatic failure, and eventually death (85). Fatal hepatic reactions have occurred primarily among patients with serious underlying medical conditions, predominantly AIDS (112).

Intravenous administration of acyclovir has been associated with neurologic symptoms such as lethargy, obtundation, tremors, confusion, hallucinations, agitation, seizures, or transient hemiparesis (116). Therefore, acyclovir should be used with caution in treating patients who have underlying neurologic abnormalities, and those who have significant hypoxia or serious renal, hepatic, or electrolyte abnormalities (85).

CONCLUSION In conclusion, despite the medical advances that have been made due to the development of antibiotics, significant complications can occur from their use. Practitioners must remain abreast of the literature regarding the potential complications of the prescribed therapy.

Polyenes Amphotericin B is produced by Streptomyces nodosus and is one of the drugs of choice for the systemic treatment of invasive fungal infections. The most important side effect of amphotericin B is nephrotoxicity, which results from a reduction in renal blood flow and the glomerular filtration rate. This toxicity may be attenuated by the use of liposomal or lipid emulsion preparations. Renal acidosis occurs more frequently among patients who have received a total dose of 0.5 to 1.0 g or more, and is usually reversible after therapy has been discontinued (3). Although rare, anaphylaxis has been associated with both amphotericin– deoxycholate and liposomal amphotericin B (113).

Antiviral Drugs Ganciclovir Ganciclovir should not be administered if the absolute neutrophil count is less than 500, or the platelet count is less than 25,000 (3). Granulocytopenia (neutropenia), anemia, and thrombocytopenia have been observed in patients treated with ganciclovir (Cytovene1). The frequency and severity of these adverse events vary widely among different patient populations. Cell counts usually begin to recover within three to seven days of cessation of drug therapy (85). Colony-stimulating factors have been shown to increase neutrophil and white blood cell counts among patients receiving ganciclovir for the treatment of cytomegalovirus retinitis (114).

Acyclovir Precipitation of acyclovir crystals in renal tubules can occur if the drug is administered by bolus injection (115). The abnormal renal function that can result from acyclovir administration depends on the state of the patient’s hydration, other treatments, and the rate of drug administration. Concomitant use of other nephrotoxic drugs, preexisting renal disease, and dehydration make further renal impairment with acyclovir more likely (85).

REFERENCES 1. Gleckman RA, Borrego F. Adverse reactions to antibiotics. Clues for recognizing, understanding, and avoiding them. Postgrad Med 1997; 101:97–98, 101–104, 107–108. 2. Rieder MJ. Mechanisms of unpredictable adverse drug reactions. Drug Saf 1994; 11(3):196–212. 3. Kuckers A. Penicillin G. In: Kuckers A, Crowe SM, Grayson ML, Hoy JF, eds. The Use of Antibiotics. A Clinical Review of Antibacterial, Antifungal and Antiviral Drugs. 5th ed. Oxford: Butterworth-Heinemann, 1997:3–70. 4. Mayhall GG. Complications of antibiotic administration. In: Greenfield LG, ed. Complications in Surgery and Trauma. Philadelphia: J.B. Lippincott Company, 1984:193–214. 5. Austen KF. Systemic anaphylaxis in the human being. N Engl J Med 1974; 291:661. 6. Kelly JF, Patterson R. Anaphylaxis. Course, mechanisms and treatment. JAMA 1974; 227:1431. 7. Lin RY. A perspective on penicillin allergy. Arch Intern Med 1992; 152:930. 8. Weinstein L, Weinstein AJ. The pathophysiology and pathoanatomy of reactions to antimicrobial agents. Adv Intern Med 1974; 19:109–134. 9. Mandell GL. Cephalosporins. In: Mandell GL, Douglas RG, Bennett JE, eds. Anti-infective Therapy. New York: John Wiley & Sons, 1985:76–94. 10. Parker CW. Drug allergy. N Engl J Med 1975; 292: 511–514. 11. Vaudaux P, Waldvogel FA. Gentamicin inactivation in purulent exudates: role of cell lysis. J Infect Dis 1980; 142:586–593. 12. Finland M. Changing patterns of susceptibility of common bacterial pathogens to antimicrobial agents. Ann Intern Med 1972; 76:1009–1036. 13. Gross RJ, Ward LR, Threlfall EJ, King H, Rowe B. Drug resistance among infantile enteropathogenic Escherichia coli strains isolated in the United Kingdom. Br Med J 1982; 285:472–473. 14. Levy SB. Microbial resistance to antibiotics. An evolving and persistent problem. Lancet 1982; 2(8289):83–88. 15. Lorian V. The antibacterial range: an approach to the assessment of in vitro activity of antibiotics. In: Lorian V, ed. Significance of Medical Microbiology in

Chapter 3:

16. 17. 18. 19. 20. 21.

22. 23. 24.

25. 26. 27. 28.

29. 30.

31. 32. 33. 34. 35. 36. 37. 38.

the Care of Patients. 2nd ed. Baltimore: Williams and Wilkins, 1982:370–382. Fry D. Mechanisms of antibiotic resistance. In: Fry DE, ed. Surgical Infections. 1st ed. Boston: Little, Brown and Company, 1995:73–83. Russell AD. Plasmids and bacterial resistance to biocides. J Appl Microbiol 1997; 83:155–165. Jevons MP, Coe AW, Parker MT. Methicillin resistance staphylococci. Lancet 1963; 1:904–907. National Nosocomial Infections Surveillance Antimicrobial Resistance Report. Atlanta: Centers for Disease Control and Prevention, 1998. Medeiros AA. Beta-lactamases: quality and resistance. Smith TL, Pearson ML, Wilcox KR, et al. Emergence of vancomycin resistance in Staphylococcus aureus. Glycopeptide-Intermediate Staphylococcus aureus Working Group. N Engl J Med 1999; 340(7):493–501. Yowler CJ, Blinkhorn RJ, Fratianne RB. Vancomycindependent enterococcal strains: case report and review. J Trauma 2000; 48(4):783–785. Klastersky J. Antibiotic synergy and antagonism. In: Ristuccia AM, Cunha BA, eds. Antimicrobial Therapy. New York: Raven Press, 1984:37–53. Sabath LD. Theoretical basis of antibiotic: synergy and antagonism. In: Klastersky J, Staquet MJ, eds. Combination Antibiotic Therapy in the Compromised Host. Baltimore: Lippincott Williams & Wilkins, 1982:59–63. Jawetz E. The use of combinations of antimicrobial drugs. Ann Rev Pharmacol 1968; 8:151–170. Best WR. Drug-associated blood dyscrasias. Recent additions to the Registry. JAMA 1963; 185:286–290. Caulfield JB, Burke JE. Inhibition of wound healing by chloramphenicol. Arch Pathol 1971; 92:119–125. Di Berardino L, Giangostino P, Silvestri LG. Antibodies against rifampin patients with tuberculosis after discontinuation of daily treatment. Am Rev Respir Dis 1976; 114:1189–1190. Koeffler HP, Golde DW. Amphotericin inhibition of hematopoiesis in vitro. Am J Hematol 1977; 3:57–62. Spath P, Garratty G, Petz L. Studies on the immune response to penicillin and cephalothin in humans. II. Immunochematologic reactions to cephalothin administration. J Immunol 1971; 107:860–869. Yunis AA. Chloramphenicol-induced bone marrow suppression. Semin Hematol 1973; 10:225–234. Ghilchik MW, Morris AS, Reeves DS. Immunosuppressive powers of the antimicrobial agent trimethoprim. Nature 1970; 227:393–394. Hartmann G, Honikel KO, Knusel J. The specific inhibition of the DNA-directed RNA synthesis by rifamycin. Biochim Biophys Acta 1967; 145:843–844. Buss WC, Morgan R, Guttman J, Barela T, Stalter K. Rifampicin inhibition of protein synthesis in mammalian cells. Science 1978; 200:432–434. Easmon CSF. The effect of antibiotics on the intracellular survival of Staphylococcus aureus in vitro. Br J Exp Pathol 1979; 60:24–28. Johnson JD, Hand WL, Francis JB, King-Thompson N, Corwin RW. Antibiotic uptake by alveolar macrophages. J Lab Clin Med 1980; 95:429–439. Klempner MS, Styrt B. Clindamycin uptake by human neutrophils. J Infect Dis 1981; 144:472–479. Lobo MC, Mandell GL. The effect of antibiotics on Escherichia coli ingested by macrophages. Proc Soc Exp Biol Med 1973; 142:1048–1050.

Complications of Antibiotic Therapy

37

39. Mandell GL. Interaction of intraleukocytic bacteria and antibiotics. J Clin Invest 1973; 52:1673–1679. 40. Mohindra JK, Rauth AW. Increased cell killing by metronidazole and nitrofurazone of phyoxic compared to aerobic mammalian cells. Cancer Res 1975; 36:930–936. 41. Pious DA, Hawley P. Effect of antibiotics on respiration in human cells. Pediatr Res 1972; 6:687–692. 42. Alexander JW. Antibiotic agents and the immune mechanisms of defense. Bull NY Acad Med 1975; 51: 1939–1945. 43. Finch R. Immunomodulating effects of antimicrobial agents. J Antimicrob Chemother 1980; 6:691–699. 44. Gillissen G. Antibiotika und immunantwortbegleiteffekte der chemotherapie. Immun Infekt 1980; 8:79–88. 45. Hauser WE Jr, Remington JS. Effect of antibiotics on the immune response. Am J Med 1982; 72:711–716. 46. Mandell LA. Effects of antimicrobial and antineoplastic drugs on the phagocytic and microbicidal function of the polymorphonuclear leukocyte. Rev Infect Dis 1982; 4:683–697. 47. Raeburn JA. Antibiotics and immunodeficiency. Lancet 1972; 2:954–955. 48. Stites DP. Clinical laboratory methods of detecting cellular immunity. In: Stites DP, Terr AI, Parslow TG, eds. Basic and Clinical Immunology. 8th ed. Norwalk, Connecticut: Appleton & Lange, 1994:195–215. 49. Lafferty WE, Brewer LA, Corey L. Alteration of lymphocyte transformation response to herpes simplex virus infection by acyclovir therapy. Antimicrob Ag Chemother 1984; 26:887. 50. Terr AI. Inflammation. In: Stites DP, Terr AI, Parslow TG, eds. Basic and Clinical Immunology. 8th ed. Norwalk, Connecticut: Appleton & Lange, 1994:137–150. 51. Moshkowitz A, Goldblum N, Heller E. Studies on the antiviral effect of rifampin in volunteers. Nature 1971; 229:422–424. 52. Desen P, Clark RA, Nauseef WM. Granulocytic phagocytes. In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas and Bennett’s Principles and Practice of Infectious Diseases. 4th ed. New York: Churchill Livingstone, 1995:78–101. 53. Herberman RB, Holden HT. Natural cell-mediated immunity. Adv Cancer Res 1978; 27:305–377. 54. Hauser WE Jr, Sharma SD, Remington JS. Natural killer cells induced by acute and chronic toxoplasma infection. Cell Immunol 1982; 69:330–346. 55. Hauser WE Jr, Remington JS. Effect of amphotericin B on natural killer cell activity in vitro. J Antimicrob Chemother 1983; 11:257–262. 56. Griffin GE. New insights into the pathophysiology of the Jarisch-Herxheimer reaction. J Antimicrob Chemother 1992; 29:613. 57. Gefland JA, Elin RJ, Berry FW Jr, Frank MM. Endotoxemia associated with the Jarisch-Herxheimer reaction. N Engl J Med 1976; 295:211. 58. Stevenson J, Mandal BK. Ampicillin and the fifth day rash. Br Med J 1966; 1:1359. 59. Prospective study of ampicillin rash. Report of a Collaborative Study Group. Br Med J 1973; 1(5844):7–9. 60. Bass JW, Crowley DM, Steele RW, et al. Adverse effects of orally administered ampicillin. J Pediatr 1973; 83: 106. 61. Reading C, Cole M. Calvulanic acid: a beta-lactamase inhibiting beta-lactam from Streptomyces calvuligerus. Antimicrob Ag Chemother 1977; 11:852.

38

Fahim and Namias

62. Reading C, Farmer T, Cole M. The beta-lactamase stability of amoxycillin with b-lactamase inhibitor, calvulanic acid. J Antimicrob Chemother 1983; 11:27. 63. Williams ME, Thomas D, Harman CP, et al. Positive direct antiglobulin tests due to calvulanic acid. Antimicrob Ag Chemother 1985; 27:125. 64. Foulds G. Pharmacokinetics of sulbactam/ampicillin in humans: a review. Rev In Dis 1986; 8(suppl 5):503. 65. Kuye O, Teal J, De Vries VG, et al. Safety profile of piperacillin/tazobactam in phase I and III clinical studies. J Antimicrob Chemother 1993; 31(suppl A):113. 66. Fass RJ, Copelan EA, Brandt JT, et al. Plateletmediated bleeding caused by broad-spectrum penicillins. J Infect Dis 1987; 155:1242. 67. Ferres H, Nunn B. Penicillin metabolites and platelet function. Lancet 1983; ii:226. 68. Gentry LO, Wood BA, Natelson EA. Effects of apalcillin on platelet function in normal volunteers. Antimicrob Ag Chemother 1985; 27:683. 69. Levine BB. Antigenecity and cross-reactivity of penicillins and cephalosporins. J Infect Dis 1973; 128(suppl): 364. 70. Petz LD. Immunologic cross-reactivity between penicillins and cephalosporins: a review. J Infect Dis 1978; 137(suppl):74. 71. Dash CH. Penicillin allergy and the cephalosporins. J Anitmicrob Chemother 1975; 1(suppl):107. 72. Boguniewicz M, Leung DYM. Hypersensitivity reactions to antibiotics used commonly in children. Pediatr Inf Dis J 1995; 14:221. 73. Ries K, Levison ME, Kaye D. Clinical and in vitro evaluation of cefazolin, a new cephalosporins antibiotic. Antmicrob Ag Chemother 1973; 3:168. 74. Moellering RC Jr, Swartz MN. Drug therapy: the newer cephalosporins. N Engl J Med 1976; 294:24. 75. Turck M. Cephalosporins and related antibiotics: an overview. Rev Inf Dis 1982; 4(suppl):281. 76. Cazzola M, Matera MG, Santangelo G, et al. Effects of some cephalosporins and teicoplanin on platelet aggregation. Int J Clin Pharmacol Res 1993; 13(2):69–73. 77. Trollfors B, Alestig K, Norrby R. Local and gastrointestinal reactions to intravenously administered cefoxitin and cefuroxime Scan. J Infect Dis 1979; 11:315. 78. Norrby R, Foord RD, Hedlund P. Clinical and pharmacokinetic studies on cefuroxime. J Antimicrob Chemother 1977; 3:355. 79. Manzella JP, Clark JK. Effects of Moxalactam and cefuroxime on mitogen-stimulated human mononuclear leucocytes. Antimicrob Ag Chemother 1983; 23: 360. 80. Kuckers A. Cefotetan. In: Kuckers A, Crowe SM, Grayson ML, Hoy JF, eds. The Use of Antibiotics. A Clinical Review of Antibacterial, Antifungal and Antiviral Drugs. 5th ed. Oxford: Butterworth-Heinemann, 1997:395–397. 81. Dunn GL. Ceftizoxime and other third generation cephalosporins: structure-activity relationships. J Antimicrob Chemother 1982; 10(suppl C):1. 82. Skyes RB, Bonner DP. Discovery and development of monobactams. Rev Infect Dis 1985; 7(suppl 4):279. 83. Tunkel AR, Scheld WM. Aztreonam. Infect Control Hosp Epidemiol 1990; 11:486. 84. El-Touny M, El Guinaidy M, Barry MA, et al. Pharmacokinetics of aztreonam in patients with liver cirrhosis and ascites. J Antimicrob Chemother 1992; 30:387.

85. The PDR1 Electronic Library. Montvale, NJ: Medical Economics Company, Inc., 2001. 86. Patel JB, Giles RE. Meropenem: evidence of lack of proconvulsive tendency in mice. J Antimicrob Chemother 1989; 24(suppl A):307. 87. Bedikian A, Okamoto MP, Nkahiro RK, et al. Pharmakokinetics of meropenem in patients with intraabdominal infections. Antimicrob Ag Chemother 1994; 38:151. 88. Appel GB, Neu HC. Gentamicin in 1978. Ann Intem Med 1978; 89:528. 89. Meyer RD. Amikacin. Ann Intern Med 1981; 95:328. 90. Hall DR, McGibbon DH, Evans CC, Meadows GA. Gentamicin, tubocurarine, lignocaine and neuromuscular blockade. Br J Anaesth 1972; 44:1329. 91. Khan T, Stein RM. Gentamycin and renal failure. Lancet 1972; I:498. 92. Itoh Z, Suzuki T, Nakaya M, et al. Gastrointestinal motor-stimulating activity of macrolide antibiotics and analysis of their side-effects on the canine gut. Antimicrob Ag Chemother 1984; 26:863. 93. Young LS, Wiviott L, Wu M, et al. Azithromycin for treatment of Mycobacterium avium-intracellulare complex infection in patients wit AIDS. Lancet 1991; 338:1107. 94. Christ W. Central nervous system toxicity of quinolones: human and animal findings. J Antimicrob Chemother 1990; 26(suppl B):219. 95. Tack KJ, Smith JA. The safety profile of ofloxacin. Am J Med 1989; 87:78S. 96. Blum MD, Graham DJ, McCloskey CA. Temafloxacin syndrome: review of 95 cases. Clin Infect Dis 1994; 18:946. 97. Claxton RC. A review of 31 cases of Steven-Johnson syndrome. Med J Aust 1963; 1:963. 98. Newfield P, Roizen MF. Hazards of rapid administration of vancomycin. Ann Intern Med 1979; 91:581. 99. Glicklich D, Figura I. Vancomycin and cardiac arrest. Ann Intern Med 1984; 101:808. 100. Polk RE, Healy DP, Schwartz LB, et al. Vancomycin and the red man syndrome: pharmacodynamics and histamine release. J Infect Dis 1988; 157:502. 101. Brummett RE, Fox KE, Jacobs F, et al. Augmented gentamicin ototoxicity induced by vancomycin in guinea pigs. Arch Otolaryngol Head Neck Surg 1990; 116:61. 102. Batts DH, Martin D, Holmes R. Treatment of antibioticassociated Clostridium difficile diarrhea with oral vancomycin. J Pediatr 1980; 97:151. 103. Keighley MRB, Burdon DW, Arabi Y. Randomized controlled trial of vancomycin for pseudomembranous colitis and postoperative diarrhea. Br Med J 1978; 2:1667. 104. Zyvox Prescribing Information, Pharmacia Upjohn, 2002. 105. Rustia M, Shubik P. Induction of lung tumours and malignant lymphomas in mice by mitronidazole. J Natl Cancer Inst 1972; 48:721. 106. Frytak S, Moertel CG, Childs DS, Albers JW. Neurologic toxicity associated with high dose metronidazole therapy. Am Int Med 1978; 88:361. 107. Bradley WG, Karlsson IJ, Rassol CG. Metronidazole neuropathy. Br Med J 1977; 2:610. 108. Mckendrick MW, Geddes AM. Neutropenia associated with metronidazole. Br Med J 1979; 2:795. 109. Polak BCP, Wesseliny H, Dicu S. Blood dyscriasis attributed to Cholramphenicol. Acta Med Scan 1972; 192:409–418.

Chapter 3:

110. Storb R, Thomas ED, Buckner CD. Marrow transplantation in thirty ‘‘untransfused’’ patients with severe aplastic anemia. Am Int Med 1980; 92:30. 111. Sutherland JM. Fatal cardiovascular collapse of infants receiving large amounts of chloramphenicol. Am. J Dis Child 1959; 97:761. 112. Jacobson MA, Hanks DK, Ferrell LD. Fatal acute hepatic necrosis due to fluconazole. Am J Med 1994; 96:188. 113. Bates CM, Carey PB, Hind CR. Anaphylaxis due to liposomal amphotericin (AmBiosome). Genitourin Med 1995; 71:414.

Complications of Antibiotic Therapy

39

114. Hardy D, Spector S, Polsky B, et al. Combination of gancylovir and granulocyte-macrophage colony stimulating factor in the treatment of cytomegalovirus retinitis in AIDS patients. The ACTG 073 team. Eur J Clin Microbiol Infect Dis 1994; 13:S34. 115. Tucker WE Jr. Preclinical toxicology of acyclovir: an overview. Am J Med 1982; 73:27. 116. Johnson GL, Limon L, Trikha G, Wall H. Acute renal failure and neurotoxicity following oral acyclovir. Ann Pharmacother 1994; 28:460.

4 Complications of Blood and Blood-Product Transfusion Igor Jeroukhimov Department of Surgery, Assaf Harophe Medical Center, Tel Aviv University, Zerifin, Israel Mauricio Lynn Division of Trauma and Surgical Critical Care, DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Transfusion therapy continues to be widely discussed throughout all of medicine but particularly in surgery, where great regional and local variation in blood and blood product use can be documented (1). Each year in the United States, approximately 11 million units of blood are transfused into about four million patients (2). The transfusion of blood products can cause numerous serious complications, even death, and therefore, blood products should be considered potentially dangerous substances. The main perceived risks of blood and blood products are primarily related to transmission of infectious diseases. Although these risks are actually decreasing steadily as newer screening tests are introduced and better antiviral processing and storage capabilities evolve, the risks of blood-related transmission of infectious disease continues to be a problem (3). In recent years, transfusion therapy has also been linked to serious adverse consequences of immune suppression, which involve postoperative infectious complications or tumor recurrence. Survival is decreased in patients with breast or colon cancer who received blood (4). Finally, it must be recognized that a blood-banking ‘‘system’’ is in large part dependent on human input for decisions and processing and thus subject to human error. In the case of fatal transfusion reactions— basically, transfusion of incompatible blood—‘‘clerical error’’ continues to be the primary cause. Awareness of all these risks has forced the medical community to consider alternatives to transfusion, which frequently include a ‘‘transfusion avoidance strategy’’ (5).

In June 1667, Jean Baptiste Denis of Paris performed the first blood transfusion from animal to man. However, after the death of one of his patients, the French Parliament banned the practice of blood transfusion by issuing an edict on April 17, 1668 (7). This injunction halted the further development of blood transfusion for the next 150 years. In 1818, a group of London obstetricians headed by James Blundell initiated direct man-to-man transfusion for acute hemorrhage. However, Blundell and later the German physician Landois discovered the undesirability of heterologous transfusion, primarily because of the high incidence of hemolysis and renal shutdown. Lindois’ work from 1875 marked the end of any further heterogeneous transfusion. In 1900, Karl Landsteiner from the University of Vienna discovered three blood groups (8), and two years later Von De Castello and Sturli identified the fourth and rarest one (9). Clotting of blood during transfusion plagued all the early experiments and probably prevented many unfortunate outcomes of the procedure. In 1914, M. Hustin of Belgium introduced the use of citrate in collecting blood (10), and in 1916, Rous added glucose to lengthen the life span of the erythrocytes, thus permitting ‘‘delayed’’ transfusion (10). Many other improvements in blood storage and preservations were implemented before blood transfusion appeared to be relatively safe.

PHYSIOLOGY OF TRANSFUSION THERAPY HISTORICAL BACKGROUND Since ancient times, blood has been regarded as synonymous with life. The ancients considered blood ‘‘the seat of the soul’’ and believed that it carried the physical and mental qualities of the person through whom it flowed (6).

The indications for transfusion can be divided into two broad categories. The most common reason for transfusion is to enhance the oxygen-carrying capability of blood by expanding the red blood cell (RBC) mass. The second most common reason is to replace clotting factors that are lost, consumed, or not produced (11).

42

Jeroukhimov and Lynn

Enhancement of Oxygen-Carrying Capacity Most oxygen atoms in arterial blood bind reversibly, which serve as a carrier for oxygen. If cardiac output is adequate, increasing the RBC mass will increase the oxygen-carrying capacity of the blood. Nevertheless, the release of oxygen to the tissue is dependent on many factors. The most important is probably oxygen saturation of hemoglobin. As the degree of saturation increases, the binding affinity decreases and the release of oxygen to the tissues is enhanced. Adequate tissue oxygenation depends not only on adequate delivery of oxygen, but also on the oxygen demands of the tissue. Under normal circumstances, the rate of oxygen delivery (1000 mL/min) is substantially greater than the rate of oxygen consumption (250 mL/min). Despite this large difference, clinical circumstances occur in which oxygen consumption exceeds delivery; one example of such a circumstance is massive multisystem injury and sepsis. Traditionally, a hemoglobin concentration of 10 g/dL has been considered the lowest value acceptable before elective surgery. However, the limits of acceptability are currently being challenged. It is well known that hemoglobin concentration of 6 to 7 g/dL is well tolerated in patients with chronic renal failure. A hemoglobin concentration of 7 to 8 g/dL has been demonstrated to be adequate, except for patients with coronary artery disease or chronic obstructive pulmonary disease. Although there has been no widely accepted agreement on the lowest acceptable hemoglobin concentration before elective surgery, it is clear that the rate and magnitude of blood loss, the state of tissue perfusion, and the presence of cardiopulmonary disease affect patients’ abilities to tolerate low concentrations of hemoglobin. Decreased concentrations of 2,3-diphosphoglycerate increase the binding affinity between oxygen and hemoglobin. In fact, 2,3-DPG levels may decrease by 30% in blood stored for greater than two weeks, and by 60% to 70% in three weeks.

Enhancement of Hemostasis The second most common reason for transfusion of blood products is to replace hemostatic agents. Surgeons should understand the pathophysiology of hemostasis before using replacement products. These products should be used only in preparation for elective surgery or for treating patients with clinically significant abnormalities in hemostasis, such as disorders of consumption or production of fibrinogen, extrinsic or intrinsic coagulation factor defects, and platelet dysfunction.

ROLE OF COMPONENT THERAPY The primary indications for transfusion of packed RBCs, fresh frozen plasma (FFP), and platelets are

Table 1 Indications and Contraindications for Blood and Its Component Transfusion Blood products

Major indications

Whole blood Packed RBCs

Acute blood loss Symptomatic anemia

Washed RBCs

Symptomatic anemia, febrile reaction to packed RBCs Symptomatic anemia, RBC incompatibility Thrombocytopenia

Frozen RBCs Platelets concentrate

Fresh frozen plasma Cryoprecipitate

Deficiency of labile coagulation factors Hemophilia A, von Willebrand’s disease, Hypofibrinogenemia

Major contraindications Chronic anemia Anemia responsive to proper medication Anemia responsive to proper medication Anemia responsive to proper medication Presence of antiplatelet antibodies Need for high-potency coagulation factors Need for high-potency coagulation factors

Abbreviations: RBCs, red blood cells.

acute hemorrhage, coagulopathy, and thrombocytopenia, respectively. Contraindications for the use of these products are chronic anemia and anemia responsive to the other therapy. Indications and contraindications for other blood products are listed in Table 1.

PATHOPHYSIOLOGY OF BLOOD TRANSFUSION Complications associated with blood transfusion can be serious and sometimes fatal. These complications can be categorized as immunologic reactions, metabolic disturbances, and infectious complications (Fig. 1).

Immunologic Complications The immunologic consequences of receiving blood from donors are of great importance. Patients can become sensitized to antigens on transfused RBCs and other accompanying blood cells [platelets and white blood cells (WBC)]. Women can also become sensitized and form antibodies to these blood group antigens from exposure to blood cells carried by their babies during pregnancy. Because sensitization to red cell antigens can induce hemolytic disease of the newborn, transfusions of blood from blood banks should be avoided, whenever possible, by women who may later choose to bear children (12). Immunologic complications of blood transfusion are characterized as hemolytic (red cell type) and nonhemolytic (non–red cell type). The nonhemolytic reactions are the most frequent and lead to the development of fever and urticaria, hives, or asthma after the administration of blood. The incidence of these reactions is approximately 2% to 10% (13). They may be related to leukocytes or proteins, and the incidence of these reactions may be reduced by the use of packed RBC or special filters for the removal of leukocytes from blood (14).

Chapter 4: Complications of Blood and Blood-Product Transfusion

43

Figure 1 Complications of blood transfusion. Abbreviations: AIDS, acquired immunodeficiency syndrome; EBV, Epstein-Barr virus; CMV, Cytomegalovirus.

Febrile Reactions Elevation in basal core temperature occurs in as many as 7% of transfusion recipients, but is usually selflimiting (11). Having undergone transfusions previously contributes to the development of fever-causing antibodies. The incidence of antileukocyte antibody development is high and increases with repeated exposure to blood and blood products. Febrile reactions are diagnosed by exclusion. Therefore, when fever occurs during transfusion, the transfusion should be stopped and studies should be conducted to rule out hemolysis. If hemolysis is not a possibility, the patient may be given antipyretic agents (15). Patients may also develop hypotension, cyanosis, tachypnea, transient leucopoenia, or a syndrome of self-limiting fibrinolysis. A high level of leukoagglutinins in donor plasma may contribute to the reaction. The severity of the reaction depends on the magnitude of the antibody titer, the degree of increase in the number of leukocytes, and the rate at which the blood was transfused. The differential diagnosis includes red cell incompatibility, bacterial contamination, and unrelated disease process. Antihistamines are not effective in the treatment of febrile reaction; therefore, premedication is not indicated with future transfusions (16).

Allergic Reaction Two percent of transfusion reactions are classified as allergic reactions that result from the transfusion of

antigen or immunoglobulin to which patient has preexisting antibodies (17). A blood transfusion recipient can experience an allergic reaction to either medications or food ingested by the donor. The reaction can vary in severity from urticaria to anaphylaxis. Furthermore, allergic reaction can develop as a result of the passive transfer of sensitizing antibodies. When the recipient subsequently encounters the allergen to which the donor has produced antibodies, the recipient can experience an allergic reaction to that allergen. Unlike delayed hemolytic or febrile reactions, an allergic reaction does not require prior exposure to blood. The initial symptoms associated with allergic reactions are usually urticaria and pruritus. In mild cases, antihistamines may be used to decrease the number and severity of symptoms. Reactions of moderate severity have been described as ‘‘anaphylactoid.’’ Less common is a full anaphylactic response characterized by hypotension, cutaneous flushing, and bronchospasm. In this situation, the transfusion should be stopped, and epinephrine may be required to support hemodynamics and relive bronchospasm (15). Reactions have been reported to be especially severe if anti-IgA antibodies are the cause of the crisis. IgA-associated reactions are also characterized by the initial symptoms of diarrhea and abdominal pain, in addition to hemodynamic instability and bronchospasm (18). With future transfusions, the use of washed RBCs may decrease the incidence of anaphylactic reactions. Furthermore, the customary type and

44

Jeroukhimov and Lynn

cross-match test should be supplemented with an assay for complement-activating activity.

Graft vs. Host Disease Graft versus host disease (GVHD) occurs after the infusion of immunocompetent cells into a recipient whose immune system is incapable of rejecting the foreign cells (e.g., a patient with a deficiency in cellmediated immunity). Consequently, the infused immunocompetent cells initiate rejection of the normal host tissues and thus cause GVHD. Acute GVHD also occurs in recipients of allogenic bone marrow transplants and in persons with primary immunodeficiencies who receive viable allogenic lymphocytes (19). Blood products that place patients at risk of GVHD are whole blood, packed RBCs, fresh plasma, granulocytes, and platelets. GVHD has not been seen after transfusion of frozen blood, FFP, cryoprecipitate, or washed red cells. Irradiated blood products also may be given safely (20). The initial signs and symptoms of GVHD involve many organ systems, including the skin, liver, gastrointestinal tract, and bone marrow. Fever occurs first; 24 to 48 hours later, a generalized erythroderma begins to appear on the face (frequently behind the ears), and then spreads to the trunk and extremities. Bullous formations can occur. Skin biopsy shows extensive lymphocytic infiltration. Other signs and symptoms that may occur are anorexia, nausea, vomiting, diarrhea, and hepatocellular dysfunction due to the increased activity of liver enzymes and pancytopenia. GVHD usually occurs 2 to 30 days after transfusion (21). Fifty percent of patient with reported GVHD received a cationic drug before transfusion. The high mortality rate associated with GVHD results from severe bone marrow hypoplasia or aplasia, or from failure in bone marrow regeneration. Secondary infections occur as a result of agranulocytosis. There have been no reports of the occurrence of GVHD among immunologically competent persons or among those lacking only humoral immunity (22).

Posttransfusion Purpura Posttransfusion purpura is a rare form of acute hemorrhagic thrombocytopenia that appears one week after transfusion and primarily affects multiparous women who lack the platelet-specific antigen (PLA) 1 [human platelet antigen (HPA) 1] (23). Anti-PLA1 antibodies are detected in the serum of most affected patients, but the precise mechanism of platelet destruction is unclear. Thrombocytopenia (platelet counts less than 50,000) results in bleeding from the skin and mucous membranes. Additional transfusions, even with platelets, accentuate the thrombocytopenia, but plasmapheresis and exchange transfusion have resulted in some benefit. The syndrome spontaneously abates in several days to months (16).

Transfusion-Related Acute Lung Injury Transfusion-related acute lung injury is an acute respiratory distress syndrome that develops within four hours after transfusion of blood and is characterized by dyspnea and hypoxia due to noncardiogenic pulmonary edema. Although the actual incidence is not well known and its occurrence is almost certainly underreported, this injury has been estimated to occur once in approximately 5000 transfusions (24). This injury is most likely the result of several possible mechanisms. In some cases, blood donor’s antibodies with specificity to human leucocyte antigen (HLA) or neutrophilic antigenics react with recipient’s neutrophils; thus, this reaction leads to increased permeability of the pulmonary microcirculation. Reactive lipid products that arise during storage of blood products have been recently been implicated in the pathophysiology of transfusion-related acute lung injury (25). Such substances are capable of neutrophil priming, with subsequent damage to pulmonary capillary endothelium in the recipient, particularly in the setting of sepsis. As in other cases of acute respiratory distress syndrome, therapy is indicated; at least 90% of the patients with transfusionrelated acute lung injury recover (26).

Hemolytic Reaction Hemolytic (RBC type) reactions occur as a result of the interaction between the antibodies in the plasma of the recipient and the antigens on the RBCs of the donor. These reactions can be acute (i.e., appear within minutes after beginning of the transfusion) or delayed (causing clinical symptoms 3 to 21 days after transfusion). Compared with other reactions, severe hemolytic reactions are associated with the highest morbidity and mortality rates.

Acute Hemolytic Reaction Forty-five percent of acute hemolytic reactions are the result of errors in which blood samples were incorrectly identified, or the wrong type of blood was administered to the patient. Half of fatal hemolytic reactions occur among patients with O-negative blood and previously unidentified antibodies; the other half occur among patients with non–O blood who continued to receive type O blood and thus developed ‘‘admixture’’ blood types, which caused difficulty in subsequent crossmatching (14). To avoid these errors, the American Association of Blood Banks recommends that two unique identifiers (e.g., name and hospital identification number) always be used when linking blood products and blood samples to the intended patient (27). Acute hemolytic reaction begins with the transfusion of as little as few milliliters of incompatible blood, and its severity is proportional to the volume of blood to which the recipient is exposed (28). This type of reaction is characterized by pain at the

Chapter 4: Complications of Blood and Blood-Product Transfusion

infusion site, fever, chills, back and substernal pain, mental status changes, dyspnea, hypotension, facial flashing, cyanosis, and a bleeding diathesis (11,28). During a surgical procedure, the only evidence of an acute hemolytic reaction may be hypertension and myoglobinuria. Most of these manifestations result from RBC–antibody complexes that activate complement and liberate anaphylatoxins, histamine, and serotonin (28). These complexes also activate Factor XII, with the release of bradykinin and activation of the extrinsic coagulation cascade, with resultant disseminated intravascular coagulation (DIC) (17). Severe hemolytic reactions are also associated with renal cortical ischemia and subsequent redistribution of blood flow from the cortex to the medulla. Volume restoration corrects hypovolemia, but reperfusion induces oxygen free-radical formation, which further interfere with renal function (29). By several mechanisms, hemoglobinuria decreases renal function. Acidosis and sluggish glomerular filtration (associated with shock) promote the precipitation of hemoglobin, with blockage of renal tubules. Filtered antigen–antibody complexes may activate compliment and induce a local inflammatory response. In addition, ferric ions associated with hemoglobin promote the oxygen free radical (Huber– Weiss reaction); such free-radical formation may account for the synergistic deterioration in renal function that occurs with hypotension and concomitant hemoglobinuria (30). When acute hemolytic reaction is suspected, the transfusion should be stopped immediately. The free hemoglobin and haptoglobin concentrations in the patient’s serum should be assessed, and a direct Coombs’ test should be performed. Serum haptoglobin concentration decreases markedly after hemolysis because the protein binds to free hemoglobin to augment its clearance through the reticuloendothelial system. The direct Coombs’ test, in which antiglobulin antibodies are mixed with the patient’s RBCs, detects RBC–antibody complexes. The blood product in question should be returned to the blood bank for repeated type- and cross-match tests (16). Patients who are experiencing an acute hemolytic reaction characterized by urticaria or other allergic phenomena should receive diphenhydramine (Benadryl1 50 mg) intramuscularly or intravenously immediately and every six hours as needed. In addition, Ringer’s lactate solution should be administered intravenously with 12.5 to 25 mg of mannitol to ensure copious output of urine (100 to 200 mL/hour). One to two ampules of sodium bicarbonate can be added to each liter of fluid to alkalinize the urine to a pH of at least 6.5. If shock occurs, hydrocortisone and additional fluids should be administered (14).

Delayed Hemolytic Reaction Delayed hemolytic reactions are usually mild and appear 3 to 21 days after transfusion. The incidence

45

of this type of reaction is 1 per 4000 transfusions. Delayed hemolytic reactions affect patients who have previously been exposed to blood products (e.g., patients who have been pregnant or have received transfusions) and have experienced a quick anamnestic response as a result of the previous sensitization. Several days after transfusion, patients experience hemolysis, which is characterized by jaundice, hemoglobinuria, and decreased hematocrit. However, as many as 35% of patients experience no symptoms (31). The diagnosis is confirmed by an elevation in the serum concentration of indirect bilirubin, hemoglobinuria, and a decrease in the serum concentration of haptoglobin. Repeated type and screen tests of the donor’s blood often detect previously unrecognized serum antibody to Kidd (JKa and JKb), Rh (D, E, and C), Duffy (Fya and Fyb), or Kell (K) antigens (32). Furthermore, a repeated cross-match test using the patient’s current serum often shows transfusion to be incompatible. Delayed hemolytic reactions are selflimiting, require no specific treatment, and do not affect the patient’s compatibility in relation to future transfusions (16).

Iatrogenic Complications Two other forms of red cell reactions are worth mentioning. Iatrogenic hemolysis is induced if blood is administered with 5% glucose and water, rather than with saline. These patients will display signs and symptoms of immediate massive hemolysis. Iatrogenic clotting will occur if blood is administered with a calcium-containing solution such as Ringer’s lactate solution. If these clots are pushed into the circulation, massive pulmonary emboli will result (11).

Immune Suppression The association between blood transfusion and immune suppression is receiving increased attention. Nonviable RBC and particulate matter found in blood products before transfusions may impede the reticuloendothelial system’s ability to clear bacteria; such a block thereby predisposes the recipient to sepsis (33).

Tumor Recurrence The fact that patients who received allogenic blood transfusions are at increased risk of postoperative infection and cancer recurrence suggests that such transfusions are associated with clinically significant immunomodulatory effects (34,35). Indeed, numerous reports have suggested an increase in tumor recurrence and a decrease in survival for transfusion recipients who have colorectal, breast, cervical, lung, prostate, or head and neck tumors. Because allogenic transfusions increase humoral immune responses and decrease cell-mediated responses, the mechanism of allogenic transfusion-induced immunomodulation may involve altered cytokine regulation that results in shift toward a type 2 (Th2) immune response (36–38).

46

Jeroukhimov and Lynn

Septic Complications

Transmission of Hepatitis B and C Viruses

Transfusions of RBCs are associated with an increase in septic complications among patients undergoing surgical procedures for carcinoma of the colon (38,39) and among patients who have suffered from multiple trauma (40). Among patients undergoing hip replacement or spine surgery, the postoperative infection rate with allogenic blood transfusion appears to be 7 to 10 times higher than that associated with autologous blood or absence of transfusion.

The primary perceived risks associated with the transfusion of blood and blood products are related to the transmission of infectious diseases. Although these risks are steadily decreasing as newer screening tests are introduced and better antiviral processing and storage capabilities evolve, such risks continue to be a problem (4,5). Recent estimates of the infectious risk per unit of blood are shown in (Table 2) (26).

In 1975, the implementation of third-generation screening tests for the surface antigens of hepatitis B virus (HBV) led to marked reduction in transfusiontransmitted HBV infection. Today, this infection accounts for approximately 10% of all cases of hepatitis after transfusion (46). Although an acute disease appeared in approximately 35% of infected persons, chronic infection develops in only 1% to 10% of patients. The estimated risk of transfusion-transmitted HBV infection is 7 to 32 per million units of blood transfused, or 1 case per 30,000 to 250,000 units transfused blood (3). The estimated risk of transfusion-transmitted hepatitis C virus (HCV) is now in 103,000 transfusions (3). However, if one considers the unlikely possibility of chronic, immunologically silent state of infection, the risk of HCV may be as great as 1 in 30,000. Transfusion-transmitted HCV infection is important because the infection becomes chronic in 85% of cases, leads to cirrhosis in 20%, and results in hepatocellular carcinoma in 1% to 5% of cases (47).

Transmission of HIV

Transmission of Other Viruses

The first description of transfusion-associated human immunodeficiency virus (HIV) infection occurred in 1983 (41,42). After the implementation of HIVantibody testing in 1985 (43), approximately five cases of transfusion-related HIV infection per year were recorded, whereas 714 cases were reported in 1984 (44). To decrease the risk of transfusion-transmitted HIV disease, blood banks began to test donors for p24 antigen in late 1995 (45). Nevertheless, the estimated risk of transfusion-associated HIV infection is now calculated as 1 per 200,000 to 2,000,000 units of blood transfused; thus, transfusion-associated HIV infection results in 0.5 to 5 deaths per million units of blood transfused (3).

The incidence of hepatitis G viremia in donors in the United States is 1% to 2%. Although the virus can be transmitted by transfusion, there is no evidence that it is particularly hepatotrophic or causes disease (48). Transmission of hepatitis A virus by blood transfusion has been estimated to occur in one case per one million units of blood transfused (49). Hepatitis A infection is not commonly associated with blood transfusion because the absence of chronic carrier state and the presence of the symptoms that would eliminate the infected person as a potential donor during the brief viremic phase of the illness. The risk of transfusion-related transmission of parvovirus B 19 is highly variable from year to year (50,51). Infection is usually not clinically significant except in certain populations such as pregnant women (in whom the virus can cause hydrops fetalis), patients with hemolytic anemia (in whom aplastic crisis may develop), and immunocompromised patients (in whom chronic aplastic anemia may develop). Infection will develop in 20% to 60% of recipients of blood infected by human T-lymphotropic virus-I (HTLV-I) or HTLV-II (3). The rate of transmission is affected by the length of time that blood has been stored and by the number of WBCs in the unit. Blood that has been stored for more than 14 days and noncellular blood products such as cryoprecipitate and FFP do not appear to be infectious (52). The estimated risk of transfusionrelated HTLV-I or HTLV-II infection is 0.5 to 4 cases per 1,000,000 units (53). Myelopathy can occur in persons infected with HTLV-I or HTLV-II (54).

Infectious Complications

Table 2 Risk of Blood Transfusion Estimated per million units

Frequency per actual unit

No. of deaths per million units

Viral Infection Viral Hepatitis A Hepatitis B

1 7–32

0 0–0.14

Hepatitis C

4–36

HIV

0.4–5

100

1/1,000,000 1/30,000– 1/250,000 1/30,000– 1/150,000 1/200,00– 1/2,000,000 1/250,00– 1/2,000,000 1/10,000

2 83

1/500,000 1/12,000

Risk factor

HTLV types I

and II

Parvovirus B 19 Bacterial contamination Red cells Platelets

0.5–4

0.5–17 0.5–5 0 0 0.1–0.25 21

Abbreviations: HIV, human immunodeficiency virus; HTLV, human T lymphotropic virus.

Bacterial Contamination The organism most commonly implicated in bacterial contamination of red cells is Yersinia enterocolitica (3).

Chapter 4: Complications of Blood and Blood-Product Transfusion

Bacterial contamination of blood units is directly related to the length of storage, but Yersinia sp. RBC sepsis has been observed after transfusion of RBCs that had been stored as few as 7 to 14 days. In the United States, fewer than one per one million RBC units is contaminated. Grossly contaminated units of RBCs can sometimes be identified by comparing the color of the blood bag with the color of the blood in the attached tubing; contaminated blood in the bag will appear darker as a result of hemolysis and decreased oxygen content (55). The risk of platelet-related sepsis is estimated to be 1 case per 12,000 units transfused, but it is greater with transfusion of pooled platelet concentrates from multiple donors than for transfusion of platelet units obtained by aphaeresis from the single donor (56). Because of the risk of bacterial overgrowth with time, the shelf life of platelets stored at 20 C to 24 C is five days. In descending order, the organisms most commonly implicated in death due to bacterial contamination are Staphylococcus aureus, Klebsiella pneumonia, Serratia marcescens, and Staphylococcus epidermidis (57). The initial clinical signs associated with platelet-related sepsis vary more than those associated with transfusion of bacterially contaminated RBCs do and can range from mild fever (which can be undistinguishable from that seen with nonhemolytic febrile transfusion reactions) to acute sepsis, hypotension, and death. The overall mortality rate associated with platelet-associated sepsis is 26% (56). To date, there is no widely accepted test method or device that can identify bacterially contaminated blood products. A promising approach is the use of psoralens and ultraviolet light to produce not only nonimmunogenic but also sterile blood products (58). In any patient in whom fever develops within six hours after platelet transfusion, the possibility of bacterial contamination should be examined, and empiric antibiotic therapy should be considered.

COMPLICATION OF MASSIVE TRANSFUSION Massive transfusion is arbitrarily defined as the replacement of a patient’s total blood volume in less than 24 hours, or as the acute administration of more than half the patient’s estimated blood volume per hour. The complications of massive transfusion are the same as those associated with any blood transfusion and also include acidosis, abnormal hemostasis, changes in oxygen affinity, citrate toxicity, hyperkalemia, and hypothermia.

Acidosis In traumatic shock, the severity of metabolic acidosis and the rapidity of its correction are correlated with the likelihood of survival. The predominant cause of acidosis is inadequate tissue perfusion, and the occurrence of this complication indicates a need for further

47

resuscitation. Stored red cells are acidotic, but the metabolism of citrate produces alkalosis. Therefore, a pH imbalance to RBC transfusion itself is rare.

Hemostasis Abnormality Hemostasis may already be impaired in transfusion recipients because of an underlying condition. During massive transfusion of red cells, the number of platelets decreases because few functioning platelets exist in blood that has been stored more than 48 hours. The concentrations of factors V and VIII are reduced after storage for a few days, and the remaining concentration can be diluted if large volumes of crystalloid or colloid are given. In addition, DIC can be initiated by the release of thromboplastin-like material from platelets, WBCs, and RBCs broken down during storage and by the partial activation of coagulation factors. The extent of hemostatic derangement varies widely and is not predictable in relation to the volume of red cells transfused. Therefore, the use of prophylactic replacement formulas (e.g., administration of platelets and FFP after the transfusion of every eight units of red cells) is not recommended. It is preferable to monitor hemostasis and use FFP when abnormalities of the coagulation system appear. The use of cryoprecipitate is indicated for patients with DIC, when fibrinogen concentration is below 0.8 g/L. Thrombocytopenia below 50  103/L contributes to microvascular bleeding from mucosal surfaces, wounds, and puncture sites. A standard adult dose of platelets is usually six to eight units, which equals approximately one unit of platelets per 10 kg body weight (59).

Changes in Oxygen Affinity Stored hemoglobin has a high affinity for oxygen. Therefore, massive transfusion of stored RBCs with high oxygen affinity adversely affects oxygen delivery to the tissues. It seems wise to use fairly fresh red cell transfusions (less than one week old), but evidence supporting this practice has not been reported. Using fresh (less than 24 hours old) blood is not indicated. The concentration of 2,3-DPG rises rapidly after transfusion, and normal oxygen affinity is usually restored in a few hours.

Citrate Toxicity Coagulation and normal cardiac function require that serum contain ionized calcium. Each unit of blood contains approximately 3 g of citrate, which binds the ionized calcium. Because citrate is commonly used as an anticoagulant for blood storage, rapid infusion of blood may induce hypocalcemia if excess citrate binds with serum calcium. Normothermic patients can metabolize citrate from stored blood when it is transfused at a rate of 150 mL/70 kg/min (i.e., one unit given every five minutes) (60).

48

Jeroukhimov and Lynn

However, the metabolism of citrate is decreased in patients with hypothermia, liver dysfunction, shock states, or hypocarbia (17). All of these conditions that may be present in the exsanguinating patient significantly increase the patient’s risk of hypocalcemia. Abnormal calcium concentrations are more closely related to the rate of blood transfusion than to the total volume transfused (61). Empiric calcium administration of 1 g of calcium gluconate repletion for every four to six units of blood transfused has been suggested for rapid transfusion. However, calciumrelated hemostasis must be judiciously maintained because of the risk of poor contractility due to hypocalcemia and arrhythmia due to hypercalcemia (62,63). Serum calcium concentration should be measured prospectively. The effect of treatment is gauged by ionized calcium concentration rather than by total serum calcium concentrations because more than half of serum calcium is protein bound. Therefore, a low total serum calcium concentration may reflect only hypoalbuminemia due to dilution and not a physiologic hypocalcemia.

Hyperkalemia Hyperkalemia is an often discussed but rarely documented complication of blood transfusion (64). Potassium escapes from RBCs during storage: the plasma potassium concentration can reach 70 mEq/L in a stored unit of packed RBCs. However, this is usually clinically insignificant because each unit of packed red cells contains only 10 to 20 mL of plasma (1 mEq of potassium), and red cells resorb much of this leaked cation upon transfusion. Even so, death due to hyperkalemia has been associated with massive transfusion; therefore, the serum potassium concentration should be closely monitored (65).

Hypothermia The infusion of room-temperature crystalloid solutions decreases the patient’s core temperature. Transfusions with blood received from the blood bank at 4 C to 5 C can decrease the core temperature from 0.5 C to 1 C per unit transfused (17). Hypothermia leads to a reduction in citrate and lactate metabolism (leading to hypocalcemia and metabolic acidosis), an increased affinity of hemoglobin for oxygen, an impairment of red cell deformability, platelet dysfunction, and an increased tendency to cardiac dysrhythmias. Hypothermia can be avoided or at least minimized by warming all fluids, particularly blood, before they are infused. If transfusion-related hypothermia occurs, rapid infusion of warmed blood will help alleviate its deleterious effects.

VOLUME OVERLOAD Overload of circulatory system occurs when blood, especially whole blood and plasma, is infused in too

large a quantity or at too fast a rate. Elderly and neonatal patients are most sensitive to rapid volume shifts and constitute the largest groups in which this condition commonly develops. Also, patients with impaired cardiac function require careful monitoring so as to prevent circulatory overload during transfusion. Initially, volume overload is indicated clinically by labored respiration, but pulmonary edema can rapidly occur if the necessary measurements are not promptly undertaken. In mild cases of volume overload, decreasing the rate of transfusion or stopping the transfusion will allow equilibration of circulatory system. In more severe instances, oxygen may be required to prevent hypoxemia, and aggressive diuresis should be provided.

DIFFERENTIAL DIAGNOSIS AND ROLE OF LABORATORY TESTS The main purpose in evaluating a possible transfusion reaction is to recognize the early manifestations of life-threatening complications such as anaphylactic or endotoxic shock or acute hemolysis. These complications may be accompanied by apparently insignificant signs and symptoms rather than by those more commonly attributed to the reaction. Therefore, even minimal signs and symptoms have to be evaluated promptly, so that the blood transfusion can be terminated if necessary and appropriate management can be started. The first step in evaluating any transfusion reaction is a clerical check of all identifying information on the unit of blood, laboratory slips, and the patient’s chart. Next, the most crucial laboratory tests, i.e., those that can quickly provide evidence that supports or rules out a lethal transfusion reaction, should be performed. Examination of a clotted or anticoagulated blood sample drawn at the time of the suspected reaction may provide the earliest evidence of a hemolytic transfusion reaction. Hemolysis of as little as 10 mL to 20 mL of blood may be visually apparent in the blood sample at the bedside, but the blood sample should be centrifuged and the plasma or serum compared to a suitable pretransfusion sample from the patient, so that the occurrence of hemolysis can be ascertained. Finding of hemolysis does not necessarily confirm immunologic destruction; instead hemolysis may have been produced by physical agents. A direct Coombs’ test, which can be readily performed at the time that blood sample is examined visually, can rapidly determine whether the hemolysis is due to an immunologic process. When sequential negative results from pretransfusion direct Coombs’ tests are followed by a positive result after transfusion, the presence of RBC incompatibility is strongly suggested. The possibility of false-positive result from a Coombs’ test should be considered if the patient has a history

Chapter 4: Complications of Blood and Blood-Product Transfusion

of long-term treatment with Aldomet1 (Merck & Co., Inc., New Jersey, U.S.A.) or other drugs. False-negative results from a Coombs’ test can occur when incompatible RBCs are completely destroyed; such destruction is not uncommon in samples from recipients of ABO-incompatible transfusion. If a clerical error is discovered or if the results of a direct Coombs’ test performed on the reaction sample obtained after the transfusion are positive, crossmatching tests of blood samples obtained before and after transfusion should be repeated. When clerical checks and laboratory tests fail to identify the cause of the suspected transfusion reaction, the possibility that the blood or blood product is contaminated by bacteria should be evaluated. This evaluation can be made quickly by performing a gram stain of a dried blood smear or, even better, of the plasma supernatant from the unit of blood or blood product. The presence of bacteria should be presumptive evidence of contamination; however, units containing Clostridium spp. may be rarely found. Examination of urine for hemoglobinuria may be useful in documenting the presence of significant hemolysis, but this is usually delayed because filtration of hemoglobin occurs only after the plasma haptoglobin is saturated (about 100 mg hemoglobin/dL of plasma) and the level of free hemoglobin exceeds approximately 150 mg/dL. Quantitative determination of hemoglobin and haptoglobin requires additional time and testing facilities and is usually not a useful initial study. Furthermore, both levels can be elevated or decreased as a result of the patient’s state of nutrition, the presence of inflammatory process, and the effect of recent blood transfusions. Serum bilirubin levels reach their maximum five to seven hours following acute hemolysis. Most often they provide evidence supporting hemolysis, although there are many other intrahepatic and extrahepatic causes for hyperbilirubinemia (66). The laboratory assessment including prothrombin time (PT), partial thromboplastin time (PTT), fibrinogen, bleeding time, and platelet count should provide useful information for guiding therapeutic intervention (67), but provide no information about ongoing hemolysis.

PREVENTION OF TRANSFUSION COMPLICATIONS Preoperative Autologous Donation Preoperative autologous donation was rarely used before the recognition that HIV could be transmitted via blood transfusion. Both autologous blood donation and transfusion are associated with risks (Table 3). In one study (68), 1 in 16,783 autologous donations was associated with an adverse reaction severe enough to require hospitalization; this risk is 12 times as high as the risk associated with voluntary donations by healthy persons. Myocardial ischemia events have also been reported in association with, but not necessarily

49

Table 3 Advantages and Disadvantages of Autologous Blood Donation Advantages Prevents transfusion-transmitted disease Avoids red cell alloimmunization

Supplements the blood supply Provides compatible blood for patients with alloantibodies Prevents some adverse transfusion reactions

Disadvantages Does not eliminate the risk of bacterial contamination and volume overload Does not eliminate the risk of administrative error, resulting in ABO incompatibility Costs more than allogenic blood donation Results in discarding of blood that is not transfused Causes perioperative anemia and increases the likelihood of transfusion

as a result of, autologous blood transfusion (69,70). The transfusion of autologous blood has many of the same complications as transfusion of allogenic units (Table 3). Published guidelines have been considered on the types of patients for whom autologous donation is most appropriate (71). Most commonly, the number of units of autologous blood obtained preoperatively is based on the number of units that would be crossmatched before surgery if allogenic blood were being used (72). However, even for such procedures as joint replacement or radical prostatectomy, as much as 50% of autologous blood goes unused (73,74). When autologous blood is collected for procedures that seldom require transfusion, such as hysterectomy, vaginal delivery, and transuretheral resection of prostate, up to 90% of the units collected before these procedures go unused (75). A recent British consensus conference on autologous transfusion stated that autologous blood donation should be considered only if the likelihood of transfusion exceeds 50% (76).

Acute Normovolemic Hemodilution Acute normovolemic hemodilution entails the removal of whole blood from a patient immediately before surgery and simultaneous replacement with acellular fluid such as crystalloid or colloid to maintain normovolemia. Blood is collected in standard blood bags containing anticoagulants, remains in operating room, and is reinfused after major loss of blood has ceased, or sooner if indicated. Recent guidelines state that acute normovolemic hemodilution should be considered when the potential surgical blood loss is likely to exceed 20% of the blood volume in patients who have a preoperative hemoglobin level of more than 10 g/dL and who do not have severe myocardial disease, such as moderate-to-severe left ventricular impairment, unstable angina, severe aortic stenosis, or critical left main coronary artery disease (77). Acute normovolemic hemodilution has several advantages over autologous blood donation. First, the units procured by hemodilution require no testing, so that the costs are substantially lower than those of autologous blood donation (78). Second because the units of blood are not removed from

50

Jeroukhimov and Lynn

the operating room, the possibility of administrative error that could lead to ABO-incompatible blood transfusion is theoretically eliminated, as is the risk of bacterial contamination. Third, blood obtained by hemodilution does not require an additional investment of time by the patient because it is done at the time of surgery, nor does it prolong the duration of surgery or anesthesia (79).

Intraoperative Recovery of Blood Intraoperative recovery of blood involves collection and reinfusion of autologous red cells lost by patient during the surgery. A cell-washing device can provide the equivalent of 10 units of banked blood per hour to a patient with massive bleeding. The length of survival of recovered red cells appears to be similar to that of transfused allogenic red cells (80). Relative contraindications include the potential to aspiration of malignant cells, the presence of infection, and presence of other contaminants as amniotic or ascitic fluid in the operative field. Because washing does not completely remove bacteria from the recovered blood, intraoperative recovery should not be used if operative field has gross bacterial contamination (71).

Inactivation of Microbes in Platelet Units The inactivation of viruses in a unit of platelets while retaining the viability and hemostatic properties of these blood cells has proved to be a formidable challenge. Viral inactivation in a unit of platelets by means of exposure to psoralen derivatives followed by exposure to ultraviolet A has been intensively investigated and can greatly reduce the levels of HIV and hepatitis viruses. This treatment appears to inactivate any contaminating bacteria (81) and reduce or eliminate immunomodulation due to lymphocytes.

Use of Plasma with Reduced Viral Infectivity Treatment of plasma with a solvent-detergent process provides a means to inactivate all the viruses with lipid envelopes, including HIV, HBV, and HCV (82). The contents of plasma appear to be unchanged except for the procoagulant activity, which is reduced by about 15%, and the levels of large multimers von Willebrand factor and some other factors, including protein S, are decreased by over 50%. The pooling of plasma from so many donors as part of the solventdetergent process has aroused concern about the possible transmission of nonenveloped viruses (e.g., hepatitis A and parvovirus B19) that are not inactivated by the process. Transmission of parvovirus B19 is a potential problem for some transfusion recipients such as patients with sickle-cell disease or thalassemia (71). Clinicians are also concerned that the pooled product may transmit viruses that are yet undiscovered. Furthermore, the high cost of the product has slowed widespread acceptance. Other

improvement to the product, such as screening for parvovirus infectivity by PCR testing and removal of ABO system antibodies, are in progress (83).

Viral Inactivation Because donor testing and screening will never be perfect, viral inactivation has been pursued and is becoming available. Leukodepletion can limit the transmission of cell-borne viruses such as cytomegalovirus. Data have been obtained in bone marrow transplant recipients, which shows that removing white cells from transfused blood components is equivalent to providing blood components from donors who are seronegative to cytomegalovirus (84). To accomplish this degree of viral safety, it is necessary to decrease the concentration of contaminating white cells below 5  106; this represents a 1000-fold reduction, which is best accomplished by new filters that can be applied in blood centers to provide prestorage leukocytedepleted blood products. Because the removal of white cells to this concentration also reduces the risk of alloimmunization to HLA factors and reduces the rate of other immunologic transfusion reactions in blood recipients, the implementation of universal leukodepletion is progressing in transfusion medicine practice (85).

Pharmacologic Agents In addition to the autologous option, other transfusion alternatives have been pursued. Physicians are now encroached to use drugs without biohazardous risks in favor of blood components. Patients with von Willebrand disease are given 1-desamino-8-D-arginine-vasopressin (DDAVP) rather than cryoprecipitate to improve hemostasis (86); aprotinin and other fibrinolytic inhibitors have been used in high-blood loss surgeries (87), and hematopoetic growth factors such as erythropoietin are advocated for dialysis patients or for patients undergoing elective surgery (88). Recombinant factor VIIa was successfully used for the treatment of acute bleeding episodes in hemophilia patients, as well in patients with liver cirrhosis, who underwent major surgery (89,90).

Use of Red Cell Substitute A number of blood substitutes have completed safety trials and are now undergoing efficacy evaluation. There are numerous potential advantages to the use of these solutions compared to RBCs. They are readily available and have a long shelf life, do not require type- and crossmatching, are free of viral or bacterial contamination, have a much lower viscosity than blood, and may lack the immunosuppressive effect of blood. These products appear to be safe and have potential for extensive use, particularly in the surgical patient. Optimization of oxygen delivery to individuals suffering from organ ischemia may ultimately represent the greatest potential use of these solutions,

Chapter 4: Complications of Blood and Blood-Product Transfusion

but a short circulation time may limit their utility in emergent resuscitation (91,92).

TREATMENT AXIOMS &

&

&

Major hemolytic transfusion reaction generally appears clinically before 50 to 100 mL of blood has been infused. Nonhemolytic febrile or allergic reactions may not occur until all units of blood have been administrated. Most allergic or febrile reactions occur in spite of a satisfactory type and crossmatch. If a major transfusion reaction is suspected, the transfusion should be stopped immediately, and the remainder of the stored blood and the sample of the patient’s blood should be sent to the blood bank for repeated crossmatch and analysis.

REFERENCES 1. Sirchia G, Giovanetti AM, eds. Safe and Good Use of Blood in Surgery (Sanguis): Use of Blood Products and Artificial Colloids in 43 European Hospitals. Vol. 153. Luxembourg: The European Commission, 1994. 2. Wallace EL, Churchill WH, Surgenor DM, Cho GS, McGurk S. Collection and transfusion of blood and blood components in the United States, 1994. Transfusion. 1998; 38(7):625–636. 3. Schreider GB, Busch MP, Kleinman SH, Korelitz JJ. The risk of transfusion-transmitted viral infection. The Retrovirus Epidemiology Donor Study. N. Engl J Med 1996; 334(26):1685–1690. 4. Klein HG. Allogeneic transfusion risks in the surgical patient. Am J Surg. 1995; 170(suppl 6A):21S–26S. 5. Greenburg AG. New transfusion strategies volume. Am J Surg 1997; 173:49–52. 6. Hutchin P. History of blood transfusion: a tercentennial look. Surgery 1967; 64(3):685–700. 7. Hoff HE, Guillemin R. The first experiments on transfusion in France. J Hist Med 1963; 18:103. 8. Mollison PL, Engelfriet CP. Marcelo contreras immunology of red cells. In: Mollison PL, Engelfriet CP, eds. Marcelo Contreras Blood Transfusion in Clinical Medicine. 9th ed. Oxford: Oxford Publication, 1993: 76–147. 9. Race RR, Sanger R. The ABO blood groups. In: Race RR, Sanger R, eds. Blood Groups in Men. 5th ed. Oxford, Edinburgh: Blackwell Scientific Publication, 1968:9–63. 10. Rossi EC, Simon TL, Moss GS, Gould SA. Transfusion into the next millennium. In: Rossi EC, Simon TL, Moss GS, Gould SA, eds. Principles of Transfusion Therapy. 6th ed. Baltimore: Williams & Wilkins, 1996:1–11. 11. Rotondo MF. Physiology of transfusion therapy. In: Savage EB, Fishman SJ, Miller LD, eds. Essentials of Basic Science in Surgery. Philadelphia: JB Lippincott, 1993:61–67. 12. Ness PM. Transfusion medicine: an overview and update. Clin Chem 2000; 46:1270–1276. 13. Barnes A. Transfusion of universal donor and uncrossmatched blood. Bibl Haematol 1980; (46):132–142. 14. Wilson RF, ed. Handbook of Trauma: Pitfalls and Pearls. Vol. 4. Philadelphia: JB Lippincott, Williams & Wilkins, 1999:32–33.

51

15. Barton JC. Nonhemolytic, noninfectious transfusion reactions. Semin Hematol 1981; 18(2):95–121. 16. Phillips G, Rotondo M, Schwab CW. Transfusion therapy. In: Maull KI, Rodrigez A, Wiles CE III, eds. Complications in Trauma and Critical Care. Philadelphia: WB, Saunders, 1996: 73–79. 17. Edelman B, Heyman MR. Blood component therapy for trauma patients. In: Stene JK, Grande CM, eds. Trauma Anesthesia. Baltimore: Williams and Wilkins, 1991: 133–176. 18. Miller WV, Holland PV, Sugarbaker E, Strober W, Waldman TA, Neil SY. Anaphylactic reactions to IgA: a difficult transfusion problem. Am J Clin Pathol 1970; 54(4):618–621. 19. Dennis RC, Clas D, Niehoff JM, et al. Transfusion therapy. In: Civetta JM, Taylor RW, Kirby R, eds. Critical Care. Philadelphia: Lippincott Williams & Wilkins, 1997:653–654. 20. Brubaker DB. Human posttransfusion graft-versus-host disease. Vox Sang 1983; 45(6):401–420. 21. Brubaker DB. Immunopathogenic mechanism of posttransfusion graft-vs-host disease. Proc Soc Exp Biol Med 1993; 202(2):122–147. 22. Hong M, Gatti RA, Good RA. Hazards and potential benefits of blood transfusion in immunologic deficiency. Lancet 1968; 2:388–389. 23. Nauck MS, Gierens H, Nauck MA, Marz W, Wieland H. Rapid genotyping of human platelet antigen 1 (HPA-1) with fluorophore-labelled hybridization probes on the LightCycler. Br J Haematol 1999; 105(3): 803–810. 24. Popovsky MA, Moore SB. Diagnostic and pathogenetic considerations in transfusion-related acute lung injury. Transfusion 1985; 25(6):573–577. 25. Silliman CC, Paterson AJ, Dickey WO, et al. The association of biologically active lipids with the development of transfusion-related acute lung injury: a retrospective study. Transfusion 1997; 37(7):719–726. 26. Goodnough LT, Brecher ME, Kanter MH, AuBuchon JP. Transfusion medicine. First of two parts – blood transfusion. N Eng J Med 1999; 340(6):438–447. 27. Standards for Blood Banks and Transfusion Services. 15th ed. Bethesda: American Association of Blood Banks, 1993:23. 28. Greenwalt TJ. Pathogenesis and management of hemolytic transfusion reactions. Semin Hematol 1981; 18(2): 84–94. 29. Paller MS, Hoidal JR, Ferris TF. Oxygen free radicals in ischemic acute renal failure in the rat. J Clin Invest 1984; 74:1156–1164. 30. Paller MS. Hemoglobin- and myoglobin-induced acute renal failure in rats: role of iron in nephrotoxicity. Am J Physiol 1988; 255:F539–F544. 31. Moore SB, Taswell HF, Pineda AA, Sonnenberg CL. Delayed hemolytic transfusion reaction: evidence of the need for an improved pre-transfusion compatibility test. Am J Clin Pathol 1980; 74:94–97. 32. Pineda AA, Taswell HF, Brzica SM Jr. Transfusion reaction: an immunologic hazard of blood transfusion. Transfusion 1978; 18(1):1–7. 33. Rutledge R, Sheldon GF, Collins ML. Massive transfusion. Criti Care Clinics 1986; 2(4):791–805. 34. Blumberg N. Allogeneic transfusion and infection: economic and clinical implications. Seminars in Hematology 1997; 34(3 suppl 2):34–40.

52

Jeroukhimov and Lynn

35. Wu HS, Little AG. Perioperative blood transfusion and cancer recurrence. J Clin Oncol 1988; 6(8):1348–1354. 36. van Aken WG. Does perioperative blood transfusion promote tumor growth? Transfus Med Rev 1989; 3(4): 243–252. 37. Tartter PI. The association of perioperative blood transfusion with colorectal cancer recurrence. Ann Surg 1992; 216(6):633–638. 38. Blumberg N, Heal JM. Transfusion and host defenses against cancer recurrence and infection. Transfusion 1989; 29(3):236–245. 39. Tartter PI. Blood transfusion and infectious complications following colorectal cancer surgery. Br J Surg 1988; 75(8):789–792. 40. Nichols RL, Smith JW, Klein DB, et al. Risk of infection after penetrating abdominal trauma. N Eng J Med 1984; 311(17):1065–1070. 41. Joint statement on acquired immune deficiency syndrome (AIDS) related to transfusion. Transfusion 1983; 23:87–88. 42. Provisional Public Health Service inter-agency recommendations for screening donated blood and plasma for antibody to the virus causing acquired immunodeficiency syndrome. Morb Mortal Wkly Rep 1985; 34:1–5. 43. Selik RM, Ward JW, Buehler JW. Trends in transfusionassociated acquired immune deficiency syndrome in the United States, 1982 through 1991. Transfusion 1993; 33(11):890–893. 44. Stramer SL, Aberle-Grasse J, Brodsky JP, Busch MP, Lackritz EM. US blood donor screening with p24 antigen (Ag): one year experience [abs.]. Transfusion 1997; 37 (suppl):1S. 45. Domen RE. Paid-versus-volunteer blood donation in the United States: a historical review. Transfus Med Rev 1995; 9(1):53–59. 46. Conry-Cantilena C, VanRaden M, Gibble J, et al. Routes of infection, viremia, and liver disease in blood donors found to have hepatitis C virus infection. N Eng J Med 1996; 334(26):1691–1696. 47. Tong MJ, el-Farra NS, Reikes AR, Co RL. Clinical outcomes after transfusion-associated hepatitis C. N Eng J Med 1995; 332(22):1463–1466. 48. Alter HJ, Nakatsuji Y, Melpolder J, et al. The incidence of transfusion-associated hepatitis G virus infection and its relation to liver disease. N Eng J Med 1997; 336(11):747–754. 49. Dodd RY. Adverse consequences of blood transfusion: quantitative risk estimates. In: Nance ST, ed. Blood Supply: Risk Perceptions, and Prospects for the Future. Bethesda, MD: American Association of Blood Banks, 1994:1–24. 50. Risk associated with human parvovirus B 19 infection. Morb Mortal Wkly Rep 1989; 38:81–88, 93–97. 51. Luban NL. Human parvoviruses: implication for transfusion medicine. Transfusion 1994; 34:821–827. 52. Guidelines for counseling persons infected with human T-lymphotropic virus type-I (HTLV-I) and type II (HTLV-II). Centers for Disease Control and Prevention and the U.S.P.H.S Working Group. Ann Intern Med 1993; 118:448–454. 53. Manns A, Wilks RJ, Murphy EL, et al. A prospective study of transmission by transfusion of HTLV-I and risk factors associated with seroconversion. Int J Cancer 1992; 51(6):886–891.

54. Gout O, Baulac M, Gessain A, et al. The Rapid development of myelopathy after HTLV-I infection acquired by transfusion during cardiac transplantation. N Eng J Med 1990; 322(6):383–388. 55. Kim DM, Brecher ME, Bland LA, Estes TJ, Carmen RA, Nelson EJ. Visual identification of bacterially contaminated red cells. Transfusion 1992; 32(3):221–225. 56. Chiu EK, Yuen KY, Lie AK, et al. A prospective study of symptomatic bacteremia following platelet transfusion and of its management. Transfusion 1994; 34(11): 950–954. 57. Goldman M, Blajchman MA. Blood product-associated bacterial sepsis. Transfus Med Rev 1991; 5(1):73–83. 58. Lin L, Cook DN, Wiesehahn GP, Alfonso R, et al. Photochemical inactivation of viruses and bacteria in platelets concentrates by use of a novel psoralen and long-wavelength ultraviolet light. Transfusion 1997; 37(4):423–435. 59. Hewitt P, Regan F. Blood transfusion. In: Webb AR, Shapiro MJ, Singer M, Sater PM, eds. Oxford Textbook of Critical Care. Oxford: Oxford University Press, 1999:691–694. 60. Rudowski WJ. Blood transfusion: yesterday, today and tomorrow. World J Surg 1987; 11(1):86–93. 61. Denlinger JK, Nahrwold ML, Gibbs PS, Lecky JH. Hypocalcemia during rapid blood transfusion in anaesthetized man. Br J Anesthes 1976; 48(10):995–1000. 62. Denlinger JK, Nahrwold ML. Cardiac failure associated with hypocalcemia. Anesthesia Analgesia 1976; 55(1): 34–36. 63. Olinger GN, Hottenrott C, Mulder DG, et al. Acute clinical hypocalcemic myocardial depression during rapid blood transfusion and postoperative hemodialysis: a preventable complication. J Thorac Cardiovasc Surg 1976; 72(4):503–511. 64. Wilson RF, Binkley LE, Sabo FM Jr, et al. Electrolyte and acid-base changes with massive blood transfusions. Am Surg 1992; 58(9):535–544. 65. Jameson LC, Popic PM, Harms BA. Hypercalcemic death during use of high capacity fluid warmer for massive transfusion. Anesthesiology 1990; 73(5): 1050–1052. 66. Sharp DE. Complication of blood and blood-product transfusion. In: Greenfeld LJ, ed. Complications in Surgery and Trauma. Philadelphia: JB Lippincot, 1984: 148–152. 67. Mannucci PM, Federici AB, Sirchia G. Hemostasis testing during massive blood replacement. A study of 172 cases. Vox Sanguinis 1982; 42(3):113–123. 68. Popovsky MA, Whitaker B, Arnold NL. Severe outcomes of allogenic and autologous blood donation: frequency and characterization. Transfusion 1995; 35(9):734–737. 69. Goodnough LT, Monk TG. Evolving concepts in autologous blood procurement and transfusion: case reports of perisurgical anemia complicated by myocardial infarction. Am J Med 1996; 101(2A):33S–37S. 70. Kasper SM, Ellering J, Stachwitz P, Lynch J, Grunenberg R, Buzello W. All adverse events in autologous blood donors with cardiac disease are not necessarily caused by blood donation. Transfusion 1998; 38(7):669–673. 71. Goodnough LT, Brecher ME, Kanter MH, AuBuchon JP. Transfusion medicine. Second of two parts – blood conservation. N Eng J Med 1999; 340(7):525–533.

Chapter 4: Complications of Blood and Blood-Product Transfusion

72. Axelrod FB, Pepkowitz SH, Goldfinger D. Establishment of a schedule of optimal preoperative collection of autologous blood. Transfusion 1989; 29(8):677–680. 73. Renner SW, Howanitz PJ, Bachner P. Preoperative autologous blood donation in 612 hospitals. A College of American Pathologists’ Q-Probes study of quality issues in transfusion practice. Arch Pathol Laboratory Med 1992; 116(6):613–619. 74. Goodnough LT, Saha P, Hirschler NV, Yomtovian R. Autologous blood donation in nonorthopaedic surgical procedures as a blood conservation strategy. Vox Sanguinis 1992; 63(2):96–101. 75. AuBuchon JP, Gettinger A, Littenberg B. Determinants of physician ordering of preoperative autologous donations. Vox Sanguinis 1994; 66(3):176–181. 76. Thomas MJ, Gillon J, Desmond MJ. Consensus conference on autologous transfusion. Preoperative autologous donation. Transfusion 1996; 36(7):633–639. 77. Napier JA, Bruce M, Chapman J, et al. Guidelines for autologous transfusion II. Perioperative hemodilution and cell salvage. British Committee for Standards in Hematology Blood Transfusion Task Force. Autologous Transfusion Working Party. Br J Anesth 1997; 78(6): 768–771. 78. Monk TG, Goodnough LT, Birkmeyer JD, Brecher ME, Catalona WJ. Acute normovolemic hemodilution is a cost-effective alternative to preoperative autologous blood donation by patients undergoing radical retropubic prostatectomy. Transfusion 1995; 35(7):559–565. 79. Goodnough LT, Despotis GJ, Merkel K, Monk TG. A randomized trial comparing acute normovolemic hemodilution and preoperative autologous blood donation in total hip arthroplasty. Transfusion 2000; 40(9):1054–1057. 80. Williamson KR, Taswell HF. Intraoperative blood salvage: a review. Transfusion 1991; 31(7):662–675. 81. Grass JA, Hei DJ, Metchette K, et al. Inactivation of leukocytes in platelets concentrates by photochemical treatment with psoralen plus UVA. Blood 1998; 91(6): 2180–2188. 82. Klein HG, Dodd RY, Dzik WH, et al. Current status of solvent/detergent-treated frozen plasma. Transfusion 1998; 38(1):102–107.

53

83. Judd WJ, Davenport RD, Downs T, Hammond DJ, Chin S, Pehta JC. Isohemagglutinin-depleted solvent detergent plasma: a universal viral-inactivated plasma for patients of any ABO type. Blood 1999; 94:375a. 84. Bowden RA, Slichter SJ, Sayers MH, Mori M, Cays M, Meyers JD. Use of leukocyte-depleted platelets and cytomegaloviruses-seronegative red blood cell store prevention of primary cytomegalovirus infection after marrow transplant. Blood 1991; 78(1): 246–250. 85. Blajchman MA. Transfusion-associated immunomodulation and universal white cell reduction: are we putting the cart before the horse? Transfusion 1999; 39(7):665–670. 86. Mannucci PM, Canciani MT, Rota L, Donovan B. Response of factor VIII/von Willebrand factor to DDAVP in healthy subjects and patients with haemophilia A and von Willebrand disease. Br J Haematol 1981;47(2):283–293.100. 87. Salzman EW, Weinstein MJ, Weintraub RM, et al. Treatment with desmopressin acetate to reduce blood loss after cardiac surgery. A double blind randomized trial. N Eng J Med 1986; 314(22):1402–1406. 88. Eschbach JW, Egrie JC, Downing MR, Browne JK, Adamson JW. Correction of the anemia of end-stage renal disease with recombinant human erythropoetin. Results of combined phase I and II clinical trial. N Eng J Med 1987; 316(2):73–78. 89. Negrier C, Lienhart A. Overall experience with NovoSeven. Blood Coagul Fibrinolysis 2000; 11(suppl 1): 19–24. 90. Papatheodoridis GV, Chung S, Keshav S, et al. Correction of both prothrombin time and primary hemostasis by recombinant factor VII during therapeutic alcohol injection of hepatocellular cancer in liver cirrhosis. J Hepatol 1999; (31):747–750. 91. Cohn S. Blood substitutes in surgery. Surgery 2000; 127(6):599–602. 92. Goodnough LT, Monk TG, Brecher ME. Autologous blood procurement in surgical setting: lessons learned in the last 10 years. Vox Sanguinis 1996; 71(3): 133–141.

5 Hypovolemic and Septic Shock Matthew O. Dolich Division of Trauma and Surgical Critical Care, University of California, Irvine School of Medicine, Irvine, California, U.S.A. Don H. Van Boerum Section of Trauma Surgery, Department of Surgery, Sutter Roseville Medical Center, Roseville, California, U.S.A.

Shock is a term used to broadly categorize a group of physiologic states, which, if left untreated, ultimately result in cardiovascular failure and death. Credit for the earliest use of this term goes to the French surgeon, Le Dran, in 1737. In his manuscript ‘‘A Treatise of Reflections Drawn from Experience with Gunshot Wounds,’’ the word ‘‘choc’’ was used to describe a severe jolt or impact (1). Over a century later, Gross eloquently described shock as ‘‘a rude unhinging of the machinery of life’’ (2), a sentiment shared by many who followed him, frustrated at the difficulty of resuscitating a patient from severe shock. At the end of the 19th century, shock was referred to as ‘‘a momentary pause in the act of death’’ by Warren (3). At approximately the same time, Crile noted a decrease in central venous pressure in response to hemorrhage, as well as a survival benefit associated with saline resuscitation (4). Blalock, in one of the first ‘‘modern’’ descriptions, defined shock as ‘‘peripheral circulatory failure, from a discrepancy between the size of the vascular bed and the volume of intravascular fluid’’ (5). Blalock’s description of shock has served as a starting point toward our current understanding of shock, which is characterized by the presence of inadequate tissue perfusion and oxygenation. As mentioned previously, shock is a very broad term that necessitates further subclassification. Four types of shock that are generally described are hypovolemic, distributive, cardiogenic, and obstructive. Hypovolemic shock most commonly results from hemorrhage, but may also be caused by any process that results in dehydration or volume loss. Distributive shock, resulting from alterations in vasomotor tone, is most commonly caused by sepsis. Less common causes of distributive shock include spinal cord injury, anaphylaxis, and adrenocortical insufficiency. Cardiogenic shock may be caused by any process that results in decreased myocardial contractility (e.g., myocardial infarction, cardiomyopathy, and blunt cardiac injury), and obstructive shock results from mechanical etiologies such as pericardial tamponade, tension pneumothorax, constrictive pericarditis, and ventricular outflow

obstruction. The remainder of this chapter will specifically address the diagnosis, management, and complications of hypovolemic and septic shock.

HYPOVOLEMIC SHOCK Hypovolemic shock may be caused by any process that ultimately results in depletion of intravascular volume. Most commonly, hypovolemic shock is related to hemorrhage, but it may also result from fluid sequestration in the extravascular space, which occurs in the setting of severe pancreatitis or burns. Insensible losses and volume loss via the gastrointestinal or urinary tract may also produce hypovolemic shock. Hypovolemic shock and, in particular, hemorrhagic shock are further classified by the percentage of lost intravascular volume (Table 1). It is particularly important to note that normal blood pressure may be maintained despite intravascular volume losses of up to 30%.

Diagnosis Uncompensated hypovolemic shock, comprising class III and class IV fluid loss, is usually diagnosed readily by alterations in blood pressure, heart rate, and urinary output. This determination is especially straightforward when an obvious source of hemorrhage is present. Compensated shock, which may be present despite significant intravascular volume depletion, requires a much higher index of suspicion because the diagnosis may be masked by relatively normal vital signs and urinary output. In compensated shock, regional alterations in blood flow occur because the body attempts to maintain perfusion of the brain and heart. While teleologically adaptive for short-term survival, this response occurs at the expense of other organ systems, in particular, the mesenteric circulation. As regional perfusion decreases, aerobic glycolysis is replaced at the cellular level by anaerobic metabolism.

56

Dolich and Van Boerum

Table 1 Classification of Volume Loss in Hypovolemic Shock

Percentage volume loss Blood pressure Heart rate Urinary output

Class I

Class II

Class III

Class IV

15% Normal Normal Normal

15–30% Normal " #

30–40% # "" ##

>40% ## """ ###

Source: From Ref. 6.

During this process, pyruvate is hydrolyzed into lactate, with a resultant rise in serum lactate levels. Excess Hþ ions generated during this reaction contribute to a metabolic lactic acidosis, which is one of the clinical hallmarks of hypovolemic shock. A complete description of the physiologic alterations associated with hypovolemic shock falls outside the scope of this chapter and the reader is directed to one of the many excellent texts on this subject. The incidence of compensated hemorrhagic shock is high in victims of trauma, as much as 85% in some studies (7,8). Early diagnosis is crucial because persistence of malperfusion associated with compensated shock has been shown to be highly associated with multiple organ dysfunction syndrome (MODS) and death (9). Because vital signs and urinary output may be normal despite derangements in perfusion, other modalities must be utilized to establish an early diagnosis of compensated shock. Among the simplest diagnostic measures are determination of biochemical parameters including serum lactate and arterial base deficit. Serum lactate levels provide a global estimate of anaerobic metabolism. Many studies have shown that high initial or maximal serum lactate levels correlate with severity of hypovolemic shock and subsequent mortality (10–12). Additionally, the duration of time for normalization of lactate levels after resuscitation has been directly associated with mortality. Failure to achieve a normal serum lactate level within 48 hours has been associated with mortality rates as high as 86% in some studies (12). Lactic acidosis resulting from hypovolemic shock may also be measured indirectly by arterial blood gas analysis. The arterial base deficit, as calculated by the Siggard–Anderson nomogram, is defined as the amount of additional base required to titrate 1 L of blood to a pH of 7.40, assuming a partial pressure of CO2 of 40 mmHg and a temperature of 37
C. In the absence of preexisting metabolic derangements, the arterial base deficit will vary directly with the level of lactic acidemia. Several studies have verified the clinical utility of arterial base deficit determinations in establishing the severity of hypovolemic shock, as well as assessing the adequacy of resuscitation (13,14). Thus, lactate and arterial base deficit determination represent simple, global diagnostic modalities in suspected hypovolemic shock. Invasive hemodynamic monitoring provides another avenue for establishing the presence of compensated hypovolemic shock. Pulmonary artery catheterization is an effective diagnostic maneuver

that is easily performed percutaneously at the bedside in most intensive care units (ICUs) today. Although some recent evidence (15) has associated pulmonary artery catheterization with increased mortality rates, the results of another large study suggest that pulmonary artery catheter-directed therapy significantly improves outcome (16). Despite some controversy in the literature, most intensivists consider the pulmonary artery catheter to be an indispensable tool in the clinical management of critically ill patients. Although pulmonary artery catheterization does not definitively establish the presence or absence of compensated hypovolemic shock, several parameters may lead the clinician to consider this diagnosis. Typically, hypovolemic shock is associated with low pulmonary artery pressures and central venous pressure. Pulmonary artery occlusion pressure (PAOP), measured distal to a balloon ‘‘wedged’’ in a proximal pulmonary vessel, provides a close approximation of left atrial pressure, a surrogate for ventricular end-diastolic volume (EDV) traditionally referred to as preload. Thus, in hypovolemia, PAOP is typically low. Cardiac output, easily measured by thermodilution, is secondarily low due to decreased preload, although compensatory sinus tachycardia may mitigate this finding. Systemic vascular resistance (SVR) is usually increased in an attempt to maintain central circulatory volume and pressure. Current pulmonary artery catheters also have the ability to continuously measure mixed venous oxygen saturation (SvO2) by reflectance oximetry. Decreased SvO2 levels generally indicate inadequate oxygen delivery, and thus may be associated with hypovolemia prior to changes in vital signs or urinary output. Newer pulmonary artery catheters also provide continuous measurement of right ventricular EDV and right ventricular ejection fraction. While the measured EDV of the right ventricle typically overestimates the left ventricular EDV, it may in fact be a more accurate predictor of preload than PAOP. Normal cardiovascular pressures are listed in Table 2. Normal hemodynamic parameters and their derivations are outlined in Table 3. In addition to global markers of hypoperfusion associated with hypovolemic shock, several newer modalities have emerged that allow better assessment of regional perfusion. Because gut perfusion decreases in the early stages of compensated hypovolemia and is restored relatively late during resuscitation, Table 2 Normal Cardiovascular Pressures Parameter Systolic blood pressure Diastolic blood pressure Mean arterial pressure Central venous pressure Pulmonary artery systolic Pulmonary artery diastolic Mean pulmonary arterial pressure Pulmonary artery occlusion pressure

Normal range (mmHg) 100–140 60–90 70–100 0–6 15–30 6–12 10–18 6–12

Chapter 5: Hypovolemic and Septic Shock

Table 3 Normal Hemodynamic Parameters Parameter

Derivation

Stroke volume  heart rate CI CO/body surface area SVR (MAP – CVP)/CO  80 SVR index (MAP – CVP)/CI  80 PVR (MPAP – PAOP)/CO  80 PVR index (MPAP – PAOP)/CI  80 Mixed venous Direct or oximetric oxygen measurement saturation RVEF Thermodilution RVEDV Stroke volume/RVEF RVEDV index Stroke volume index/RVEF CO

Normal value

57

however, at this time it is unclear whether NIRS-directed therapy will result in improved outcomes.

4–8 L/min

Treatment 2.5–4 L/min/m2 800–1400 dyne/sec/cm5 1200–2500 dyne/sec/cm5/m2 80–200 dyne/sec/cm5 200–400 dyne/sec/cm5/m2 68–72%

35–60% 150–225 mL 80–110 mL

Abbreviations: CO, cardiac output; CI, cardiac index; SVR, systemic vascular resistance; PVR, pulmonary vascular resistance; RVEF, right ventricular ejection fraction; RVEDV, right ventricular end-diastolic volume; MAP, mean arterial pressure; CVP, central venous pressure; PAOP, pulmonary artery occlusion pressure; MPAP, mean pulmonary artery pressure.

recent attention has been directed toward assessment of this parameter. One technique that has shown some promise involves measurement of gastric intramucosal pH (pHi) by method of gastric tonometry. This is accomplished by insertion of a specialized nasogastric tube constructed with a semipermeable membrane balloon tip. Carbon dioxide is able to equilibrate across this membrane, allowing determination of the regional CO2 concentration within the gastric mucosa. During periods of hypoperfusion, the regional mucosal CO2 rises with a resultant decrease in pHi, as calculated by the Henderson–Hasselbalch equation: pHi ¼ 6:1 þ logðHCO 3 =PCO2  0:03Þ It is important to note that gastric tonometry, by detecting changes in regional CO2 concentration, measures pHi (which is related to perfusion), as opposed to ‘‘intraluminal’’ pH (which is related to gastric acid secretion). Thus, gastric tonometry can be used as a window through which the perfusion status of the foregut can be assessed. Studies in animals and humans have verified that gastric tonometry provides an accurate assessment of mucosal perfusion in hypovolemia and shock. However, controlled trials of pHi-directed interventions have, in general, failed to demonstrate improved outcomes in most studies to date (17,18). Another modality by which regional perfusion can be estimated is near-infrared spectroscopy (NIRS). This technique relies on differential absorption and reflectance patterns of mitochondrial cytochrome a and cytochrome a3 that vary depending on the state of cellular respiration. As photons in the near-infrared spectrum pass through bone and soft tissue, NIRS transducers have been designed to noninvasively assess various tissue beds ranging from gastric mucosa to the brain. NIRS has been shown to reflect alterations in mesenteric blood flow associated with hemorrhagic shock (19);

Therapy for hypovolemic shock is both deceptively simple and frustratingly controversial. The two mainstays of treatment are correcting the source of intravascular fluid loss (e.g., hemorrhage) and restoration of adequate circulating volume. Once a source of hypovolemia is identified, rectification is usually straightforward and may involve surgical or angiographic hemostasis, medical intervention to decrease gastrointestinal or urinary losses, or measures to combat insensible fluid loss. However, controversies persist regarding the appropriate intravenous fluid for resuscitation, as well as the timing and volume of fluid to be given. Four basic categories of resuscitative fluid may be utilized in hypovolemic shock: crystalloids, colloids, hypertonic solutions, and oxygen-carrying solutions such as blood products. Each has relative advantages and disadvantages; the past four decades have witnessed fervent debate regarding the appropriate choice for treating hypovolemic shock. Resuscitative efforts should be toward achieving two main goals: restoration of adequate circulating volume, and maintenance of adequate oxygen delivery.

Crystalloid Solutions Isotonic crystalloid solutions such as lactated Ringer’s (LR) and normal saline (NS) comprise the initial mainstay of therapy for hypovolemic shock. These solutions are cheap, safe, and readily available. In hemorrhagic shock, both are equally effective as resuscitative fluids when given in a ratio of threevolume units of crystalloid solution per unit volume of shed blood. However, each fluid does have certain advantages and disadvantages. LR infusion may ameliorate the metabolic acidosis associated with hypovolemic shock, as the lactate it contains is rapidly metabolized to bicarbonate by the liver. Unfortunately, infusion of large volumes of LR also appears to have a negative immunologic effect, in particular, the activation of neutrophils in the blood stream. This neutrophil priming effect may be associated with development of systemic inflammatory response syndrome (SIRS) and MODS (20). Additionally, because LR is mildly hypotonic (Na ¼ 130), infusion of large volumes of this fluid in the setting of traumatic brain injury may be deleterious due to worsening cerebral edema, intracranial pressure, and secondary brain injury. Thus NS, due to its mild hypertonicity (Na ¼ 154 mmol/L), may be superior to LR in this setting. An additional advantage of NS is that it may be safely infused with banked blood, whereas the calcium content of LR overwhelms the chelating ability of citrate in stored blood. This may result in clot formation; thus, infusion of blood and LR together is

58

Dolich and Van Boerum

relatively contraindicated. Conversely, administration of large volumes of NS may result in a hyperchloremic metabolic acidosis, an undesirable side effect in the hypovolemic, acidemic patient.

Colloid Solutions Colloid solutions, including human serum albumin, modified starches, and dextrans, have provided alternatives to crystalloids for over half a century. They offer, at least theoretically, the advantage of increased intravascular colloid oncotic pressure, resulting in a net influx of fluid from the interstitium into the vascular space. Multiple studies have shown that resuscitation with colloid entails significantly less total volume infusion when compared with isotonic crystalloid. A secondary benefit of decreased volume requirement is decreased resuscitation time. Hundreds of studies comparing crystalloid with colloid have been performed over the last 25 years; however, most large reviews and meta-analyses of the available prospective data fail to show any survival benefit favoring colloid solutions in the setting of hypovolemic shock (21). In fact, albumin administration in hypovolemic shock may increase mortality rate (22). Colloid resuscitation is extremely expensive when compared with crystalloid; thus, in the absence of clear benefit, it is difficult to recommend the routine use of colloids in hypovolemic shock.

Hypertonic Solutions Because resuscitation with isotonic crystalloid inherently involves relatively large volumes of fluid, recent interest has focused on hypertonic solutions, which have the benefit of requiring much smaller volume infusion. Decreased infusion volumes may provide significant advantage in prehospital and military settings, where carriage of large volumes of intravenous fluid is impractical. The best studied of these fluids are 7.5% NaCl hypertonic saline (HTS) and 7.5% NaCl/6% dextran 70 solution (HTS-D). Hypertonic solutions may provide additional benefits in patients with traumatic brain injury, as the high osmolarity of these solutions may ameliorate the cerebral edema and decreased cerebral perfusion that accompany large-volume fluid resuscitation (23). In addition, HTS causes less neutrophil activation and adhesion when compared with LR (20). Thus, it is possible that large-volume resuscitation with HTS or HTS-D may reduce the inflammatory response that accompanies LR infusion, but human data are lacking at this time.

Blood and Blood Substitutes In hypovolemic shock due to significant hemorrhage, the asanguinous fluids mentioned in the previous paragraphs may be inadequate for complete resuscitation. In this circumstance, oxygen delivery must be augmented with an oxygen-carrying solution. Most commonly, this is accomplished by allogenic blood transfusion, typically in the form of packed red

blood cells (RBC). Although RBC transfusion is frequently life saving, it is also associated with many pitfalls and complications. Despite improvements in screening techniques, small but finite risks of infectious complications, including infections with human immunodeficiency virus, human T-lymphotrophic virus, hepatitis C, and hepatitis B, persist. Blood transfusion is also associated with immunosuppression and increased frequency of wound infection, and has been shown to increase the risk of developing MODS (24). Perhaps most importantly, blood is an extremely limited and costly resource. In light of the limitations of RBC transfusion, recent efforts have focused on other oxygen-carrying blood substitutes. The goal has been to identify a blood substitute that has a long shelf life, is universally compatible, has no potential for disease transmission, and has oxygen-carrying and dissociation characteristics similar to blood. Although initially promising, perfluorocarbon-based blood substitutes have had disappointing results, and most current efforts center on the clinical utility of hemoglobin-based oxygen carriers (HBOC). Early stroma-free hemoglobin solutions contained significant amounts of monomeric and dimeric hemoglobin, which led to nephrotoxicity. Subsequent formulations included cross-linking of hemoglobin, reducing the incidence of this complication. However, pulmonary and systemic hypertension, thought secondary to binding of nitric oxide by hemoglobin, continued to be undesirable side effects. More recently, this effect has been addressed by creation of polymerized hemoglobin solutions, which appear to have decreased affinity for nitric oxide and fewer undesirable vasoactive effects, while maintaining oxygen affinity. Currently, three HBOCs are undergoing phase III clinical trials: human glutaraldehyde polymerized hemoglobin, human o-raffinose polymerized hemoglobin, and ultrapurified polymerized bovine hemoglobin.

Delayed Fluid Resuscitation For most of the last half century, immediate administration of large volumes of resuscitative fluid has been the mainstay of treatment in hypovolemic shock. Upon further reflection, this practice may be questionable in the setting of uncontrolled hemorrhage. In theory, hemorrhagic shock is teleologically adaptive in the short term, as blood is directed to the brain and heart, and lower blood pressure allows hemostasis at the site of hemorrhage. Standard treatment with large volumes of crystalloid results in a dilutional coagulopathy, hypothermia, decreased blood viscosity, and elevated blood pressure, all of which may favor recurrent hemorrhage from an unsecured vessel. This has proven to be the case in animal studies, where immediate resuscitation with LR resulted in an increased blood loss following aortotomy (25). A recent prospective clinical trial has supported this hypothesis; overall survival and hospital stay were

Chapter 5: Hypovolemic and Septic Shock

improved in penetrating truncal trauma victims receiving delayed fluid resuscitation after surgical hemostasis, as compared with patients receiving immediate fluid resuscitation in the prehospital setting (26).

SEPTIC SHOCK Septic shock is a state of complete homeostatic deterioration resulting from a host’s response to infection. It is a form of distributive shock, characterized by inadequate tissue perfusion due to microcirculatory alterations in the blood flow. Severe sepsis is estimated to affect 750,000 patients in the United States annually, with an associated mortality rate of approximately 30%. By comparison, a similar number of people in the United States die of acute myocardial infarction each year. Despite advances in critical care, the incidence of sepsis is not declining, and may indeed be on the rise. As the average age in industrialized nations increases, so does the number of sepsisrelated ICU admissions. The average length of stay for septic patients is nearly three weeks, consuming an estimated total of $16.7 billion each year in the United States (27).

Definitions To adequately address the diagnosis, pathophysiology, and management of sepsis-related disorders, appropriate terminology must be defined. The term sepsis is broad and has been traditionally used to describe a vast spectrum of disease. Over the last 20 years, attempts have been made to apply more uniform diagnostic criteria to better define specific disease entities. Fairly specific criteria now define the terms sepsis, severe sepsis, septic shock, and SIRS (28,29). ‘‘Sepsis’’ refers to the systemic response to infection, including altered temperature, heart rate, respiratory rate, and leukocyte count. ‘‘Severe sepsis’’ adds the criterion of associated end-organ dysfunction to the above. Severe sepsis with hypotension or other evidence of hypoperfusion is referred to as ‘‘septic shock.’’ ‘‘SIRS’’ is defined by the same criteria

59

as sepsis, only in the absence of documented or presumed infection. SIRS may occur in a variety of clinical circumstances, including trauma, burns, major surgery, and massive blood transfusion. Generally accepted diagnostic criteria are listed in Table 4, although considerable variability persists within the literature. Medications such as vasopressors or betablockers may alter physiologic parameters used in the definitions. This, in part, explains the inconsistency found in published reports. The remainder of this section will address the pathophysiology and management of septic shock.

Pathophysiology A complete understanding as to the pathophysiology of septic shock remains elusive. One major theory holds that sepsis represents an uncontrolled overstimulation of the inflammatory response. This theory has been challenged many times, and alternate hypotheses demonstrate evidence of immune failure. It is likely that both theories are at least partially correct. Early in the course of sepsis, the immune system appears to become activated, with high levels of circulating tumor necrosis factor (TNF)-a and interleukin (IL)-1. Increased levels of these cytokines are associated with the clinical hyperdynamic state. Later in the course of septic illnesses, there is a shift toward anti-inflammatory mediator production. Patients may eventually become anergic and more susceptible to nosocomial infections. Much research has been done, beyond the scope of this chapter, attempting to clarify the complex interaction of bacteria and their byproducts with the host production of cytokines and other inflammatory mediators. While inflammatory cytokines have long been felt to be generally deleterious in sepsis, it is also known that TNF-a plays an important role in combating infection. Studies in both animals and humans have shown worse outcomes when TNF-a was blocked (30–33). More recently, a strong association between the inflammatory response of sepsis and perturbations of coagulation has been established. Specifically, sepsis-associated alterations in conversion of protein C into its activated form may lead to microvascular thrombosis, which may, in

Table 4 Sepsis: Definitions Systemic inflammatory response syndrome (SIRS)

Sepsis Severe sepsis Septic shock

The generalized inflammatory response to a variety of clinical insults. The response is manifested by two or more of the following conditions: (i) temperature >38
C or 90 beats/min; (iii) respiratory rate >20 breaths/min or PaCO2 12,000/mm3, 10% band forms. SIRS as described above with a presumed or confirmed infection. Sepsis as described above with associated organ dysfunction, hypoperfusion, or hypotension. Perfusion abnormalities may include, but are not limited to, lactic acidosis, oliguria, or an acute alteration in mental status. Severe sepsis as described above with associated hypotension despite adequate fluid resuscitation. Patients on inotropic or vasopressor agents may not be hypotensive at the time perfusion abnormalities are measured.

Abbreviation: PaCO2, arterial pressure of carbon dioxide. Source: From Ref. 28.

60

Dolich and Van Boerum

turn, be at least partially responsible for end-organ dysfunction. Ongoing research will hopefully allow a more thorough understanding of this complex disease process.

Treatment There are five crucial components in the management of septic shock: fluid resuscitation, septic source control, appropriate use of vasopressor agents, appropriate use of inotropic agents, and rational utilization of antimicrobials. More recently, control of the perturbation of coagulation inherent in septic shock is emerging as a promising adjunctive measure that will be discussed in a later section. Once septic shock is suspected, rapid institution of these steps is necessary to avoid MODS and death. Early intervention has clearly been shown to improve outcomes and suggests that invasive hemodynamic monitoring and goal-directed therapy be initiated as early as possible (34).

Fluid Resuscitation The initial treatment of septic shock such as hypovolemic shock involves intravenous fluid administration. Accompanying the systemic vasodilation of septic shock is disruption of microvascular endothelial integrity. Although the exact mechanism has not clearly been elucidated, the result is a net egress of intravascular fluid and protein into the interstitial space. This phenomenon of fluid sequestration is commonly referred to as a ‘‘capillary leak syndrome.’’ Vasodilation and capillary leak conspire to decrease the central circulating blood volume and result in relative hypovolemia. Maintenance of adequate perfusion requires careful and repeated assessment of volume status, with fluid resuscitation guided by hemodynamic parameters. In most instances, an isotonic crystalloid such as LR solution is the preferred initial resuscitative fluid. Blood products may be required if cardiac ischemia is present or if anemia precludes adequate oxygen delivery. Transfusion of blood products should be based on physiologic variables rather than the predetermined notion of an ‘‘optimal’’ hematocrit of 30%. Inappropriate blood transfusion may actually increase the incidence of infectious complications, cardiac events, and mortality (35,36). Fluid management in septic shock mandates frequent and repeated assessments of intravascular volume. Total body weight or running totals of net fluid balance are not effective surrogates for intravascular volume status. In most instances, pulmonary artery catheter-guided fluid administration is invaluable, as fluid shifts may occur quite rapidly. Diuretic administration based on oliguria in septic shock is frequently detrimental and may increase the likelihood of death. Aggressive fluid replacement should be continued until endothelial integrity is reestablished. Onset of recovery is generally heralded by fluid mobilization and diminished volume requirements.

Septic Source Control In an ICU population, numerous potential sources for infection exist. An adequate history and careful examination of the patient coupled with appropriate use of diagnostic studies is essential to accurately localize an infectious focus serving to fuel septic shock. The process of accurately localizing a source of infection can frequently be quite challenging. In many instances, the septic source is a site of previous intervention. It is essential that all surgical and traumatic wounds be carefully examined for evidence of superficial or deep infection. Implanted prosthetic devices should always be considered as possible sources of infection. Peripheral intravenous catheters, central venous lines, prosthetic meshes, and implanted orthopedic hardware may all become infected and lead to an episode of septic shock. Because many ICU patients have undergone recent abdominal surgery, consideration of intra-abdominal abscess, anastamotic leak, bowel ischemia, or iatrogenesis should prompt appropriate diagnostic workup. Computed tomography has proven invaluable in this regard, and modern highresolution helical scanners are extremely accurate in detecting intra-abdominal sources of sepsis. Abscesses should be drained after initiating fluid resuscitation, so as to avoid cardiovascular collapse, which may result from the cytokine surge accompanying a drainage procedure. The importance of abscess drainage should not be underestimated because it is very rare for antibiotics alone to control an abscess. Drainage also facilitates culture-directed, narrowspectrum antibiotic therapy, with the important secondary goal of prevention of antibiotic resistance. Overwhelmingly, the most common site of infection in patients in modern ICUs is the respiratory tract. Ventilator-associated pneumonia and aspiration pneumonia are unfortunately common. Endotracheal intubation and mechanical ventilation for greater than 24 hours significantly increases the risk of developing pneumonia. Despite their frequency, however, respiratory infections are not common causes of septic shock. As a general rule, septic shock should not be attributed to an infiltrate on chest X ray without first giving adequate consideration to other possible sources. A common pitfall is the delayed diagnosis of severe intra-abdominal pathology in a patient being treated for presumed pneumonia, often with disastrous consequences. Infections related to central venous catheterization represent another relatively common source of septic shock. It is estimated that over five million central venous catheters are utilized yearly in the United States alone, and most critically ill patients will require central venous catheterization at some point. The advent of central venous catheterization has truly been a double-edged sword. Central venous catheterization provides an avenue for monitoring and treatment of critically ill patients, while at the same time serving as an intravascular prosthetic nidus for bacteria. The

Chapter 5: Hypovolemic and Septic Shock

risk of catheter-related infection is directly proportional to both the number of catheter lumens and the duration of catheter dwelling time. Catheters may become colonized by hematogenous seeding from other septic foci or, more commonly, via direct transcutaneous colonization at the insertion site. Catheters with 15 or more colony-forming units of bacteria, as described by Maki et al., are associated with higher relative risk of bacteremia and sepsis and should be removed (37). Further management depends on the organism isolated and the patient’s clinical status. Uncomplicated gram-positive catheter-related bacteremia in the minimally symptomatic patient may only require removal of the offending device. Gramnegative organisms, resistant microbes, or grampositive infections with persistent systemic symptoms usually require adjuvant antibiotic therapy. Prevention is the best method of treatment. Sterile technique at the time of insertion as well as at any time of manipulation is mandatory. Catheters should be removed when no longer needed, so as to limit the number of catheter days for any given patient. Occasionally, the source of septic shock remains unclear despite aggressive evaluation. Careful consideration of less common etiologies should be given to avoid unnecessary delays in diagnosis. Sinusitis, perirectal abscess, meningitis, acalculous cholecystitis, ischemic colitis, and infected decubitus ulcer may all result in a septic picture. A high index of suspicion is crucial so that infectious foci can be drained or removed.

Vasopressors Persistent hypotension despite adequate ongoing fluid administration may be an indication for vasopressor administration (38). Dopamine, epinephrine, norepinephrine, phenylephrine, and vasopressin are potential agents for use in this setting. The goal of vasopressor administration is to provide enough mean arterial pressure to maintain organ perfusion. This may be accomplished by counteracting the vasodilatory effects of circulating mediators released during SIRS. Vasoactive agents are appropriately administered after adequate volume resuscitation has been effected because failure to do so may increase morbidity and mortality by vasoconstriction of already ischemic capillary beds. Because shock, by definition, results in inadequate perfusion at the cellular level, global pharmacologic vasoconstriction in an effort to raise SVR and blood pressure also has the potential for reduction in end-organ blood flow. Teleologically, one might expect that the ability to autoregulate perfusion to ischemic tissues would give the most vital organs priority and protection. This occurs in hypovolemic or hemorrhagic shock, where blood flow is redirected from the periphery in order to maintain cerebral and cardiac perfusion. In sepsis, as well as some chronic disease states, the ability to autoregulate blood flow may be impaired. Recently

61

it has been suggested that sepsis is associated with subphysiologic levels of circulating vasopressin (39). Replacement therapy with extremely low-dose vasopressin (0.01–0.04 units/min) has resulted in hemodynamic improvement without the detrimental effects of vasoconstriction (40,41). Exogenous vasopressors must be administered with great caution because the potential for serious adverse effect exists. Cardiac, mesenteric, and digital ischemia with infarction are well-described complications of vasopressor therapy, and a potential for worsening renal insufficiency exists as well. Tachycardia is frequently observed and may be mild, or, in certain circumstances, may be severe and limit therapeutic use of these drugs. In addition to unintended local consequences, global deleterious effects may also be observed. By increasing afterload, vasopressors have the potential to lower cardiac output and thereby exacerbate tissue oxygen debt. In general, the adage of ‘‘less is more’’ is well applied here. These agents should only be used in doses required to achieve the lowest acceptable blood pressure required to maintain cerebral and coronary perfusion. Continuous mixed SvO2 combined with continuous cardiac output monitoring is often extremely helpful in titration of this therapy.

Inotropes Septic shock is frequently characterized by a hyperdynamic state, with hypotension, low SVR, and a cardiac index that is normal or elevated (42). It should be noted, however, that an elevated cardiac index does not ensure adequate oxygen delivery; thus, inotropic agents are occasionally required to augment oxygen delivery in the septic patient (43). Retrospective studies (44–46) have repeatedly shown that septic patients who were able to mount a hyperdynamic cardiovascular response have lower mortality than patients with normal or subnormal cardiac indexes. Extending the concept that supranormal cardiac performance was associated with improved outcome, Shoemaker and coworkers popularized the concept of flow-dependent oxygen consumption and suggested that resuscitation should be directed to push cardiac performance to a state of flow-independent oxygen consumption (47,48). However, randomized prospective trials that employed inotropic agents to increase cardiac indices to predetermined levels have had mixed results (49–52). Thus, the goal should be to provide enough cardiac output to meet the metabolic demands of the patient. This assessment should be made using serial lactate measurements, arterial blood gas analysis, and establishment of acceptable mixed SvO2. Complications of inotropic support are most frequently cardiac in origin. Improved inotropy comes at a cost of increased myocardial oxygen consumption. Although inotropic support is usually well tolerated in young patients, elderly patients with coronary

62

Dolich and Van Boerum

artery disease may experience myocardial ischemia or infarction. Dobutamine, one of the most commonly used inotropes, is frequently associated with hypotension due to associated vasodilation. This effect can frequently be overcome by additional fluid administration. In instances where severe hypotension persists, dobutamine may require concomitant infusion of a vasopressor. Occasionally, hypotension or tachyarrhythmia may limit or preclude dobutamine administration. In these circumstances, secondline inotropic agents such as amrinone or milrinone may be utilized.

Antibiotics After adequate resuscitation of a patient with apparent sepsis or SIRS, two challenges remain: whether to administer antibiotics, and, if so, which antibiotic to administer. As mentioned previously, it can be quite difficult to differentiate between sepsis and SIRS, as SIRS is frequently a diagnosis of exclusion. This pursuit is an important one, however, because antibiotics provide no therapeutic benefit in the patient with SIRS in the absence of infection. Patients with presumed septic shock should receive empiric antibiotic therapy directed at likely pathogens while the search for a septic focus occurs and microbiologic data are gathered. When feasible, cultures should be taken prior to administration of antibiotics. Depending on the clinical circumstance, broad-spectrum agents may be required initially. Empiric therapy may be guided by the institution’s antibiogram. When available, culture-directed antibiotic therapy should be the standard. The antibiotic spectrum should be narrowed to specifically cover the pathogens known or believed to be responsible for the infection. It is crucial that antibiotic therapy does not overshadow the need for surgical source control, if indicated. If infection is ruled out, empiric antibiotics should be discontinued.

Coagulation In Sepsis As our understanding of cellular and molecular pathophysiology has progressed, many mediators involved in the inflammatory response of septic shock have been identified. Many investigators have put forth great efforts to identify an agent capable of attenuating or enhancing these mediators in the hope of favorably altering the inflammatory response present in septic patients. The search for a ‘‘magic bullet’’ antisepsis drug has, for the most part, had disappointing results. Anti-TNF antibodies, endotoxin-binding antibodies, IL-1 antagonists, and tissue factor pathway inhibitors are a few examples of sepsis drugs that have failed to improve the outcomes (53–62). This is likely the effect of a system with many inherent redundancies, as well as the fact that the inflammatory response is exceedingly complex and, at best, only partially understood. Recent attention has turned toward the role of coagulation in sepsis, specifically regarding the

role of protein C. Protein C is best known for its anticoagulant role in maintaining microcirculatory homeostasis. Protein C is activated by the thrombin– thrombomodulin complex, which is formed by the binding of thrombin to thrombomodulin. Activated protein C then in turn limits thrombin formation and downregulates coagulation activation by inactivating factor VIIIa and factor Va, which are cofactors for factor IXa-induced factor X activation and factor Xa–induced prothrombin activation, respectively (63–65). There exists significant overlap and interaction between mediators of the coagulation cascade and promoters of inflammation (66,67). The exact mechanism by which protein C interferes with the propagation of the inflammatory response in sepsis has not yet been fully elucidated. It may be that the conversion of protein C to its activated form may be impaired during sepsis due to downregulation of thrombomodulin by inflammatory cytokines (68). Protein C has been shown to interfere with cytokine production by monocytes. In light of the above observations, a new agent that has shown promise in recent studies is recombinant human activated protein C (rHAPC). In a recent randomized, prospective, double-blind, placebo-controlled, multicenter trial (69) of nearly 1700 subjects, septic patients with evidence of end-organ dysfunction were randomized to receive a course of rHAPC or placebo. A significant mortality benefit was observed in the rHAPC-treated group. The beneficial effect of rHAPC may be secondary to an anticoagulant effect ameliorating microvascular thrombosis induced organ damage; alternately, there may be a direct anti-inflammatory effect as well. The anticoagulant effect of rHAPC can result in significant morbidity and mortality, however. Bleeding complications may severely limit the use of this agent in trauma victims or surgical patients. Additional clinical trials of rHAPC in sepsis are underway.

SPECIFIC COMPLICATIONS OF HYPOVOLEMIC AND SEPTIC SHOCK Abdominal Compartment Syndrome Compartment syndrome is defined as increasing pressure within an unyielding body cavity or compartment, resulting in organ injury, ischemia, or other undesirable physiologic effects. Commonly seen in the lower extremity following crush injury of muscle, it may also develop in the abdomen following injury, shock, or resuscitation. In abdominal compartment syndrome (ACS), visceral edema within the bony, fascial, and muscular confines of the abdominal cavity leads to rising intra-abdominal pressure. Visceral swelling may occur in directly injured tissues, or may follow large-volume resuscitation of various shock states. Elevated intra-abdominal pressure results in respiratory impairment secondary to cephalad displacement of both hemidiaphragms. Increased pressure on renal capillaries diminishes

Chapter 5: Hypovolemic and Septic Shock

kidney perfusion, resulting in renal insufficiency. Impaired splanchnic venous return may worsen visceral edema, creating a vicious cycle that, if left untreated, frequently results in MODS and death. Additionally, ACS may increase intracranial pressure in head-injured patients (70). ACS is most frequently encountered in the setting of major abdominal injury or ruptured abdominal aortic aneurysm. It may occur postoperatively or intraoperatively, as an unsuspecting surgeon struggles to close a laparotomy incision over particularly distended or edematous abdominal contents. However, ACS may occur even in the absence of abdominal injury or surgery, and has been described in burns, isolated orthopedic injuries, and septic shock from nonabdominal sources. Overall incidence in ICU patients is estimated to be between 2% and 5%, but the mortality rate of ACS is approximately 50% (71,72). Intra-abdominal hypertension leading to ACS may be diagnosed with relative ease by manometric transduction of urinary bladder pressure. Because the dome of a partially filled urinary bladder acts as a passive diaphragm to the abdominal cavity, the pressure within the bladder serves as an accurate surrogate for intra-abdominal pressure. Pressures above 20 mmHg may be associated with the physiologic derangements of ACS; treatment involves decompressive laparotomy with temporary silo closure.

Adrenocortical Insufficiency The response to shock typically involves activation of the hypothalamic-pituitary-adrenal (HPA) axis by neural and systemic pathways. Septic shock, in particular, results in HPA stimulation due to circulating cytokines such as TNF-a, IL-1, and IL-6. However, in up to 50% of septic patients, blunted HPA response may occur, resulting in relative adrenocortical insufficiency. Most commonly, this phenomenon is observed in vasopressor-dependent septic shock. Early studies of steroid administration in septic shock utilized large doses of glucocorticoids aimed at ameliorating the inflammatory cascade. The underlying hypothesis at the time was that high-dose steroid-induced interruption of inflammation might decrease the incidence of MODS. However, the results of these early trials of high-dose steroids in sepsis were uniformly disappointing, and mortality increases were even noted. However, in light of the frequent observation of HPA suppression in vasopressordependent sepsis, there has been renewed interest in low-dose steroid replacement. In a recent randomized, prospective study (73), low-dose hydrocortisone and fludrocortisone administration significantly decreased both mortality and duration of vasopressor therapy in septic patients. At the very least, septic patients with exogenous catecholamine dependency should receive a cosyntropin stimulation test; nonresponders should be considered for corticosteroid replacement therapy.

63

REFERENCES 1. Le Dran HF. A Treatise, or Reflections Drawn from Practice on Gun-Shot Wounds. Avignon: Jean Chaillot, 1790. 2. Gross SD. A System of Surgery: Pathological, Diagnostic, Therapeutique, and Operative. 4th ed. Philadelphia: H.C. Lea, 1866. 3. Warren JC. Surgical Pathology and Therapeutics. Philadelphia: WB Saunders, 1895. 4. Crile GW. An Experimental Research into Surgical Shock. Philadelphia: JB Lippincott, 1899. 5. Blalock A. Principles of Surgical Care, Shock, and Other Problems. St. Louis: CV Mosby, 1940. 6. Advanced Trauma Life Support Manual. Chicago, Illinois: American College of Surgeons, 1997. 7. Abou-Khalil B, Scalea TM, Trooskin SZ, Henry SM, Hitchcock R. Hemodynamic responses to shock in young trauma patients: need for invasive monitoring. Crit Care Med 1994; 22(4):633–639. 8. Scalea TM, Maltz S, Yelon J, Trooskin SZ, Duncan AO, Sclafani SJ. Resuscitation of multiple trauma and head injury: role of crystalloid fluids and inotropes. Crit Care Med 1994; 22(10):1610–1615. 9. Manikis P, Jankowski S, Zhang H, Kahn RJ, Vincent JL. Correlation of serial blood lactate levels to organ failure and mortality after trauma. Am J Emer Med 1995; 13(6):619–622. 10. Dunham CM, Siegel JH, Weireter L, et al. Oxygen debt and metabolic acidemia as quantitative predictors of mortality and the severity of the ischemic insult in hemorrhagic shock. Crit Care Med 1991; 19(2):231–243. 11. Vincent JL, Dufaye P, Berre J, Leeman M, Degaute JP, Kahn RJ. Serial lactate determinations during circulatory shock. Crit Care Med 1983 11(6):449–451. 12. Abramson D, Scalea TM, Hitchcock R, Trooskin SZ, Henry SM, Greenspan J. Lactate clearance and survival following injury. J Trauma 1993; 35:584–589. 13. Davis JW, Kaups K. Base deficit in the elderly: a marker of severe injury and death. J Trauma 1998; 45(5):873–877. 14. Davis JW, Shackford SR, Mackersie RC, Hoyt DB. Base deficit as a guide to volume resuscitation. J Trauma 1988; 28(10):1464–1467. 15. Connors AF Jr, Speroff T, Dawson NV, et al. The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. JAMA 1996; 276(11):889–897. 16. Mimoz O, Rauss A, Rekik N, Brun-Buisson C, Lemaire F, Brochard L. Pulmonary artery catheterization in critically ill patients: a prospective analysis of outcome changes associated with catheter-prompted changes in therapy. Crit Care Med 1994; 22:573. 17. Gomersall CD, Joynt GM, Freebairn RC, Hung V, Buckley TA, Oh TE. Resuscitation of critically ill patients based on the results of gastric tonometry: a prospective, randomized, controlled trial. Crit Care Med 2000; 28(3):607–614. 18. Ivatury RR, Simon RJ, Islam S, Fueg A, Rohman M, Stahl WM. A prospective randomized study of end points of resuscitation after major trauma: global oxygen transport indices versus organ-specific gastric mucosal pH. J Am Coll Surg 1996; 183(2):145–154. 19. Cohn SM, Varela JE, Giannotti G, et al. Splanchnic perfusion evaluation during hemorrhage and resuscitation with gastric near-infrared spectroscopy. J Trauma 2001; 50(4):629–634.

64

Dolich and Van Boerum

20. Rhee P, Wang D, Ruff P, et al. Human neutrophil activation and increased adhesion by various resuscitation fluids. Crit Care Med 2000; 28(1):264–265. 21. Alderson P, Schierhout G, Roberts I, Bunn F. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev 2000; (2). 22. Alderson P, Bunn F, Lefebvre C, et al. Human albumin solution for resuscitation and volume expansion in critically ill patients. Cochrane Database Syst Rev 2002; (1). 23. Wade CE, Grady JJ, Kramer GC, et al. Individual patient cohort analysis of the efficacy of hypertonic saline/dextran in patients with traumatic brain injury and hypotension. J Trauma 1997; 42:S61–S65. 24. Sauaia A, Moore FA, Moore EE, Haenel JB, Read RA, Lezotte DC. Early predictors of postinjury multiple organ failure. Arch Surg 1994; 129(1):39–45. 25. Holmes JF, Sakles JC, Lewis G, Wisner DH. Effects of delaying fluid resuscitation on an injury to the systemic arterial vasculature. Acad Emerg Med 2002; 9(4):267–274. 26. Bickell WH, Wall MJ, Pepe PE, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating truncal injuries. N Engl J Med 1994; 331(17):1105–1109. 27. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome and associated costs of care. Crit Care Med 2001; 29(7): 1303–1310. 28. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992; 101(6):1644–1655. 29. Balk RA. Severe sepsis and septic shock: definitions, epidemiology, and clinical manifestations. Crit Care Clin 2000; 16(2):179–192. 30. Echtenacher B, Weigl K, Lehn N, Mannel DN. Tumor necrosis factor-dependent adhesions as a major protective mechanism early in septic peritonitis in mice. Infect Immun 2001; 69(6):3550–3555. 31. Eskandari MK, Bolgos G, Miller C, Nguyen DT, DeForge LE, Remick DG. Anti-tumor necrosis factor antibody therapy fails to prevent lethality after cecal ligation and puncture or endotoxemia. J Immunol 1992; 148(9):2724–2730. 32. Opal SM, Cross AS, Jhung JW, et al. Potential hazards of combination immunotherapy in the treatment of experimental septic shock. J Infect Dis 1996; 173(6):1415–1421. 33. Fisher CJ, Agosti JM, Opal SM, et al. Treatment of septic shock with the tumor necrosis factor receptor: Fc fusion protein. N Engl J Med 1996; 334(26):1697–1702. 34. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345(19):1368–1377. 35. Hebert PC, Wells G, Blajchman MA, et al. A multicenter randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. NEJM 1999; 340(6):409–417. 36. Hebert PC, Yetisir E, Martin C, et al. Is a low transfusion threshold safe in critically ill patients with

37.

38. 39. 40. 41. 42.

43. 44.

45.

46.

47.

48.

49.

50.

51. 52. 53.

cardiovascular diseases? Crit Care Med 2001; 29(2): 227–234. Maki DG, Weise CE, Sarafin HW. A semiquantitative culture method for identifying intravenous-catheterrelated infection. N Engl J Med 1977; 296(23):1305– 1309. Rudis MI, Basha MA, Zarowitz BJ. Is it time to reposition vasopressors and inotropes in sepsis? Crit Care Med 1996; 24(3):525–537. Landry DW, Levin HR, Gallant EM, et al. Vasopressin deficiency contributes to the vasodilation of septic shock. Circulation 1997; 95(5):1122–1125. Malay MB, Ashton RC Jr, Landry DW, Townsend RN. Low-dose vasopressin in the treatment of vasodilatory septic shock. J Trauma 1999; 47(4):699–703. Landry DW, Levin HR, Gallant EM, et al. Vasopressin pressor hypersensitivity in vasodilatory septic shock. Crit Care Med 1997; 25(8):1279–1282. Parrillo JE, Parker MM, Natanson C, et al. Septic shock in humans: advances in the understanding of pathogenesis, cardiovascular dysfunction, and therapy. Ann Intern Med 1990; 113(3):227–242. Parker MM, Shelhamer JH, Bacharach SL, et al. Profound but reversible myocardial depression in patients with septic shock. Ann Intern Med 1984; 100(4):483–490. Edwards JD, Brown GCS, Nightingale P, Slater RM, Faragher EB. Use of survivors’ cardiorespiratory values as therapeutic goals in septic shock. Crit Care Med 1989; 17(11):1098–1103. Gilbert EM, Haupt MT, Mandanas RY, Huaringa AJ, Carlson RW. The effect of fluid loading, blood transfusion, and catecholamine infusion on oxygen delivery and consumption in patients with sepsis. Am Rev Respir Dis 1986; 134(5):873–878. Shoemaker WC, Appel PL, Kram HB. Hemodynamic and oxygen transport responses in survivors and nonsurvivors of high-risk surgery. Crit Care Med 1993; 21(7):977–1090. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest 1988; 94(6):1176–1186. Bishop MH, Shoemaker WC, Appel PL, et al. Prospective, randomized trial of survivor values of cardiac index, oxygen delivery, and oxygen consumption as resuscitation endpoints in severe trauma. J Trauma 1995; 38(5):780–787. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high risk surgical patients. JAMA 1993; 270(22):2699–2707. Hayes MA, Timmins AC, Yau EHS, Palazzo M, Hinds CJ, Watson D. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med 1994; 330(24):1717–1722. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-oriented hemodynamic therapy in critically ill patients. SVO2 Collaborative Group. N Engl J Med 1995; 333(16):1025–1032. Tuchschmidt J, Fried J, Astiz M, Rackow E. Elevation of cardiac output and oxygen delivery improves outcome in septic shock. Chest 1992; 102(1):216–220. Ziegler EJ, McCutchan JA, Fierer J, et al. Treatment of Gram-negative bacteremia and shock with human antiserum to a mutant Escherichia coli. N Engl J Med 1982; 307(20):1225–1230.

Chapter 5: Hypovolemic and Septic Shock

54. Calandra T, Glauser MP, Schellekens J, Verhoef J. Treatment of Gram-negative septic shock with human IgG antibody to Escherichia coli J5: a prospective, doubleblind, randomized trial. J Infect Dis 1988; 158(2):312–318. 55. Ziegler EJ, Fisher CJ Jr, Sprung CL, et al. Treatment of Gram-negative bacteremia and septic shock with HA-1A human monoclonal antibody against endotoxin. N Engl J Med 1991; 324(7):429–436. 56. Abraham E, Anzueto A, Gutierrez G, et al. Doubleblind, randomized, controlled trial of monoclonal antibody to human tumor necrosis factor in treatment of septic shock. NORASEPT II Study Group. Lancet 1998; 351(9107):929–933. 57. Abraham E, Wunderink R, Silverman H, et al. Efficacy and safety of monoclonal antibody to human tumor necrosis factor-alpha in patients with sepsis syndrome: a randomized, controlled, double-blind multicenter clinical trial. TNF-alpha MAb Sepsis Study Group. JAMA 1995; 273(12):934–941. 58. Cohen J, Carlet J. INTERSEPT: an international, multicenter, placebo-controlled trial of monoclonal antibody to human tumor necrosis factor-alpha in patients with sepsis. International Sepsis Trial Study Group. Crit Care Med 1996; 24(9):1431–1440. 59. Abraham E, Glauser MP, Butler T, et al. p55 tumor necrosis factor receptor fusion protein in the treatment of patients with severe sepsis and septic shock. JAMA 1977; 277(19):1531–1538. 60. Opal SM, Fisher CJ Jr, Khainaut Jf, et al. Confirmatory interleukin-1 receptor antagonist trial in severe sepsis: a phase III, randomized, double-blind, placebocontrolled, multicenter trial. The Interleukin-1 Receptor Antagonist Sepsis Investigator Group. Crit Care Med 1997; 25(7):1115–1124. 61. Abraham E. Tissue factor inhibition and clinical trial results of tissue factor pathway inhibitor in sepsis. Crit Care Med 2000; 28(suppl 9):S31–S33. 62. De Jonge E, Dekkers PE, Creasey AA, et al. Tissue factor pathway inhibitor does not influence inflammatory pathways during human endotoxemia. J Infect Dis 2001; 183(12):1815–1818.

65

63. Gruber A, Mori E, del Zoppo GJ, Waxman L, Griffin JH. Alteration of fibrin network by activated protein C. Blood 1994; 83(9):2541–2548. 64. De Fouw NJ, van Hinsbergh VW, de Jong YF, Haverkate F, Bertina RM. The interaction of activated protein C and thrombin with the plasminogen activator inhibitor released from human endothelial cells. Thromb Haemost 1987; 57(2):176–182. 65. Esmon CT. The roles of protein C and thrombomodulin in the regulation of blood coagulation. J Biol Chem 1989; 264(9):4743–4746. 66. Esmon CT, Taylor FB Jr, Snow TR. Inflammation and coagulation: linked processes potentially regulated through a common pathway mediated by protein C. Thromb Haemost 1991; 66(1):160–165. 67. Stouthard JM, Levi M, Hack CE, et al. Interleukin-6 stimulates coagulation, not fibrinolysis, in humans. Thromb Haemost 1996; 76(5):738–742. 68. Boehme MW, Deng Y, Raeth U, et al. Release of thrombomodulin from endothelial cells by concerted action of TNF-alpha and neutrophils: in vivo and in vitro studies. Immunology 1996; 87(1):134–140. 69. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344(10): 699–709. 70. Citero G, Vascotto E, Villa F, Celotti S, Pesenti A. Induced abdominal compartment syndrome increases intracranial pressure in neurotrauma patients: a prospective study. Crit Care Med 2001; 29(7):1466–1471. 71. Hong JJ, Cohn SM, Perez JM, Dolich MO, Brown M, McKenney MG. Incidence and outcome of intraabdominal hypertension and the abdominal compartment syndrome. Br J Surg 2002; 89(5):591–596. 72. Tremblay LN, Feliciano DV, Schmidt J, et al. Skin only or silo closure in the critically ill patient with an open abdomen. Am J Surg 2001; 182(6):670–675. 73. Annane D, Se´bille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002; 288(7):862–871.

6 Complications Associated with the Use of Invasive Devices in the Intensive Care Unit Victor Cruz Department of Surgery, Stony Brook University Hospital, Stony Brook, New York, U.S.A. J. Martin Perez Robert Wood Johnson University Hospital, New Brunswick, New Jersey, U.S.A.

Venous Air Embolus The ability of the practitioner to care for the critically ill patient is closely associated with the types and functions of the many invasive and noninvasive devices that can be used. Because the use of central venous catheterization in the intensive care unit (ICU) permits more intensive monitoring and directed interventions, improvements have been achieved in hemodynamic and metabolic assessment, administration of total parental nutrition, chemotherapies, and the performance of hemodialysis. Despite the low incidence of complications associated with these invasive and noninvasive devices, if mortality and morbidity rates are to be reduced even further, the practitioner must be aware of the many complications that may accompany the use of each type of device. Knowledge of these complications will enable the practitioner to not only implement preventive measures but also rapidly recognize and treat complications that may arise. This chapter will discuss the preventive measures and the treatment of complications associated with the use of central venous lines, arterial lines, intracerebral pressure monitors, gastric tubes, and thoracostomy tubes.

CENTRAL VENOUS ACCESS The increasing complexity of the ICU has paralleled the development and increased use of central venous catheterization. Central venous access has become a mainstay in the ICU; several million devices are used annually (1). Unlike the catheters introduced by Broviac et al. (2) and Hickman et al. (3), the catheters used in the ICU lack cuffs and do not require tunneling. Thus, they can be easily placed at the bedside with the Seldinger method or other percutaneous techniques. Although central venous lines are technically less challenging to use and care for than are the Broviac et al. (2) and Hickman et al. (3) catheters, central venous lines are associated with their own set of complications.

The complication of venous air embolism is rare; in fact, the occasional case report represents the majority of the literature about this entity. However, this complication should be suspected if a patient becomes dyspneic on insertion of a central venous line. Venous air embolism can lead to hypotension, acute pulmonary edema, and cardiac arrest. If the foramen ovale is patent, the risk of an ischemic stroke also exists. During the physical examination of a patient with a venous air embolism, a murmur with a characteristic mill wheel may be heard over the right side of the heart. Because a small pressure gradient of 4 mmHg can cause enough air to enter in one second to cause a fatal air embolus (4), prevention techniques should include attempts to increase intrathoracic pressure. Such an increase can be accomplished by placing the patient in Trendelenberg position or by asking the patient to perform a Valsalva maneuver or to hum during placement of a central venous line. Intrathoracic pressure can also increase during exchange over a guide wire. Once venous air embolism occurs, the patient should be placed in the left lateral decubitus position; a syringe should be used to aspirate fluid, air, or both from the line; and a pericardiocentesis can be performed, if necessary, by inserting a needle into the right ventricle in an attempt to aspirate air.

Pneumothorax A pneumothorax from central venous lines occurs when the needle injures lung parenchyma and air escapes into the pleural space. This type of complication occurs in approximately 1% to 4% of all central venous line attempts. Symptoms such as coughing, wheezing, chest pain, and dyspnea may be evident; however, in approximately 0.5% of cases, the appearance of symptoms may be delayed (5). Regardless of when the symptoms appear, a pneumothorax may

68

Cruz and Perez

develop into a tension pneumothorax, and patients who are on a ventilator may be at increased risk (5). In a study of patients with cancer for whom central venous access was established through the subclavian approach, older patients with a body mass index of less than 19 were more likely to experience pneumothorax (6). Pneumothoraces may occur with the same frequency whether the subclavian approach is used or the internal jugular approach is used (7). A relatively high rate of pneumothorax development is also associated with a difficulty in obtaining central venous access, such as repeated attempts at cannulation (6). The diagnosis of pneumothorax is facilitated by an expiratory chest radiograph of the upright patient; however, most chest X rays taken in the ICU are performed with the patient in the supine position. When patients are supine, air usually presents along the lung base and mediastinum, and this makes the diagnosis of a pneumothorax by X-ray relatively more difficult. The risk of a pneumothorax can be reduced if the patient lies on a rolled towel placed under the thoracic spine and between the scapulae during cannulation of the vessel. If the risk of a pneumothorax is to be eliminated, a cut-down of the cephalic vein can be performed. Rates of pneumothorax are lower if experienced physicians perform the central venous access procedure (7). Management of this complication may include observation if pneumothoraces are less than 30% and placement of a pigtail catheter or chest tube if the pneumothorax is larger than 30% or expands after initial observation, or if the patient is on a ventilator.

Hemorrhage and Hemothorax Hemorrhage resulting from central line catheterization can be categorized as either localized or regional. Localized hemorrhage is confined to the site of access, whereas regional hemorrhage occurs in the soft tissue of the neck or extends to the thoracic and mediastinal spaces. Localized hemorrhage due to central line access is uncommon, even if coagulopathy or thrombocytopenia is present (8). Carotid artery puncture, a complication that occurs during approximately 2% to 10% of attempts at internal jugular line placement (9), usually manifests itself as a hematoma when the needle is removed from the artery. Insertion of a needle into the carotid artery may be indicated by bright red, pulsatile blood that fills the hub of the syringe. Removing the needle and applying pressure usually suffices in arresting the bleeding. Close follow-up of the patient who has experienced carotid artery puncture is prudent because acute airway obstruction has been reported from a large cervical hematoma (10). Complications have also resulted from cannulation of the carotid artery (11); these complications include hematoma, arteriovenous fistula, stroke, and death (12). If cannulation of the carotid artery is suspected but the chest X-ray does not reveal it or is inconclusive, an arterial

blood gas can be obtained or catheter pressures can be transduced to confirm a venous wave form. Hemorrhage that results in a hemothorax is not commonly reported. A hemothorax can be caused by an injury to the subclavian artery or vein at the time of insertion or by the gradual erosion of the superior vena cava (SVC). A hemothorax can occur ipsilaterally or contralaterally to the insertion site. Reports of hemothorax caused by the internal jugular approach are rare. Diagnosis of hemothorax can be based on the results of plain-film radiography. However, a substantial hemothorax can cause tachycardia and hypotension. Hemothorax resulting from erosion of the SVC can be prevented by careful placement of the catheter tip so that it does not press against the SVC when placed from left side (13). Once the SVC has been injured, conservative measures are recommended for maintaining the volume and treating any coagulopathy. However, surgical intervention is usually required for both adults (14) and children (15). Bleeding into the mediastinum can occur when a vein is injured or the catheter penetrates the mediastinum. This complication most commonly appears initially on chest radiographs as a widened mediastinum, but it can also cause chest pain after line insertion. When the mediastinum is widened after the placement of a central line, additional chest radiographs should be obtained with the injection of contrast material through the central catheter. If extravasation of the contrast agent occurs and the catheter is in the mediastinum, it should be withdrawn quickly; however, if the catheter is in the vein, it can be left in place, with careful observation of the patient’s condition. According to Whitman (14), most mediastinal hematomas are self-limiting, and the venous injury will resolve without intervention. This is unlike a hemorrhage into the pleural space, which usually requires further intervention. Mediastinal hematomas occur in less than 1% of patients who undergo line placement (14). The unrecognized presence of a catheter in the mediastinum, regardless of the type of fluid being administered, is associated with high morbidity and mortality rates (16).

Cardiac Tamponade According to the 1989 Food and Drug Administration (FDA) drug bulletin, cardiac tamponade is the most commonly reported lethal complication associated with central venous access. Cardiac tamponade occurs when the catheter tip penetrates the pericardium. Cardiac tamponade is signaled by an acute onset of tachycardia, hypotension, jugular venous distention, and pulsus paradoxus. However, the symptoms of tamponade may also be delayed. Cardiac tamponade can be prevented by careful placement of the catheter tip. The FDA has stated that placing the tip into the atrium is associated with a higher risk of tamponade. There have also been

Chapter 6: Complications Associated with the Use of Invasive Devices in the Intensive Care Unit

reports of cardiac tamponade when the catheter tip is placed in the SVC (17). Cardiac tamponade is treated by rapidly increasing intravenous volume followed by subxiphoid pericardiocentesis to stabilize the patient’s condition. Surgical intervention may be warranted to repair lacerations of the atria or the SVC, as well as the need for a pericardial window as a more definitive measure.

Central Venous Device and Infection Central venous access devices are a common source for bacterial nosocomial infections in the ICU setting. In fact, catheter-related bloodstream infection (CRBSI) is the most frequently occurring type of bloodstream infection in the ICU (18,19). Approximately 250,000 patients develop CRBSI each year (18). According to one study (20), nosocomial bloodstream infection occurs in 2.7 of every 100 patients admitted to a surgical ICU. A case control analysis found that such infections are associated with a 35% increase in mortality, a 24-day increase in median length of stay, and a $40,000 increase in expense per survivor. The National Nosocomial Infections Surveillance (NNIS) System reported that a median range of bloodstream infections associated with central lines was 2.4 to 7.8 central line–patient days (21). The pathogenesis of CRBSI involves colonization of the catheter by microorganisms that inhabit the skin surface at the site of catheter insertion (18). Other important sources of catheter colonization are contaminated infusates, distant sites of infection, and hub contamination. Microorganisms migrate along the transcutaneous tract and adhere to the catheter’s biofilm. The exact sequence of events leading from biofilm colonization to bloodstream infection is poorly defined, as are the factors that determine establishment of infection. According to the NNIS, the most commonly reported organisms involved in CRBSI, in decreasing order of frequency, are coagulase-negative Staphylococcus spp. (39.3%), Staphylococcus aureus (10.7%), Enterococcus spp. (10.3%), Candida albicans (4.9%), and other gram-negative organisms. The Centers for Disease Control and Prevention define catheter colonization as growth of 15 or more colony-forming units isolated from a catheter segment and cultured by the roll-plate method. CRBSI is defined as an infection in which the same organism with a similar drug susceptibility pattern is isolated from a catheter segment and simultaneously from peripheral blood in a patient with clinical manifestations of sepsis and no other apparent source of blood stream infection (22). If signs of local infection, such as erythema, induration, tenderness, or purulent drainage, develop at the site of catheter insertion, the catheter should be removed and the catheter segment submitted for culture. If CRBSI is established, in addition to catheter removal, systemic antibiotics are usually indicated. Currently, a lack of consensus exists regarding antibiotic selection and duration of therapy;

69

however, with the increasing emergence of methicillinresistant strains, vancomycin is commonly used as firstline therapy. Most clinicians will develop a treatment plan according to the isolate, known susceptibility patterns, and response to therapy; this response is indicated by clinical variables such as fever and leukocytosis. Persistent fever, leukocytosis, and bacteremia may warrant further investigation of a deep-seated infection such as suppurative thrombophlebitis. In patients who are febrile but have no obvious clinical source of infection and no clinical signs of infection at the site of catheter insertion, the catheter segment may be submitted for quantitative culture. If the culture of the catheter segment demonstrates significant microbial growth, the replacement catheter should be removed and a new catheter should be inserted at a different site. Multiple methods have been studied for preventing CRBSI. Ruesch et al. (23) have reviewed several prospective studies demonstrating that the subclavian site of insertion is associated with a lower risk of infection than is the jugular or femoral sites. Another review (24) has shown that chlorhexidine gluconate is superior to 10% povidone iodone for skin preparation in the prevention of CRBSI. A prospective observational study (25) showed that full-barrier precautions, including sterilized gloves, gown, cap, mask, and barrier drape that covers the entire patient, reduces the incidence of CRBSI associated with the insertion of central venous devices. Several studies have demonstrated that the use of transparent occlusive dressings increases the risk of clinically significant catheter colonization. The use of antibiotic ointments at the catheter insertion site appears to increase catheter colonization with fungi. Using chlorhexidine-impregnated sponges as a dressing for central venous, pulmonary artery, and arterial catheters have been shown to significantly decrease catheter colonization and CRBSI (26). Minocycline- or rifampin-impregnated catheters are associated with lower rates of catheter colonization and CRBSI than are catheters impregnated with chlorhexidine/silver sulfadiazine (27). Patients should be asked about potential allergic reactions to the components of the catheter before it is placed. The emergence of resistant organisms in response to the use of antibiotic-impregnated catheter may have an impact on future clinical use. Preventative measures that have been proven to reduce catheter colonization and CRBSI should be implemented and adhered to as a part of an overall effort to reduce nosocomial infection rates associated with the use of central venous access devices.

Thrombosis Central venous thrombosis after placement of a central line appears to be related to several factors. Larger catheters, which are more likely to obstruct flow, have been shown to be associated with higher thrombosis rates (28). Moreover, the location of the catheter tip

70

Cruz and Perez

may be a factor. An analysis of several retrospective studies has shown a correlation between placement of catheter tips high in the SVC and higher rates of thrombosis (29). Patients with bone marrow transplants, malignancies, sickle cell disease, and renal failure are more likely to experience thrombosis (30). Other factors that play a role in thrombosis are hypercoagulable states secondary to illness and endothelial injury from catheter insertion. Venous thrombosis can be diagnosed by duplex ultrasonography. However, duplex ultrasound is unable to adequately visualize the subclavian vein deep to the clavicle. This limitation is partially overcome by the fact that the presence of subclavian thrombosis can be predicted by a lack of flow in the internal jugular or axillary vein. This observation of a lack of flow may be adequate for the diagnosis of thrombosis. However, if a patient exhibits symptoms such as superficial venous engorgement, unilateral arm swelling, pain, and discoloration, contrast venography should be performed even if the findings of venous duplex ultrasonography are negative. Management of thrombosis depends on the patient’s symptoms and on whether the catheter is functioning properly. If a patient has no symptoms, a functional catheter can safely be left in place while anticoagulation therapy is started to prevent clot propagation, or to prevent the formation of a pulmonary embolism from an upper-extremity thrombosis (the risk of such an embolism is less than 10%) (31). If the catheter is clotted or if the patient’s symptoms worsen, the catheter should be removed. Prevention of thrombosis should take into account the location of the catheter, with the optimal placement being just proximal to the right atrium. Anticoagulation has also been useful in preventing the formation of thrombosis. Both heparin (32) and low doses of Coumadin1 (warfarin sodium; 1 mg/day) (33) have been shown to effectively reduce the rate of thrombosis formation. However, no studies have compared the effectiveness of heparin with that of Coumadin in preventing thrombosis. The incidence of complications such as fracture of the catheter, loss of the guide wire, and arteriovenous fistula is unknown. However, when fracture or loss of the guide wire has occurred, the catheter can be safely removed via interventional radiologic techniques (34).

ARTERIAL LINES The radial, brachial, dorsal pedal, and femoral arteries can be safely cannulated for invasive blood pressure monitoring and frequent blood draws. The complications commonly associated with these catheters are those associated with central venous lines—infection, bleeding, and thrombosis. The most common complication associated with these catheters, however, is not infection, as is the case with central venous lines. In a

study of more than 2000 patients (35), the most common complications were vascular insufficiency (4%), bleeding (2%), and infection (0.5%). Arterial thrombosis due to catheterization can occur in as many as 25% of patients with radial arterial lines, but the occurrence of this complication is lower among patients with femoral arterial lines (36). Fortunately, thrombosis of the radial or femoral artery rarely causes ischemia to the distal extremity. Flushing the catheter with heparin has been shown to decrease the rate of thrombosis in both femoral and radial arterial lines (37). The incidence of infection is similar with femoral and radial arterial lines. Staphylococcal species continue to be the most common organism associated with catheter-related infections. Although many ICUs routinely change arterial line tubing and solution in an attempt to decrease infection rates, O’Malley and colleagues (38) demonstrated in a prospective study that routinely changing the tubing increases the likelihood of introducing contamination into the pressuremonitoring system. Preventing arterial catheter infections requires methods similar to those used to prevent infection of central venous lines. Sterile techniques should be used during catheter placement, and hygienic manipulation should be employed when the line is accessed.

GASTROSTOMY TUBES The popularity of gastrostomy tubes has increased over the years. Although their use for gastrointestinal decompression has been widely accepted for some time, it was not until the late 1970s that their usefulness in providing enteral access for nutritional support was appreciated. Kudsk et al. (39) first demonstrated the benefits of enteral nutrition over parenteral nutrition in rats. Alexander (40) reported that the outcome of children with burns to 60% of their total body surface area was improved when enteral nutrition rather than parenteral nutrition was used. Several subsequent studies continued to report the benefits of enteral nutrition over parenteral nutrition in reducing the rates of nosocomial infection. Although gastrointestinal intubation has improved clinical outcome by providing enteral nutrition and gastric decompression, the placement of nasogastric tubes (NGTs) has it own set of complications— aspiration pneumonia, esophageal perforation, sinusitis, malposition, and arterial esophageal fistula.

Aspiration Pneumonia The incidence of aspiration pneumonia increases among patients with certain risk factors (Table 1). Mullan and Roubenoff (41) have stated that the incidence of aspiration ranges from 1% to 30%. This variation is most likely due to the variety of patient populations included and the different methods used in the diagnosis of aspiration pneumonia. In one

Chapter 6: Complications Associated with the Use of Invasive Devices in the Intensive Care Unit

Table 1 Risk Factors for Aspiration Pneumonia Delayed gastric emptying Decreased gag, cough, or swallow reflexes Mechanical ventilation Large bore feeding tubes Neurologic injury

study, the incidence of aspiration among patients receiving enteral nutrition was approximately 5%, and mortality from aspiration in this population was less than 5%. A retrospective analysis using national medical claims has reported the mortality of aspiration pneumonia to be as high as 23.9%. Mortality is correlated with the amount of aspiration, acidity of aspirate, number of lobes involved, and the overall condition of the patient. A dramatic change in the appearance of the chest on radiograph can suggest the diagnosis of aspiration pneumonia; however, such changes may not be evident until 24 hours after aspiration has occurred. The diagnosis of aspiration may be aided by testing the pulmonary aspirate for glucose. Previously, Food Drug and Cosmetic Blue No. 1 dye was added to enteral nutrition formulas in order to facilitate the detection of gastric aspirate in tracheal secretions. However, reports (42) of systemic blue dye absorption and associated adverse outcomes are emerging. This has caused many hospitals to withdraw the practice of adding dye to their enteral formulas. Bronchoscopy is a helpful tool that can be used to diagnose aspiration in addition to providing immediate lavage for removing aspirate. Aspiration can be prevented by correctly placing the NGT and closely monitoring gastric residuals. Other preventive measures include elevating the head of a patient’s bed to an angle of 30 to 45 so the patient is not supine. Using small-bore NGTs has also been suggested as a means of decreasing aspiration pneumonia and reflux. Conversely, the incidence of pneumothorax and tracheal intubation with insertion of a feeding tube is increased with small-bore nasoenteral tubes (less than 5 mm diameter). Aspiration should be treated by removing the aspirate with nasotracheal suction or bronchoscopy when necessary. Antibiotics, intubation, and ventilator support may be indicated. Other treatments such as corticosteroids therapy have not proven to be beneficial.

Esophageal Perforation Esophageal perforation is a rare complication of nasogastric intubation; however, its occurrence may be fatal. One study of hospitalized patients with esophageal perforation showed that the most common cause of this complication is iatrogenic (43). The most common site of iatrogenic perforation is the thoracic esophagus. The mortality rates associated with esophageal perforation range from 16% to 30% (44), but the mortality rate is lower when

71

the cause of perforation is iatrogenic or when the injury is promptly recognized and treatment is not delayed. The risk of esophageal perforation is higher among patients with carcinoma, stricture, or altered mental status, as well as among those who have undergone tracheal intubation or have undergone multiple attempts at nasogastric intubation (45). In the critically ill, the diagnosis of esophageal perforation may be difficult and requires a high level of clinical suspicion. The most common clinical features are neck and substernal pain, fever, and subcutaneous or mediastinal emphysema that may cause nuchal crepitus or xiphisternal crepitus (also known as Hamman’s sign). Radiography of the chest or abdomen with the patient in the upright position is diagnostic in most cases. However, when the results are negative and the level of clinical suspicion remains high, esophageal study using Gastrografin1 as a contrast agent should be performed. The prognosis of esophageal perforation is dependent upon early diagnosis and treatment and the site and size of the perforation. Medical management includes the use of broad-spectrum antibiotics and nasogastric decompression. These treatments may be adequate for patients with cervical perforation, who are asymptomatic and whose condition is hemodynamically stable (44). Surgical management, if necessary, involves drainage alone, drainage and repair, or drainage and diversion.

Sinusitis The use of a NGT with tracheal intubation has been associated with an increase risk of sinusitis. In these cases, sinusitis occurs because the NGT obstructs the ostial meatal complex and impairs the drainage of mucous. According to Fasqualle et al. (46), nosocomial sinusitis occurs most commonly in the ICU. The most commonly involved organism is Pseudomonas aeruginosa, Streptococcal pneumonia, and Hemophilus influenza. Pain and pressure over the cheeks are the most common symptoms of sinusitis, but may be difficult to assess among ICU patients. Purulent nasal discharge may not be present, but if it is, the likelihood of sinusitis is increased. Often patients initially present with fever of unknown etiology. Once the most common causes of fever have been ruled out, further evaluation of the sinuses is warranted if intubation and an NGT are in use. Westergren et al. (47) have advocated the use of ultrasonography as a sensitive test for the presence of fluid and edema in the sinuses of critically ill patients. For improved diagnostic accuracy, computer tomography of the sinuses (coronal view) can be performed. Sinoscopy can aid in the diagnosis by yielding a culture specimen that can be tested for the purpose of directing antibiotic therapy; sinoscopy can also be used to treat sinusitis by creating an ostium to allow drainage of sinus secretions. Treatment consists of antibiotic therapy targeted at

72

Cruz and Perez

specific bacteria, removal of any foreign objects from the nose, using nasal decongestants, and if necessary, sinoscopy drainage.

THORACOSTOMY TUBES

Malposition Retrospective case review of complications associated with NGTs have indicated that approximately 0.5% to 1% of NGTs are malpositioned (48). The most common location of malpositioning is the tracheal bronchial tree, especially the bronchi of the right lower lobe; misplacement in the pleural space, intracranial space, and internal jugular vein is rarely reported. The types of complication of malpositioned NGTs are based on location (Table 2). NGTs are contraindicated for patients with basal skull injuries because such injuries increase the likelihood of intracerebral placement of the tube. However, this type of malpositioning has also been observed in patients without basal skull fractures (49). Diagnosis of malpositioning has been based on several commonly acceptable clinical guidelines. Proper placement of the NGT can be performed by insufflation of air in the tube with auscultation over the stomach or by aspiration of gastric contents. Aspiration prior to insufflation is preferred, in order to prevent a fatal air embolus, in the rare case of NGT malposition in a vascular structure. However, malpositioning of the NGT in the left lung base can also be detected by auscultation over the stomach, and secretions can also be suctioned from the bronchial tree, the esophagus and stomach, or the brain. Bankier et al. (48) have demonstrated that the clinical signs of NGT malpositioning may not accurately indicate the tube’s location. A chest radiograph taken with the patient in the supine position can accurately detect malpositioning. Prevention lies in the use of clinical guidelines described above and of chest radiographs when there is any doubt about the position of the NGT.

Arterial Esophageal Fistula Arterial esophageal fistula is a rare complication of gastric esophageal intubation and can involve the aorta and other great vessels in the chest. Symptoms are similar to those of aortic enteric fistulas. Sentinel bleeding, which first alerts the practitioner to the presence of this complication, is followed by a symptomfree period that precedes exsanguinations. Reports Table 2 Nasogastric Tube Malpositioning and Associated Complications Location of malposition Tracheobronchial tree Pleural space Internal jugular vein Intracranial

(50) suggest that anomalies of the aortic arch predispose patients to this complication.

Complications Loss of tidal volume Pneumonia Pneumothorax Tension pneumothorax Hypotension Anemia Death

Thoracostomy tubes have been used since the time of Hippocrates but were not popularized until the Korean War. Indications for chest tube placement are pneumothorax, tension pneumothorax, penetrating chest injury, hemothorax, empyema, chylothorax, post–thoracic surgery, and bronchopleural fistula. Thoracostomy tubes can be placed via sharp cut down, by the trochar method (the use of a sharp metal rod), or by a percutaneous Seldinger techniques. The percutaneous technique using the Seldinger method is employed when smaller pigtail-type chest tubes are used. Complications associated with chest tubes can be separated into three categories (Table 3).

Empyema Empyema occurs in 1% to 16% of patients; the higher incidence occurs among trauma patients (51). Chest tubes placed for pleural effusions are associated with higher rates of empyema. This complication is indicated by purulent exudative fluid with a low pH, low glucose concentrations, and a high white blood cell count. Empyema is treated with intravenously administered antibiotics, chest tube drainage, and, possibly, open thoracotomy. Prevention of empyema begins with the use of appropriate sterile technique. The use of antibiotics remains controversial, with early studies demonstrating no benefit or only minimal benefit in association

Table 3 Thoracostomy Tube Complications Infectious Chest tube site wound infection Tracheitisa Pneumonia Empyema Anatomic Subcutaneous emphysema Pneumothorax Residual Iatrogenic Hemothorax/pleural effusion Residual Iatrogenic Arteriovenous fistulaa Malposition Subcutaneous Intrathoracic With or without visceral injurya Intra-abdominal With or without visceral injurya Physiologic Re-expansion pulmonary edema Myocardial ischemiaa Horner’s syndromea a

Rare complications only illustrated by the rare case report.

Chapter 6: Complications Associated with the Use of Invasive Devices in the Intensive Care Unit

with their use. However, more recent studies and a meta-analysis by Evan et al. (52) of six prospective, randomized studies found that a beneficial effect was associated with the use of antibiotics effective against Staphylococcus spp.

Pneumothorax and Hemothorax Pneumothorax and hemothorax are the most common indications for chest tube placement; however, these complications may also occur in association with thoracostomy tube placement and use. Thoracostomy tube placement may fail as an effective treatment for these conditions, and pneumothorax and hemothorax may recur after the tube has been removed or may be an iatrogenic sequela of removal. A study of chest radiographs after thoracostomy tube insertion demonstrated that the most common complications were tube malposition and unresolved pneumothorax and hemothorax (53). Recurrence of a pneumothorax after thoracostomy tube removal is most likely caused by the entry of air through the wound or by reaccumulation from a small air leak. To prevent such pneumothoraces, the thoracostomy tube should be pulled out quickly while the patient is exhaling, so that intrapleural pressure is positive, thus decreasing the risk of air entry. When placing chest tubes, some practitioners place a second stitch through the site of chest tube insertion; this stitch is left untied. When the chest tube is removed, a second person places tension on this stitch to close the incision as the chest tube is being removed. To decrease the likelihood of a recurrent pneumothorax caused by an undetected small air leak, it is best to convert the tube to water seal before the tube is removed. In a randomized trial, Martino et al. demonstrated that converting thoracostomy tubes to water seal before removing them reduces the number of pneumothoraces that occur after tube removal and also reduces the need for chest tube replacement (54). There have also been reports of ipsilateral or contralateral pneumothoraces in association with the placement of chest tubes for effusion (55). Pneumothorax associated with thoracostomy tube placement can be treated in several ways. The pneumothorax can be carefully observed and may resolve even if it is not adequately drained by the tube. Increasing the suction of the thoracostomy tube, by increasing the column of fluid in a wet hemovac system or simply turning the pressure dial on a dry hemovac system, can resolve a persistent pneumothorax. A second thoracostomy tube can also be placed to resolve the pneumothorax or hemothorax. Pleurodesis may be necessary to resolve a persistent leak or effusion, but a thoracotomy is rarely necessary for adequate resolution of a pneumothorax.

73

was the reason for as many as 26% of inadequate tube placements and was the most common complication (53). Malpositioned chest tubes can be located in the subcutaneous tissue of the chest wall or in incorrect locations in the intrathoracic and intra-abdominal regions. A malpositioned thoracostomy tube may cause injury to the visceral organs and may even result in death. One study (53) suggests that chest tube placement by clinicians other than surgeons is associated with a higher rate of complications, but another study (56) refuted this finding. The use of trochars for insertion is also associated with a high rate of complications; thus, this method of tube insertion has largely been replaced by blunt dissection.

Re-expansion Pulmonary Edema Re-expansion pulmonary edema (RPE) can occur after pulmonary re-expansion by thoracostomy tube for pneumothorax, pleural effusion, or atelectasis. One retrospective review of the placement of thoracostomy tubes for pneumothorax in Japan (57) suggested that RPE occurs in approximately 14% of cases. However, in the United States, RPE is generally considered a rare complication; the mortality rate associated with RPE may be as high as 20% (58). Although the cause of RPE is unknown, certain risk factors such as young age, a large pneumothorax, and long duration of collapse are associated with RPE (59). Some patients with RPE may exhibit no symptoms other than radiographic findings; others may experience severe tachypnea, tachycardia, hypoxemia, or chest pain. The two most common symptoms are dyspnea and chest pain, which usually occur within minutes to hours of re-expansion. Diagnosis of RPE is based on chest radiography, which shows pulmonary edema in a previously collapsed lung. Cases of contralateral RPE have also been reported (58). Treatment involves supportive care, which may include hemodynamic and ventilatory support; most cases are self-limiting and resolve within a week.

INTRACRANIAL PRESSURE MONITORING Intracranial pressure (ICP) monitoring plays a key role in the management of increased ICP. ICP monitoring is frequently used for patients with brain injuries and those undergoing elective neurosurgery. Complications associated with ICP monitoring include infection and hemorrhage, as well as malfunction, obstruction, and malpositioning of the tube. Longterm morbidity and mortality associated with complications of ICP monitoring appear to be rare.

Infection Malpositioning In one radiologic evaluation of thoracostomy tubes placed in the emergency room, tube malpositioning

Infection related to ICP monitoring is characterized by positive results from either a culture of cerebrospinal fluid or a culture of material on the intracranial portion

74

Cruz and Perez

of the catheter, along with clinical signs of infection. No semiquantitative or quantitative methods exist for distinguishing infections of the ICP-monitoring device from colonization of the device, and clinical signs of infection are needed to establish the diagnosis of infection. Colonization of ICP monitors has been associated with either implantation for more than five days or irrigation of fluid-coupled devices. In one study, the incidence of bacterial colonization increases from 6% to 19% when irrigation was employed (60). Differences in the rates of colonization of ICP devices appear to be related to the type of device used. Parenchymal devices are associated with the highest average rates of colonization (14%; range 11.7–16.6%); the other types of devices associated with colonization are subdural devices (4%; range 1–10%), subarachnoid devices (5%; range, 0–10%), and ventricular devices (5%; range, 0–9.5%) (61). For all types of devices, the rates of colonization increased as time of implantation increased. A retrospective cohort study (62) yielded similar results in a pediatric subgroup of patients; the infection rate associated with Camino1 fiberoptic monitors was 0.3%. There is no evidence-based consensus regarding antibiotic prophylaxis for various classes of devices or for duration of implantation.

Hemorrhage Hemorrhage associated with ICP monitoring has not been clearly defined in many reports but appears to occur infrequently; the overall incidence of hematoma is 1.4%. In two studies, substantial hematomas requiring evacuation occurred in 0.5% of patients who required ICP monitoring (63,64).

Malfunctioning and Malpositioning of Intracranial Pressure Monitors Malfunctioning and displacement of ICP monitors are the most frequently reported complications associated with the use of these devices. Malfunction or obstruction has been observed in association with 6.3% of fluid-coupled ventricular devices, 16% of subarachnoid bolts, and 10.5% of subdural catheters (65,66). In a pediatric population, Camino fiberoptic monitors malfunctioned in 2.6% of cases and were displaced in 1% (62). The implications of malfunction and displacement include inaccurate ICP readings, potential morbidity associated with reinsertion, and additional cost.

CONCLUSION Invasive devices, such as lines and tubes, have increased the ability of the practitioner to care for critically ill patients in the ICU. However, although such devices can be invaluable tools in caring for patients, their use is also associated with significant morbidity and mortality. Understanding the potential pitfalls associated with these devices will increase the clinician’s ability to diagnose and treat such complications

and to implement strategies that may prevent the occurrence of such complications in the future.

REFERENCES 1. Mansfield PF, Hohn DC, Fornage BD, Gregurich MA, Ota DM. Complication and failures of subclavian-vein catheterization. N Eng J Med 1994; 331:1735–1738. 2. Broviac JW, Cole JJ, Scibner BH. A silicone rubber atrial catheter for prolonged parenteral alimentation. Surg Gynecol Obstet 1973; 136:602. 3. Hickman RO, Buckner CD, Cliftra, et al. A modified right atrial catheter for access to the venous system in marrow transplant recipients. Surg Gynecol Obstet 1979; 148:871. 4. Sladen A. Complications of invasive hemodynamic monitoring in the intensive care unit. Curr Probl Surg 1988; 25:69–145. 5. Plewa MC, Ledrich D, Sferra JJ. Delayed tension pneumothorax complicating central venous catheterization and positive pressure ventilation. Am J Emerg Med 1995; 13(5):532–535. 6. Harrington KJ, Pandha HS, et al. Risk factors for pneumothorax during percutaneous Hickman line insertion in patients with solid and hematologic tumors. Clin Oncol 1995; 7(6):373–376. 7. Broadwater JR, Henderson MA, Bell JL, et al. Outpatient percutaneous central venous access in cancer patients. Am J Surg 1990; 160:676–680. 8. Foster PF, Moore LR, Sankay HN, et al. Central venous catheterization in patients with coagulopathy. Arch Surg 1992; 127:273–275. 9. Seneff MG. Central venous catheterization: a comprehensive review. Intensive Care Med 1987; 2:163–175, 218–232. 10. Wiemann J, Frass M, Traindl O. Acute upper airway obstruction caused by cervical hematoma as a complication of internal jugular venous catheterization. Acta Med Aust 1990; 17(4):77–79. 11. Silverman JM, Olson KW. Avoiding unintentional arterial cannulation. Crit Care Med 1990; 18(4):460. 12. Heffner JE. A 49 year-old man with tachypnea and a rapidly enlarging pleural effusion. J Crit Illness 1994; 9:101–109. 13. Harrer J, Brtico M, Zacek P, Knap J. Hemothorax—a complication of subclavian vein cannulation. Acta Med 1997; 40(1):21–23. 14. Whitman E. Complication associated with the use of central venous access devices. Curr Probl Surg 1996; 312–317. 15. Bagwell CE, Salzbery AM, Sourrcio RE, Haynes JH. Potentially lethal complication of central venous catheter placement. J Pediatr Surg 2000; 35(5):709–713. 16. Sheep RE, Guiney WB. Fatal cardiac tamponade occurrence with other complications after left internal jugular vein catheterization. JAMA 1982; 248:1632–1635. 17. Karnauchow PN. Cardiac tamponade from central venous catheterization. J Can Med Assoc 1986; 135: 1145–1147. 18. Maki DG. Infection caused by intravascular devices used for infusion therapy: pathogenesis, prevention, and management. Infections Associated with Indwelling Medical Devices. 2nd ed. Washington, DC: ASM Press, 1994:155–205.

Chapter 6: Complications Associated with the Use of Invasive Devices in the Intensive Care Unit

19. Taynes R, Culver D, Emori T, et al. The National Nosocomial Infections Surveillance System: plans for the 1990’s and beyond. Am J Med 1991; 91(3B):116S–120S. 20. Pittet D, Tarara D, Wenzel RP. Nosocomial bloodstream infections in critically ill patients: excess length of stay, extra costs, and attributable mortality. JAMA 1994; 271:598–601. 21. National Nosocomial Infections Surveillance (NNIS) System. December 2000. Available at http://www. cdc.gov/ncidod/dhqp/pdf/nnis/DEC2000sar.PDF. Accessed June 16, 2006. 22. Pearson ML. Hospital Infection Control Advisory Committee. Guidelines for prevention on vascular device-related infections. Am J Infect Control 1996; 23: 262–277. 23. Ruesch S, Walder B, Tramer MR. Complications of Central venous catheters: internal jugular versus subclavian access—a systemic review. Crit Care Med 2002; 30(2):454–460. 24. Chaiyakunapruk N, Veenstra DL, Lipsky BA, Saint S. Chlorhexadine compared with povidone-iodine solution for vascular catheter site care: a meta-analysis. Ann Intern Med 2002; 136(11):792–801. 25. Raad II, Hohn DC, Gilbraith BJ. Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertion. Infect Control Hosp Epidemiol 1994; 15:231–238. 26. Maki DG, Ringer M, Alvarado CJ. Prospective randomized trial of povidone-iodone, alcohol, and chlorhexidine for prevention of infection associated with central venous and arterial catheters. Lancet 1991; 338:339–343. 27. Veenstra DL, Lipsky BA, Saint S. The cost effectiveness of minocycline/rifampin versus chlorhexidine/silver sulfadiazine central venous catheters. 10th Annual Meeting for the Society for Hospital Epidemiology of America, Atlanta, GA, March 5–9, 2000. 28. Puel V, Candry M, LeMetayer P, et al. Superior vena cava thrombosis related to catheter malposition in cancer chemotherapy. Cancer 1993; 72:2248–2252. 29. Dierks MM, Whitman ED. Catheter tip position in the analysis of central venous access device outcome. J Can Intra Nurses Association 1997; 13(3):7–10. 30. Whitman ED. Complications with the use of central venous access devices. Curr Probl Surg 1996; 33(4): 309–378. 31. Becker DM, Philbrick JT, Walker FB. Axillary and subclavian venous thrombosis. Arch Intern Med 1991; 151:1934–1943. 32. Randolph AG, Cook DJ, Gonzalez CA, Andrew M. Benefit of heparin in central venous and pulmonary artery catheters: a meta-analysis of randomized control trials. Chest 1998; 113(1):165–171. 33. Bern MM, Lokich JJ, Wallach SR, Bothe A, Benotti PN. Very low dose of warfarin can prevent thrombosis in central venous catheters. A randomized prospective trial. Ann Intern Med 1990; 112(6):423–428. 34. Coles CE, Whitear WP, Levay JH. Spontaneous fracture and embolization of central venous catheter: prevention and early detection. Clin Oncol 1998; 10(6):412–414. 35. Frezza EE, Mezghebe H. Indications and complications of arterial catheter use in surgical and medical intensive care units. Am Surg 1998; 64(2):127–131. 36. Sfeir R, Khoury S, Khoury G, Rustum J, Ghabash M. Ischemia of the hand after radial artery monitoring. Cardiovasc Surg 1996; 4(4):456–458.

75

37. Zevola DR, Dioso J, Maggio R. Comparison of heparinized and nonheparinized solutions for maintaining patency of arterial and pulmonary artery catheters. Am J Crit Care 1997; 6(1):52–55. 38. Omalley MK, Rhames FS, Cerra FB, McLomb RC. Value of routine pressure monitoring system changes after 72 hours of continuous use. Crit Care Med 1994; 22(9): 1424–1430. 39. Kudsk KA, Stone JM, Carpenter G, Sheldon GF. Effects of enteral versus parenteral feeding of malnourished rats on body composition. Curr Surg 1981; 38(5): 322–323. 40. Alexander JW. Nutrition and infection. New perspective for an old problem. Arch Surg 1986; 121(8): 966–972. 41. Mullan H, Roubenoff RA, Roubenoff R. Risk of pulmonary aspiration among patients receiving enteral nutrition. J Parenteral Enteral Nutr 1992; 16(2): 160–164. 42. Lucarelli MR, Shirk MB, Julian MW, Crouser ED. Toxicity of Food Drug and Cosmetic Blue No. 1 dye in critically ill patients. Chest 2004; 125(2):793–795. 43. Norman EA, Sosis M. Iatrogenic esophageal perforation due to tracheal or nasogastric intubation. J Can Anaesth Soc 1986; 33:222–226. 44. Michael L, Grillo HC, Malt RA. Operative and nonoperative management of esophageal perforations. Ann Surg 1981; 194:57–63. 45. Jackson RH, Payne DK, Bacon BR. Esophageal perforation due to nasogastric intubation. Am J Gastroenterol 1990; 37:439–442. 46. Fasqualle D, Alami M, Dumas G, Fockenier F, Sibille JP. Epidemiology of sinusitis seen in hospitalized patients. Pathologie Biologie 1998; 46(10):751–759. 47. Westergren U, Berg S, Lundreg J. Ultrasonographic bedside evaluation of maxillary sinus disease in mechanically ventilated patients. Intensive Care Med 1997; 23(4):393–398. 48. Bankier AA, Wiesmayr MN, Henk C, et al. Radiographic detection of intrabronchial malpositions of nasogastric tubes and subsequent complication in ICU patients. Intensive Care Med 1997; 23: 406–410. 49. Frei RM, Mullet ST. Inadvertent intracranial insertion of a nasogastric tube in a non trauma patient. Emerg Med J 1997; 14(1):45–47. 50. Minyard AN, Smith DM. Arterial-esophageal fistulae in patients requiring nasogastric esophageal intubation. Am J Forensic Med Pathol 2000; 21(1):74–78. 51. Etoch SW, Bar-Natam MF, Miller FB, Richardson JD. Tube thoracostomy. Factors related to complications. Arch Surg 1995; 130(5):521–525. 52. Evan JT, Green JD, Carlin PE, Barrett LO. Meta-analysis of antibiotics in tube thoracostomy. Am Surg 1995; 61(3):215–219. 53. Baldt MM, Bankier AA, Germann PS, et al. Complications after emergency tube thoracostomy: assessment with CT. Radiology 1995; 195(2):539–543. 54. Martino K, Merrit S, Sernas T, et al. Prospective of randomized trial of thoracostomy removal algorithms. J Trauma Injury Infect Crit Care 1999; 46(3): 369–373. 55. Gerard PS, Kaldarvi E, Litani V, Lenora RA, Tessler S. Right-sided pneumothorax as a result of left-sided chest tube. Chest 1993; 103(5):1602–1603.

76

Cruz and Perez

56. Chan L, Reilly KM, Henderson C, Kahn F, Salluzzo RF. Complication rates of tube thoracostomy. Am J Emerg Med 1997; 15(4):368–370. 57. Matsura Y, Nomimura T, Murakami H, et al. Clinical analysis of reexpansion pulmonary edema. Chest 1991; 100(6):1562–1566. 58. Heller BJ, Grathwohl MK. Contralateral reexpansion pulmonary edema. Southern Med J 2000; 17(3):234. 59. Murat A, Arslan A, Balci AE. Re-expansion pulmonary Edema. Acta Radiol 2004; 45(4):431–433. 60. Aucoin PJ, Kotilainen HR, Gantz NM, et al. Intracranial pressure monitors. Epidemiologic study of risk factors and infections. Am J Med 1986; 80:369–376. 61. Bullock R, Chestnut RM, Clifton G, et al. Guidelines for the Management of Severe Head Injury. New York, NY: Brain Trauma Foundation, 1995.

62. Pople IK, Muhlbauer MS, Sanford RA, et al. Results and complications of intracranial pressure monitoring in 303 children. Pediatr Neurosurg 1995; 23(2): 64–67. 63. Narayan RK, Kishore P RS, Becker DP, et al. Intracranial pressure: to monitor or not to monitor? J Neurosurg 1982; 56:650–659. 64. Paramore CG, Turner DA. Relative risks of ventriculostomy infection and morbidity. Acta Neurochir (Wien) 1994; 127:79–84. 65. Barlow P, Mendelow AD, Lawrence AE, et al. Clinical evaluation of two methods of subdural pressure monitoring. J Neurosurg 1985; 63:578–582. 66. North B, Reilly P. Comparison among three methods of intracranial pressure recording. Neurosurgery 1986; 18:730.

PART II Gastrointestinal Surgery Complications

7 Complications of Abdominal Wall Surgery and Hernia Repair James C. Doherty Division of Trauma Surgery, Advocate Christ Medical Center, Oak Lawn, and Department of Surgery, University of Illinois College of Medicine at Chicago, Chicago, Illinois, U.S.A. Robert W. Bailey Division of Laparoscopic and Bariatric Surgery, Daughtry Family Department of Surgery, Miller School of Medicine, University of Miami, Miami, Florida, U.S.A.

Abdominal wall closure, including ventral and inguinal hernia repair, are among the most frequently performed of all procedures in general surgery. Although the complication rates associated with these procedures are relatively low, the number of procedures performed renders their associated surgical complications among the most common encountered in clinical practice.

postoperative period must raise concern for infection with Clostridium perfringens or B-hemolytic streptococci. Such infections can rapidly progress to life-threatening necrotizing soft-tissue infections requiring aggressive surgical debridement. Fortunately, most postoperative wound infections are uncomplicated infections of the skin, the superficial subcutaneous tissue, or both. Such infections are usually controlled by opening, draining, debriding, and observing the wound. In most cases, neither antibiotics nor wound cultures are necessary.

COMPLICATIONS OF ABDOMINAL WOUND CLOSURE

Acute Wound Failure

The complications associated with abdominal wound closure can be broadly classified as infection (superficial or deep), acute wound failure (dehiscence or evisceration), and incisional hernia.

The incidence of dehiscence of laparotomy wounds has been reported to be 0.2% to 2.3% (2). In most cases, acute wound failure is preceded by acute drainage of serosanguineous fluid from the incision site. In rare cases, the first sign of dehiscence is acute evisceration. The timing of evisceration is quite variable, but the average time of occurrence is the seventh postoperative day. A number of patient factors have been associated with abdominal wound dehiscence. These are: age greater than 65 years, anemia, emergency procedure, pulmonary disease, hypoproteinemia, hemodynamic instability, sepsis, obesity, uremia, malignancy, ascites, steroid therapy, and hypertension (2,3). The same studies have also identified postoperative risk factors for dehiscence, such as vomiting, prolonged ileus, urinary retention, and cough. Despite the presence of these patient factors and postoperative factors, the most common cause of acute wound failure is the technical failure of suture material tearing through the fascia. Sutures tied too tightly, sutures placed too close together or too far apart, and sutures placed with inadequate fascial bites are all at risk of tearing through the fascia with subsequent fascial dehiscence (4). The management of acute wound failure is immediate wound exploration with reduction of

Wound Infection The infection rate for laparotomy wounds has been reported to range from 1.5% to 40% (1). Well-described factors contributing to abdominal wound infection include the degree of bacterial wound contamination, host immune status, lack of adequate antiseptic skin preparation, and lack of appropriate antimicrobial prophylaxis. Additional factors are length of operation, hair removal, tissue trauma, and break in sterile technique. Wounds are classified into distinct classes based on the degree of potential bacterial contamination. The individual classes with their respective infection rates are as follows: Class I, clean (1.5%); Class II, contaminated (7.5%); Class III, clean/contaminated (15%); and Class IV, dirty (40%) (1). The diagnosis of wound infection is usually made on the basis of signs and symptoms of inflammation locally (erythema, warmth, swelling, tenderness, and purulence) and systemically (fever and sepsis). Infections occurring during the first 48 hours of the

78

Doherty and Bailey

evisceration, wound irrigation, and primary closure if possible. Retention sutures are frequently necessary. Occasionally, fascial dehiscence occurs in only a short segment of the fascial closure, with no evidence of evisceration. Acutely, this so-called ‘‘controlled dehiscence’’ can be managed nonoperatively with careful local wound care and observation. However, if bowel is exposed, there is a risk of fistula. The bowel must be protected either by closing the fascia, using adhesive packs, or by applying a split-thickness skin graft. The surgeon who uses such an approach must recognize that the wound will almost always progress to form a clinically evident incisional hernia that may require repair at a later date. The prognostic importance of fascial dehiscence cannot be overemphasized. Reported mortality rates range from 18% to 36% (2); cardiorespiratory failure and peritonitis are the most common direct causes of mortality. Other complications of dehiscence are recurrent dehiscence (2–5%), incisional hernia (14–48%), infection (14%), fistula (6%), and intraabdominal abscess (4%).

Incisional Hernia Incisional hernia is the best studied of the complications of abdominal wound closure. The incidence of incisional hernia after laparotomy has been reported to be 4% to 20%. In 5.9% of cases, the hernia becomes evident within one year of surgery; in 78% of cases, it becomes evident within two years; and in 90% of cases, within three years (5). The key risk factor for the development of incisional hernia is infection (5). Additional nontechnical factors that increase the risk of hernia are similar to those associated with dehiscence: advanced age, poor nutrition, obesity, jaundice, early reoperation, pulmonary disease, abdominal distension, emergency surgery, and midline laparotomy. A number of prospective studies, retrospective studies, and meta-analyses have found that several technical factors are associated with an increased risk of incisional hernia (6,7). These studies demonstrate that the incidence of hernia is higher when absorbable suture materials are used than when nonabsorbable materials are used, as well as when layered closure rather than mass closure is performed. The studies have not found that the use of interrupted closure rather than continuous closure substantially reduces the risk of hernia (6,7). Other studies have associated incisional hernia with factors such as the ratio of suture length to wound length (SL/WL), a measure of suture tension and size of tissue bites, and with stitch length (ratio of SL to number of stitches), another measure of wound tension. Specifically, an SL/WL ratio less than four has been associated with an increased risk of incisional hernia, and a stitch length of five or greater has been associated with an increased risk of wound infection (8,9). Excessive tension is believed to compromise local blood flow to the approximated fascial edges, and this reduction

in blood flow results in necrosis and inadequate healing. Fascial bites of inadequate size are thought to be more prone to tearing, especially with changes in intra-abdominal pressure. Techniques of incisional hernia repair and their respective complications are discussed later in this chapter.

The Difficult Abdominal Wound Closing large abdominal wall defects after emergent surgery is particularly challenging. Massive resuscitation-induced visceral edema and substantial fascial loss are the most common clinical scenarios producing such difficult wounds. In addition to increasing the risk of dehiscence and hernia, closing wounds under excessive tension can cause intraabdominal hypertension and the abdominal compartment syndrome. This syndrome is characterized by decreased blood flow to abdominal viscera, impaired venous return to the heart, renal dysfunction, and compromised diaphragmatic function, manifested as elevated airway pressures and impaired ventilation. The diagnosis of abdominal compartment syndrome is a clinical diagnosis and is based on the presence of the above-mentioned signs and symptoms in concert with intra-abdominal hypertension. Intraabdominal hypertension is best measured by transduction of the bladder pressure via an indwelling bladder catheter. Urgent abdominal decompression is recommended if bladder pressure exceeds 25 mmHg (10). Laparotomy wounds left open or those opened emergently to decompress an abdominal compartment syndrome present a difficult clinical problem. Because of the loss of abdominal wall integrity, such wounds are prone to infection and desiccation of abdominal viscera. A number of techniques for temporary wound closure have been advocated, such as towel clip skin closure, temporary silos, zippers, absorbable prosthetic mesh closure, and permanent mesh closure. In general, a staged approach to such wounds is best and often requires the application of temporary prosthetic meshes before definitive abdominal wall reconstruction. When absorbable mesh is used, it is only a temporary measure for restoring abdominal wall integrity, and hernia formation is an expected outcome. Complications of temporary prosthetic closure are mesh extrusion, wound sepsis, adhesions, and enterocutaneous fistula. The incidence of each of these complications depends on a number of factors, including the type of mesh used, the timing of mesh placement, and the degree of wound contamination.

Other Complications Relatively minor, local complications of the incisional wound are hypertrophic scar formation, scar ossification, chronic incisional pain, peri-incisional numbness (resulting from cutaneous sensory nerve injury), and stitch abscesses or stitch fistulae. Other, less common

Chapter 7: Complications of Abdominal Wall Surgery and Hernia Repair

complications are enterocutaneous fistulae and inadvertent visceral injury during closure. These complications are associated with potentially high morbidity and mortality rates and can present challenging management issues. Fortunately, these complications occur relatively infrequently and are mentioned here only for the sake of completeness.

COMPLICATIONS OF VENTRAL HERNIA REPAIR The term ‘‘ventral hernia’’ refers to a heterogeneous group of ventral abdominal wall defects that range in size and complexity from the small, uncomplicated umbilical hernia to the large, complex incisional hernia. For the most part, the present discussion will focus on the latter group. In general, larger hernias are associated with increased atrophy of the abdominal musculature, increased trophic changes in the overlying skin, and greater alteration in pulmonary mechanics. As a result, repair of large ventral hernias often requires the use of prosthetic material and frequently mandates extensive preoperative preparation, so that the incidence of complications can be reduced. At present, two operative approaches to the repair of such hernias are available to the surgeon: open ventral herniorrhaphy and laparoscopic ventral herniorrhaphy.

Complications of Open Ventral Herniorrhaphy The most common complication of open ventral hernia repair is hernia recurrence. The risk factors for recurrence after open ventral hernia repair are identical to the previously mentioned risk factors for hernia formation after laparotomy. These include patient factors, perioperative factors, and technical factors. Because infection is associated with the development of incisional hernias and substantially increases the rate of recurrence after repair, measures should be taken to minimize the risk of infection. Such measures include antimicrobial prophylaxis, avoidance of intraoperative wound contamination, and antiseptic skin preparation. In addition, host immune status must be optimized by correcting nutritional deficiencies, aggressively managing perioperative diabetes, and timing elective surgery to occur remotely from the administration of immunosuppressive drug therapies such as cancer chemotherapy and corticosteroids. As is true of successful laparotomy closure, successful repair of large incisional hernias requires adherence to a number of basic technical principles. The fascia incorporated in the repair must be healthy and must be approximated with minimal tension. Fascial bites should incorporate healthy fascia approximately 1 cm from the fascial edge and should be placed approximately 1 cm apart. To further limit tension, the repair must be undertaken with anesthesia that provides sufficient relaxation of the abdominal wall. Before the availability of prosthetic mesh materials, the recurrence rates for incisional hernia repair were

79

reported to be as high as 30% to 50%, but after the use of prosthetic materials became widespread, the recurrence rates decreased to their current level of 0% to 10% in most studies (11). This decrease is directly attributable to the use of tension-free prosthetic repairs. Despite this dramatic reduction in recurrence rates, mesh repairs should not be employed indiscriminately. The use of mesh should be limited to instances in which primary closure cannot be performed without excessive wound tension. To avoid the complications associated with the use of mesh for ventral hernia repair, several modifications of primary closure have been advocated, such as using internal retention sutures, placing incisions to relax the anterior rectus sheath, and mobilizing the inferior aspect of the rectus musculature (11). When mesh is needed, the ideal prosthetic material should exhibit two fundamental properties: retention of high-intrinsic tensile strength and allowance of external tissue ingrowth or incorporation. Prosthetic materials are classified as absorbable or nonabsorbable. The absorbable meshes are polyglyactin (Vicryl1, ETHICON, INC., Johnson & Johnson Corporation, New Brunswick, New Jersey, U.S.A.) and polyglycolic acid (Dexon1, Syneture, a division of United States Surgical Corporation, Norwalk, Conneticut, U.S.A.). The nonabsorbable meshes include two types of polypropylene mesh: Marlex1/Bard1, Murry Hill, New Jersey, U.S.A.; and Prolene1, ETHICON, INC., Johnson & Johnson Corporation, New Brunswick, New Jersey, U.S.A.). Other nonabsorbable meshes are polyester (Mersilene1, ETHICON, INC., Johnson & Johnson Corporation, New Brunswick, New Jersey, U.S.A.), and expanded polytetrafluoroethylene (ePTFE; Gore-Tex1, W.L. Gore & Associates, Inc., Newark, Delaware, U.S.A.). The nonabsorbable materials have the advantages of long-term retention of tensile strength and, with the notable exception of ePTFE, extensive tissue ingrowth. The absorbable materials have the advantage of being safer for use in the presence of contamination or active infection. As mentioned previously, tension-free mesh repairs have the advantage of being associated with lower recurrence rates than primary closure under tension. On the other hand, both animal experiments and clinical trials have demonstrated that the use of prosthetic materials is associated with a higher incidence of local wound complications. In a recent large retrospective study, Leber et al. (12) reported a variety of early and late complications associated with the use of prosthetic mesh for incisional hernia repair. Early postoperative complications were ileus (8%), cellulitis (7%), wound drainage (4%), hematoma/seroma (3%), pneumonia (1%), pulmonary embolus (1%), and deep venous thrombosis (0.5%). Late complications were hernia recurrence (16.8%, occurring at a median of one year), chronic infection/sinus (6%, at a median of six months), small-bowel obstruction (5.5%, at a mediian of 18 months), and enterocutaneous fistulae. Other

80

Doherty and Bailey

factors contributing to the increased risk of fistula were incarceration, obstruction, upper abdominal location of hernia, and history of previous wound infection. The formation of enterocutaneous fistulae has also been associated with failure to interpose tissue between prosthetic mesh and underlying bowel. This risk of fistula and the related risk of adhesions between bowel and mesh have led many authors to recommend preserving a portion of the dissected hernia sac for use as an extra layer of the closure, which can isolate the abdominal contents from the mesh. When the hernia sac is not available, free peritoneal grafts and commercially available bioabsorbable membranes and dual-sided mesh products have been advocated for this purpose. The local wound complication rates associated with polypropylene and ePTFE nonabsorbable meshes are low (0–6% for each specific complication), as are the recurrence rates (0–10%), provided that these meshes are placed in uninfected and uncontaminated wounds. With ePTFE, an infection rate as low as 3% has been reported, but this rate triples when infection is present and quadruples when gross local purulence is present (13). A number of studies have shown that the technique employed for mesh fixation does not affect complication rates. Nevertheless, onlay techniques in which mesh is placed anterior to a completed fascial closure are not recommended because of the risk of inadvertent bowel injury. An onlay technique in which an anteriorly placed mesh is sutured under direct vision before fascial approximation and is secured afterward is a safe alternative. The mesh position can be extrafascial, subfascial, or intraperitoneal, with no significant difference in recurrence rates. Low risk of recurrence depends more on secure suturing and generous (4 to 8 cm) overlapping of the mesh and fascia than on position.

Complications of Laparoscopic Ventral Herniorrhaphy In recent years, a laparoscopic alternative to open ventral herniorrhaphy has emerged. Although several variations of the procedure exist, the basic approach involves laparoscopic reduction of the hernia contents, intraperitoneal placement of prosthetic mesh, and fixation of the mesh to the anterior abdominal wall. Although small defects can occasionally be repaired primarily by using laparoscopic intracorporeal suturing techniques, laparoscopic ventral hernia repair usually requires intraperitoneal mesh placement. To limit the potential for complications arising from inflammatory reactions between the mesh and the abdominal contents (adhesions and fistulae), the least reactive prosthetic material, ePTFE, is usually used for laparoscopic repair. A large number of retrospective and prospective studies (14–18) have demonstrated complication rates of 5% to 20% for the laparoscopic approach. Intraoperative complications are subcutaneous emphysema

and hypercarbia from abdominal insufflation (0.7%), respiratory failure (0.7%), and inadvertent enterotomy (0–4.8%). Postoperative complications are seroma (1.2–16%), mesh infection (0–3.6%), trocar site infection (0–3.3%), ileus (2–10%), small-bowel obstruction (0–4.8%), pulmonary compromise (0–4.8%), and chronic pain (0–3.6%). Excluding the complications unique to the laparoscopic approach, nearly all of these complications occur at lower rates than those reported for open herniorrhaphy. The exception is seroma, which appears to be related to the use of nonreactive ePTFE mesh. Some surgeons advocate the use of dual-sided (smooth inner surface and rough outer surface) ePTFE mesh to reduce the high incidence of seroma. The rates of recurrence for laparoscopic ventral herniorrhaphy are also quite low (1–4%), much lower than the rates generally reported for open repair. Laparoscopic ventral hernia repair also has the advantages of decreased hospital length of stay, decreased postoperative pain, and more rapid return to baseline level of physical activity (14–18).

Complications of Ventral Hernia Repair for Patients with Cirrhosis Although preoperative control of ascites is strongly recommended, patients with cirrhosis can usually safely undergo inguinal herniorrhaphy without a substantial increase in the rates of local wound complications or recurrence (19). For patients with uncontrolled cirrhosis, however, ventral herniorrhaphy is associated with high morbidity and mortality rates; the rates increase to 15% in association with rupture and to 26% is association with strangulation (20). The most common intraoperative complication is bleeding; it occurs as the result of the coagulopathy, thrombocytopenia, and periumbilical hypervascularity often associated with cirrhosis. When ascites is present, postoperative complications usually relate to ascites leak. These complications are impaired wound healing, wound infection, and peritonitis. These complications and the large amount of tension placed on the repair by massive ascites result in recurrence rates as high as 50% to 60%. Therefore, ascites should be aggressively corrected preoperatively. In cases of intractable ascites, herniorrhaphy may be delayed until after peritoneovenous shunting, portasystemic shunting, transjugular intrahepatic portacaval shunting, or liver transplantation has been performed. If a patient with ascites requires emergent herniorrhaphy and no intra-abdominal infection or contamination is present, simultaneous peritoneovenous shunt placement can be performed. The hernia repair itself should include separate closure of the peritoneum and primary approximation of the fascial edges with absorbable suture, and tight closure of the skin with a continuous nylon suture or a subcuticular suture. Afterwards, the wound should be carefully observed for any signs of infection or leakage; such findings mandate immediate wound exploration.

Chapter 7: Complications of Abdominal Wall Surgery and Hernia Repair

81

Loss of Domain

Intraoperative Complications

Large, chronic incisional and inguinal hernias can cause the abdominal viscera to lose the ‘‘right of domain’’ in the abdominal cavity. Under such circumstances, reduction of the hernia contents into the abdomen and subsequent repair of the hernia defect may cause intra-abdominal hypertension and may precipitate abdominal compartment syndrome. Several maneuvers can limit the risk of this highly morbid complication. One such maneuver is progressive preoperative pneumoperitoneum, the gradual enlargement of the peritoneal cavity by repetitive air insufflation over a period of two to three weeks (21). The goal of therapy is to expand the capacity of the abdominal cavity, so that it can accommodate the hernia contents and allow a safe reduction and repair of the hernia. Another technique often used in conjunction with progressive pneumoperitoneum for repair of large inguinal and incisional hernias is the placement of a large sheet of mesh that extends well beyond the edges of the defect. This technique (the ‘‘Stoppa repair’’) provides wide adhesion of the mesh to the abdominal wall and renders the visceral sac indistensible (22). Regardless of the technique used to repair such large hernias, whenever loss of domain is suspected the patient should undergo preoperative pulmonary function testing and preoperative optimization of pulmonary status. Aggressive respiratory physiotherapy, cardiovascular conditioning, and smoking cessation can improve pulmonary function and thus limit adverse pulmonary sequelae of postoperative intra-abdominal hypertension and diaphragmatic embarrassment.

Intraoperative complications are injury to vascular structures, spermatic cord transection, injury to the vas deferens, nerve injury, testicular devascularization, and injury to the viscera (23). As will be discussed below, these complications are often not recognized at the time of surgery and may occur as complications during the postoperative period. Nevertheless, they are discussed here as intraoperative complications because the initial technical error occurs during the surgical procedure itself. Hemorrhage can result from injury to any of multiple vascular structures in the inguinal region: the pubic branch of the obturator artery, the cremasteric artery, the inferior, deep epigastric vessels, the deep circumflex iliac vessels, the external iliac vessels, and the femoral vessels. Injuries to the epigastric, pubic, and cremasteric vessels can usually be managed safely and effectively with direct ligation of the bleeding vessel. Injuries to the deep circumflex iliac, external iliac, and femoral vessels usually result from careless suture placement and are best managed by removing the offending suture and applying direct pressure. If direct pressure fails to provide adequate hemostasis, wide exploration of the femoral sheath is necessary for providing access for more effective manual compression or for future suture repair when necessary. When a vascular injury occurs intraoperatively, careful postoperative observation is necessary so that arterial or venous thrombosis or thromboembolic events can be detected. Perioperative venous thrombosis has been associated with thrombophlebitis of the dorsal vein of the penis, a complication with an incidence as high as 0.65% (23). Delayed complications of vascular injury are arterial or venous stenosis, pseudoaneurysm, and arteriovenous fistula. Failure to detect or adequately address small vascular injuries at the time of surgery often results in the formation of postoperative hematomas in the wound or in the scrotum. These hematomas usually resolve spontaneously and rarely require exploration, but they can cause substantial discomfort for the patient and can become secondarily infected. Hernia repair may occasionally require intentional sacrifice of the spermatic cord. This maneuver is usually reserved for particularly difficult large or recurrent inguinal hernias in elderly men. After inguinal herniorrhaphy, inadvertent cord transection usually causes fever and testicular swelling and tenderness; it may cause the long-term complications of testicular atrophy or hydrocele formation. These complications are discussed in detail below. The nerves at risk of injury during open inguinal hernia repair are the ilioinguinal nerve, the iliohypogastric nerve, and the genital and femoral branches of the genitofemoral nerve. The ilioinguinal nerve lies beneath the external oblique aponeurosis along the surface of the spermatic cord. Injury most commonly occurs when the external oblique is opened for

COMPLICATIONS OF INGUINAL HERNIA REPAIR Despite the fact that it is one of the most commonly performed procedures in general surgery, inguinal herniorrhaphy presents the surgeon with a unique set of challenges. Because of the complex anatomical relationships in the inguinal region, a number of important structures are susceptible to injury, and the surgeon will be faced with a variety of technical options for repair. The present discussion is intended to outline the main complications of open and laparoscopic inguinal herniorrhaphy. Specific approaches and techniques are discussed only with reference to their association with specific complications and to the differences in their rates of recurrence.

Complications of Open Inguinal Herniorrhaphy Complications of open inguinal herniorrhaphy can be either intraoperative or postoperative. In general, both intraoperative and postoperative complications are technical in nature and can be avoided by precise knowledge of inguinal anatomy, the experience of the surgeon, and keen attention to detail during the performance of the operation.

82

Doherty and Bailey

exposure of the inguinal canal; it results in loss of sensation to the base of the penis, upper scrotum, and inner thigh. The iliohypogastric nerve can be injured by relaxing incisions in the rectus sheath or by medial dissection during preperitoneal hernia repair. Such injury usually causes sensory loss to the suprapubic area. The genitofemoral nerve perforates the internal oblique muscle at the origin of the cremaster muscle. Injury to this nerve causes motor weakness of the cremaster muscle and cutaneous sensory loss in the penis and scrotum. The femoral branch of this nerve lies deep to the inguinal canal; injury to this branch causes sensory loss to the lateral thigh. Injury to any of these nerves usually produces only temporary symptoms that characteristically resolve within six months. Reports indicate that as many as 18% to 20% of patients with hernias experience neurapraxia and hyperesthesia (23). When nerve transection is recognized during surgery, the severed nerve end should be ligated so as to reduce the possibility of the formation of a painful neuroma. Unlike nerve transection, nerve entrapment can result in the development of serious long-term pain syndromes. Chronic pain after herniorrhaphy has been reported to occur in as many as 5% of cases (23). Genitofemoral neuralgia is a well-described chronic pain syndrome associated with inguinal herniorrhaphy. Symptoms are hyperesthesia in the cutaneous distribution of the genitofemoral nerve and chronic inguinal pain extending to the genitalia and upper thigh. This pain is often exacerbated by walking, hip extension, and pubic tubercle pressure and can frequently be relieved by hip flexion at rest. Pain and paresthesias associated with nerve entrapment and neuroma formation can initially be managed with local nerve blocks. The iliohypogastric and ilioinguinal nerves can be blocked by using an L1 and L2 block. Persistent symptoms may require reexploration, with ligation and severance of the involved nerve. The presence of a short-lived response to a local block can help guide therapy toward a specific nerve, but in the absence of such evidence, therapy is best directed empirically at all three nerves. Occasionally, the condition does not respond to appropriate nonsurgical and surgical therapies; in such cases, patients should be referred to a chronic pain specialist. The blood supply to the testis is primarily derived from the internal spermatic artery, which is part of the spermatic cord. In the event of any interruption of flow in the internal spermatic artery, collateral circulatory input provided by branches of the vesical, prostatic, and deferential arteries can prevent testicular ischemia. If the collateral circulation is also disrupted, acute testicular necrosis or testicular atrophy may result. Thus, care must be taken to preserve the collateral vessels and the internal spermatic artery during inguinal herniorrhaphy. Because of its potential impact on fertility, injury to the vas deferens is a serious concern of all surgeons performing inguinal hernia surgery. When it occurs,

transection of the vas deferens mandates immediate repair. Approximately 50% of such repairs yield a functional result. Improper handling of the vas deferens can cause injury in the absence of transection. Such injury may involve obstruction of the lumen of the vas, a lesion that can cause painful ejaculatory dysfunction. Injury to abdominal viscera during inguinal herniorrhaphy usually occurs in association with sliding hernias involving bladder or bowel wall. The wall of the urinary bladder can participate as a sliding component of the medial aspect of a direct inguinal hernia. As such, it can be injured during the placement of medial sutures during the hernia repair. When recognized intraoperatively, injury to the urinary bladder should be repaired immediately, and the repair should be protected with bladder decompression via an external bladder catheter. Bowel injury can occur during high ligation of an indirect hernia sac, when the bowel wall is a component of the sac. The injury can be a simple enterotomy or a mesenteric injury with segmental vascular compromise. In either case, potential sequelae are bowel obstruction, fistula, and abscess formation. Enterotomies are best managed with primary repair, wound irrigation, and hernia repair without prosthetic material if possible. Devascularization of bowel may require resection, with or without laparotomy, and proximal diversion may occasionally be required for colon injures.

Postoperative Complications In addition to the postoperative complications of scrotal ecchymosis, testicular atrophy, and neuroma mentioned in the preceding discussion, a number of other postoperative complications are associated with open inguinal herniorrhaphy. These are urinary retention, osteitis pubis, testicular swelling, hydrocele, infection, missed hernia, and recurrence. Urinary retention can occur in association with as many as one-third of all open inguinal hernia repairs; it appears to be more common when bilateral hernia repair is performed. It also appears to be directly related to administration of increasing doses of postoperative narcotics. This complication is exacerbated by the presence of other conditions associated with urinary retention, such as benign prostatic enlargement. Initial management of postoperative urinary retention is intermittent bladder catheterization until the patient is able to void spontaneously. In some cases, long-term bladder catheterization and referral to a urologist may be necessary. To avoid this complication, urology referral and treatment of known or suspected prostatic disease may be advisable prior to elective hernia repair (24). The placement of suture material in the periosteum of the pubic bone during inguinal herniorrhaphy has been associated with persistent inflammation of the periosteal layer, a condition known as osteitis pubis. This painful syndrome is usually self-limiting

Chapter 7: Complications of Abdominal Wall Surgery and Hernia Repair

and responds to the administration of nonsteroidal antiinflammatory agents. Although osteitis pubis was once a well-described complication of inguinal hernia repair, modern techniques of inguinal herniorrhaphy have rendered this complication exceedingly rare. A swollen testis after inguinal hernia repair is usually caused by tight closure of the tissues around the spermatic cord. Occasionally, such swelling results from lymphatic injury, venous injury, or venous thrombosis induced during the surgical dissection of the cord; other causes are hematoma or seroma. Regardless of its cause, postoperative testicular swelling is usually self-limiting; the management involves scrotal support and pain control until swelling resolves. Postoperative swelling associated with severe testicular pain and fever indicates ischemic orchitis. In the absence of obvious arterial injury, ischemic orchitis may result from venous thrombosis of the cord vessels. In severe cases, the pain may last for as long as six weeks, and the condition may progress to testicular atrophy. The incidence of testicular atrophy has been estimated to be 0.036% after primary repairs and 0.46% after repair of recurrent hernias (23). The development of fluid collections along the course of the spermatic cord is not uncommon after inguinal hernia repair; it occurs with an incidence of 0.7% (23). These collections are commonly referred to as hydroceles, but they actually result from a number of different causes such as retained distal hernia sac, impaired lymphatic or venous drainage, and inflammatory fluid accumulations in proximity to mesh. Postoperative formation of seromas has been associated with the degree of tissue trauma and the use of prosthetic mesh material; its incidence ranges from 0% to 17.6% (23). Hydroceles and seromas rarely require operative intervention unless they become infected. Those that persist for more than six to eight weeks may require aspiration, open drainage, or both. The incidence of wound infection after open inguinal herniorrhaphy is approximately 1%. Factors that increase infection rates are advanced age (3.2-fold increase), female sex (2.1-fold increase), presence of a drain (9-fold increase), duration of operation (increased infection risk from 2.7% to 9.9% with increased duration from 30–90 minutes), incarceration (7.8% infection rate), and recurrent hernia (10.8% infection rate) (23). The use of prosthetic mesh does not appear to increase these rates, but it has been associated with the phenomenon of delayed infection months to years after herniorrhaphy. Moreover, the presence of infection or contamination preludes the use of mesh. Wound infections are managed with drainage and local wound care. Antibiotics are rarely indicated. The presence of deep infections after inguinal herniorrhaphy increases the risk of hernia recurrence. Occasionally, a new hernia will be noted after an apparently successful inguinal hernia repair. Such cases usually involve a small indirect or femoral hernia not appreciated at the time of direct inguinal herniorrhaphy. A second hernia is present in at least

83

13% of cases (23). This complication is entirely preventable and can be avoided by careful palpation for other hernias at the time of repair and by complete opening of the floor of the inguinal canal, a maneuver not performed by a large number of surgeons. Although fascial weakness may contribute to some hernia recurrences, the cause of recurrence is almost always a technical error. The primary technical factor influencing hernia recurrence is the degree of tension on the repair. Tension causes impaired healing and renders the repair susceptible to disruption and subsequent recurrence. The differences in recurrence rates among the various techniques of open inguinal hernia repair probably reflect differences in the degree of intrinsic tension of the repair. A summary of these recurrence rates appears in Table 1 (23). As this table demonstrates, the recurrence and rerecurrence rates associated with the two techniques of open inguinal hernia repair most commonly used today, the tension-free mesh technique and the mesh plug technique, are similarly low. The widespread acceptance of these two tension-free approaches to inguinal herniorrhaphy is due primarily to the widespread recognition of the important relationship between tension and recurrence.

Complications of Laparoscopic Inguinal Herniorrhaphy Laparoscopic inguinal hernia repair recently entered its second decade of existence, and substantial improvements in techniques and instrumentation continue to advance the field. The three main techniques of laparoscopic inguinal herniorrhaphy are the intraperitoneal onlay method (IPOM), the transabdominal preperitoneal repair (TAPP), and the total extraperitoneal repair (TEP). The evolution of laparoscopic inguinal herniorrhaphy from IPOM to TAPP to TEP has been driven by the desire to maintain low Table 1 Rates of Recurrence and Re-recurrence of Hernia for Various Techniques of Open Inguinal Hernia Repair Technique Bassini Shouldice McVay Nyhus Nyhus mesh buttress Rives mesh Stoppa mesh Tension-free mesh (Lichtenstein) Mesh plug repairs Bassini femoral Bassini-Kushner femoral Moschowitz femoral Nyhus femoral McVay femoral Mesh femoral (Stoppa, Wantz, Bendavid, Lichtenstein, Rutkow) Source: From Ref. 23.

Recurrence rate (%)

Re-recurrence rate (%)

2.9–25 0.2–2.7 1.5–15.55 3.2–21.0 0–1.7 0–9.9 0–7 0–1.7 0–1.6 2.3 2–6.5 0.9 0–0.95 0–3.1 0–1.1

6.5–13.4 2.9–6.4 2.4–5.5 9.5–27.0 0–1.7 1.7–3.2 0–8 0–3.4 0.5–1.6 – – – – – –

84

Doherty and Bailey

recurrence rates while reducing operative time, cost, complications, and anesthetic risk. The IPOM technique involves placing an intra-abdominal sheet of mesh directly over the hernia defect. Although IPOM remains important from a historical perspective, the TAPP and TEP repairs have become the favored approaches of nearly all laparoscopic surgeons. The TAPP repair involves transabdominal laparoscopic dissection of the anterior abdominal wall and secure coverage of the mesh with peritoneum. This approach has the advantage of preperitoneal mesh placement, but it has the disadvantage of requiring entry into the peritoneal cavity and the use of general anesthesia. The TEP repair involves extraperitoneal dissection to separate the peritoneum from the inguinal area, laparoscopic dissection and reduction of the hernia, and placement of mesh between the peritoneum and the transversalis fascia defect. Because laparoscopic inguinal herniorrhaphy is a relatively new procedure and is subject to continual modification and refinement, the complication rates associated with it have steadily declined over the past decade. The reported overall complication rates for both TAPP and TEP range from 5% to 32%, and the recurrence rates range from 0% to 2% (25). The complications of laparoscopic inguinal herniorrhaphy can be classified as those related to patient selection, those related to laparoscopy, those common to both TAPP and TEP, and those unique to each specific technique.

Complications Related to Patient Selection With respect to patient selection, the European Association for Endoscopic Surgery approves the use of laparoscopic inguinal hernia repair for Nyhus Type IIIA to IIIC and Type IV inguinal hernias. For these types of hernias, the laparoscopic approach is a safe and reasonable alternative to traditional open approaches. Contraindications to the laparoscopic approach are based on the potential for complications arising from difficult laparoscopic dissections and from the mandatory use of mesh. Relative contraindications to the laparoscopic approach are prior ipsilateral laparoscopic repair, prior groin irradiation or inflammation, large scrotal hernia, small congenital hernia not requiring mesh, and morbid obesity. Absolute contraindications are contraindication to general anesthesia, contraindication to the use of mesh (i.e., infection), and incarceration.

Complications Related to Laparoscopy The complications related to laparoscopy itself are those related to the access technique and those related to pneumoperitoneum. Those related to access technique are Veress needle or trocar injuries to abdominal viscera with the TEP approach. Complications related to CO2 insufflation are pneumothorax, hypercarbia, and subcutaneous emphysema.

Complications Related to Transabdominal Preperitoneal Repair Complications unique to the TAPP approach are those related to the administration of general anesthesia and those related to mesh placement and fixation. The TAPP approach requires general anesthesia to allow pneumoperitoneum and adequate operative exposure. Thus, unlike open inguinal herniorrhaphy and the laparoscopic TEP approach, TAPP cannot be performed with local or regional anesthesia alone. Although most patients tolerate general anesthesia without serious adverse sequelae, the use of general anesthesia for high-risk patients, such as those with severe cardiac and pulmonary disease, is associated with considerable risk. The TAPP approach may therefore expose such patients to substantial risk for perioperative cardiac and pulmonary complications. The TAPP approach also requires that the mesh prosthesis be adequately secured and isolated from the abdominal contents by the peritoneum. Failure to achieve adequate fixation or coverage may result in fistula, internal hernia, bowel obstruction, or hernia recurrence.

Complications Related to Total Extraperitoneal Repair Complications related to the TEP approach are primarily those associated with extraperitoneal dissection and exposure of the inguinal area. Injury to the epigastric vessels can be avoided by midline port placement and preservation of the vessel’s location on the abdominal wall. Peritoneal injury can cause loss of exposure due to the leakage of insufflated CO2 into the peritoneal cavity. Although this loss of exposure may precipitate conversion to TAPP or open repair, restoration of the operative exposure may be possible by venting the pneumoperitoneum with a Veress needle and repairing the peritoneal defect laparoscopically.

Complications Common to Transabdominal Preperitoneal Repair and Total Extraperitoneal Repair In general, the complications common to TAPP and TEP are also shared with open techniques of inguinal herniorrhaphy. These are vascular injury, visceral injury, testicular atrophy, nerve injury, and complications related to the use of mesh. Vascular structures at particular risk of injury during laparoscopic hernia repair are the iliac, iliopubic, and accessory obturator vessels. Bladder injury can occur with either TAPP or TEP, but is more common with TAPP. Bladder injury during TAPP can be avoided by limiting dissection to the area lateral to the medial umbilical ligament. The management principles for bladder injuries sustained during laparoscopic herniorrhaphy are the same as those for bladder injuries that occur during open inguinal herniorrhaphy. The bladder should be repaired primarily by using laparoscopic techniques, and bladder decompression should be maintained

Chapter 7: Complications of Abdominal Wall Surgery and Hernia Repair

postoperatively via an indwelling bladder catheter. Although ischemic orchitis and testicular atrophy are relatively common complications of open inguinal herniorrhaphy, these complications are rarely seen after laparoscopic repair. The low incidence of these complications is mainly due to the fact that both TAPP and TEP laparoscopic techniques use preperitoneal approaches and avoid excessive cord dissection, thus limiting the risk of testicular devascularization. The overall incidence of nerve injuries in association with laparoscopic inguinal hernia repair has been reported to be 0% to 5.1%. The specific nerves at risk of injury are the genitofemoral nerve (40% of laparoscopic nerve injuries) and the lateral femoral cutaneous nerve (26.7% of injuries); the remaining one-third of nerve injuries are not attributable to a specific nerve (26). In general, nerve injuries can be avoided by leaving the distal sac in place, avoiding dissection in the area of the iliacus fascia, and placing all lateral staples above the iliopubic tract and medial to the anterior superior iliac spine. As is true of nerve injuries after open hernia repair, those that occur after laparoscopic repair usually resolve spontaneously, but nerve blocks, surgical reexploration, or both will be required. As is also true of open hernia repair, with laparoscopic repair mesh complications such as infection and erosion into adjacent structures (i.e., bowel or urinary bladder) are quite rare.

85

Recurrence As mentioned previously, the recurrence rate for laparoscopic herniorrhaphy is reported to range from 0% to 3% (34–36). Technical factors contributing to recurrence include inadequate mesh size, incomplete pelvic dissection, missed hernias, and clips pulling through tissue. Inadequate experience on the part of the surgeon also contributes to recurrence. The importance of this factor is underscored by the fact that most published studies (25,34,36) have found that the recurrence rates for both TEP and TAPP have decreased to less than 1%, probably because of the increased experience gained as these techniques have become more widespread.

Comparison of Transabdominal Preperitoneal Repair and Total Extraperitoneal Repair A number of retrospective and prospective studies (33,34,37) have compared the results achieved by TAPP with those achieved by TEP. These studies consistently demonstrate that both techniques are associated with low complication and recurrence rates, but conclude that TEP is the favored laparoscopic approach because of its substantially lower risk of intraperitoneal complications and its superior recurrence rates. Because of the contraindications to TEP, many surgeons favor TAPP (33,34).

Overall Complication Rates A number of prospective and retrospective studies (27–33) have determined the rates of specific complications after laparoscopic herniorrhaphy. These complication rates are presented in Table 2.

Table 2 Rates of Specific Complications After Laparoscopic Herniorrhaphy Type of complication Intraoperative Inadvertent enterotomy Bladder injury Bleeding Conversion to open approach or transabdominal preperitoneal repair Injury to vas deferens Equipment for extra trocars Intraoperative cardiac events Postoperative Seroma Hematoma Wound infection Urinary retention Persistent pain or paresthesias Pneumoscrotum Scrotal hematoma Hydrocele Epididymitis Internal hernia Unplanned admission Source: From Refs. 26–32.

CONCLUSION Complications of abdominal wall surgery and hernia repair are some of the most common complications encountered by general surgeons. Therefore, a thorough knowledge of the causes and implications of these complications and of their management is essential to the successful practice of general surgery.

Rate of occurrence (%)

REFERENCES 0–1 0–3.3 1–4 0–5 65 for single-lung transplantation Coronary artery disease or left ventricular dysfunction Psychosocial problems Colonization with resistant bacteria, atypical mycobacteria, or Aspergillus Abbreviation: HIV, human immunodeficiency virus. Source: Adapted from Ref. 11.

Chapter 24: Complications of Lung Transplantation

DONOR SELECTION Despite improvements in the treatment of potential donors, the lungs of brain-dead patients are often damaged, typically by aspiration, excessive administration of fluids, neurogenic edema, and ventilatorassociated pneumonia. Fewer than 20% of cadaveric donors contribute to the lung donor pool. What were once standard criteria for potential donor candidates (Table 4) are often relaxed because of the shortage of suitable donor organs. The ideal donor is less than 55 years old with a history of less than 20 pack-years of smoking, a clear chest radiograph, acceptable oxygenation (arterial oxygen pressure >350 mmHg at a fraction of inspired oxygen of 100% and a positive end-expiratory pressure of 5 cmH2O), normal airways as confirmed by bronchoscopy, and no evidence of chest trauma or contusion. The increased demand for donors has led to the use of marginal donors who were once considered unacceptable. In fact, acceptable outcomes have been achieved by transplantation of lungs with secretions that can be lavaged clear with a bronchoscope, revealing normal underlying mucosa; lungs with marginal oxygenation; and lungs with unilateral infiltrates. However, the use of marginal donors increases the risk of graft dysfunction and postoperative complications. The decision to use any particular donor is often weighed against the risks for the intended recipient. A marginal donor may be acceptable for a recipient whose condition has deteriorated to the point at which mechanical ventilation is required; on the other hand, because of the risk of reperfusion injury, a nearly perfect donor is required for patients who need SLT for pulmonary vascular disease. Techniques for procuring and preserving donor lungs are continuing to evolve. Most centers use a cold, crystallized solution [CelsiorTM (Genzyme Corporation, Cambridge, Massachusetts, U.S.A.) or PerfadexTM (Transplantation Systems, Goteborg, Sweden)] with a single-flush technique, preceded by an injection of prostaglandin E1 into the pulmonary artery or as a systemic infusion titrated to the donor’s blood pressure (19). Celsior and Perfadex are extracellular electrolyte Table 4 Criteria for Suitable Lung Donor Preliminary assessment Age 350 mmHg on FiO2 100% and PEEP 5 cmH2O Adequate size match and ABO compatibility Negative history of smoking (20 pack years), chest trauma or thoracic surgery, aspiration, or sepsis Final assessment Chest radiograph and oxygenation show no deterioration Results of bronchoscopy negative for aspiration Visual and manual assessment negative for contusions, trauma, and adhesions Abbreviations: PaO2, partial pressure of oxygen; FiO2, fraction of inspired oxygen; PEEP, positive end-expiratory pressure; ABO, blood type A,B,O. Source: Adapted from Ref. 16.

289

solutions containing low concentrations of potassium; the administration of these solutions has been shown to reduce the likelihood and severity of ischemiareperfusion injury (IRI) and to improve postischemic oxygenation and functional outcome. Prostaglandin E1, a potent pulmonary vasodilator, improves distribution of the flush solution. Instilling the flush under a monitored pressure of 20 to 40 cmHg may avoid perfusion injury (20). Furthermore, inflating the lung with 100% oxygen, using moderate pressures (21), and transporting the lung at 4 C (22) have produced better graft outcomes in experimental models.

INTRAOPERATIVE MANAGEMENT Surgical Techniques The technique of performing lung transplantation has evolved substantially over the past two decades. Single and bilateral sequential lung transplantations are the procedures of choice; BLT with tracheal anastomosis has largely been abandoned because of problems related to ischemia at the anastomotic site. The recipient’s condition is monitored with a pulmonary artery catheter, a peripheral artery catheter, and an end-tidal CO2 monitor. Anesthesia may be initiated with a single-lumen endotracheal tube, particularly for patients with septic lung diseases who require preoperative (bronchoscopic) clearance of large airway secretions. However, the preferred approach requires changing to a double-lumen endotracheal tube for the operative procedure. Transesophageal echocardiography (TEE) should be performed so that right and left ventricular function can be assessed and the need for CPB can be determined. TEE can also help in assessing the outcome of the anastomosis between the pulmonary vein and the left atrium. SLT is usually performed through a standard posterolateral thoracotomy incision, although anterior incisions have been used. The lung to be removed is deflated while ventilation to the contralateral lung is continued. BLT is performed through a transverse thoracosternotomy incision; transplantation of the right and left lungs is achieved by sequential singlelung transplant procedures. The lung with the poorer function is removed first. Clamping the pulmonary artery may worsen hypoxemia, hemodynamic instability, and echocardiographic evidence of right ventricular failure; inotropic support, pulmonary vasodilators (nitric oxide or prostaglandin E1), or both may be necessary. CPB may become necessary at this point. Removal of the lung is completed by dividing the pulmonary artery and veins as far distally as possible and resecting the bronchus at the level of the takeoff of the upper lobe. Care must be taken not to injure the phrenic, vagus, and recurrent laryngeal nerves. Graft implantation is begun with the bronchial anastomosis; next, the pulmonary artery is anastomosed; finally, the pulmonary vein is sutured to the left

290

Qi and Fertel

atrium. The donor bronchus is cut two cartilaginous rings above the takeoff of the upper lobe, and the membranous bronchi are approximated. The cartilaginous bronchi are telescoped, with intussusception of the donor and recipient bronchi. The pulmonary artery is anastomosed; then the left atrial anastomosis is performed with excision of the pulmonary venous stumps. Before the anastomosis is completed, the clamps are removed from the pulmonary artery and vein, and air is evacuated from these vessels by flush irrigation. The vessels are then perfused with leukocytedepleted blood. This sequence is repeated if BLT is performed. Hypoxemia or hemodynamic instability necessitates the use of CPB. After the chest has been closed, the double-lumen endotracheal tube is replaced by a large single-lumen endotracheal tube. The bronchial anastomosis is inspected by fiberoptic bronchoscopy and any blood clots or residual secretions are removed.

Application of Extracorporeal Membrane Oxygenation Extracorporeal membrane oxygenation (ECMO) has been used intraoperatively to support the patient with immediate graft dysfunction that is refractory to medical treatment and standard methods of ventilation. ECMO is a useful bridge treatment for patients with severe IRI; it allows time for recovery and for retransplantation if the graft completely fails. ECMO differs from CPB in its artificial lung and blood reservoirs, mode of ventilator setting, methods of anticoagulation employed, temperature, and duration (23–25). Circuits are placed either vein to vein (also called venovenous or V-V ECMO), for patients whose hemodynamic condition is stable, or vein to artery (venoarterial or V-A ECMO), for patients whose cardiac function is unstable. Access may be peripheral (via percutaneous catheter) or central (within the thoracic cavity). V-A ECMO requires minimal ventilator support, whereas V-V ECMO requires moderate ventilator settings. Small tidal volumes should be used with both methods so that the likelihood of lung injury can be reduced. The use of ECMO has been associated with an increased risk of renal and neurologic impairment (23). However, when used early to treat reversible lung injury, it can support the patient by providing adequate gas exchange and systemic perfusion (21–25).

COMPLICATIONS RELATED TO IMMEDIATE GRAFT DYSFUNCTION Hyperacute Rejection Hyperacute rejection results from preformed antibodies that target major allograft antigens; these antibodies result in rapidly progressive injury and graft failure within minutes to hours after transplantation. Pathologic findings include the presence of

fibrin and platelet thrombi in arterioles and capillaries of the allograft, in association with prominent neutrophilia. Immunohistochemical analysis shows the presence of antibodies on the endotracheal surface and within vessel walls (26). This is followed by endothelial damage, leading to edema, hemorrhage, and infarction. Clinically, the graft becomes edematous, cyanotic, and mottled. Preformed antibodies to HLA antigens are almost always found and the outcome is uniformly fatal (26,27,29). Better techniques for detecting antibodies and improvements in crossmatching techniques are needed if this devastating problem is to be eliminated.

Ischemia-Reperfusion Injury IRI is the most common cause of immediate graft dysfunction after lung transplantation. Although it has not been well defined, IRI, also called the reimplantation response, is a syndrome characterized by edema and infiltrates, worsening gas exchange, and histologic evidence of diffuse alveolar damage (28,30,31). Radiographic studies have demonstrated that as many as 97% of lung allografts exhibit perihilar edema in the immediate postoperative period (30,32). However, severe graft dysfunction has been reported among only 20% to 37% of recipients; the most severe cases produce a pattern similar to that of adult respiratory distress syndrome (30,33–35). Infection, rejection, volume overload, and pulmonary venous obstruction must be excluded. The cause of IRI appears to be increased vascular permeability related to ischemia and preservation, although no direct association with ischemic time, age, sex, or underlying disease has been found (34,35). CPB appears to increase the incidence and severity of IRI (35,36). Treatment is largely supportive, but nitric oxide has been used (37). Selective lung ventilation has been attempted if lung injury is unilateral. ECMO may be required (38) and has been useful in allowing time for graft recovery. The risk of posttransplantation morbidity is increased for patients with IRI, if mechanical ventilation and intensive care unit stay are prolonged for more than 10 days. Mortality rates as high as 40% have been reported with severe injury (28). The condition of the graft with regard to donor abnormalities (contusion, aspiration, and preservation) may play a role in the severity of IRI. Efforts aimed at improving the outcomes associated with IRI must focus on preventing the release of cytokines, platelet activating factor, and complement; on preventing vascular endothelial injury; and on developing better preservation techniques.

PERIOPERATIVE COMPLICATIONS Airway Complications Anastomotic necrosis with dehiscence, once the primary limitation of lung transplantation, is now a rare occurrence; this fact reflects improvements in surgical

Chapter 24: Complications of Lung Transplantation

techniques (40,41). Routine intussusception of the donor and recipient bronchi has helped to substantially reduce this once frequent complication (42,43). When it does occur, partial dehiscence is treated conservatively by placing a chest tube and reducing the steroid dosage; complete dehiscence may require surgical intervention (44). However, if necrosis is extensive and involves the upper lobe bronchus, resection with reanastomosis may not be possible and emergency retransplantation may be the only viable option (44). Currently, the most common airway complications are bronchomalacia and stenosis at the site of the anastomosis; they occur in approximately 12% to 17% of cases (45–47). These complications can occur weeks to months after transplantation and often produce dyspnea, wheezing, and a decline in lung function with worsening airflow limitation as detected by spirometry. The definitive diagnosis is made with bronchoscopy, which shows luminal narrowing at the site of the anastomosis (34,35,43). Bronchial ischemia is related to resection of the bronchial arteries; reliance on retrograde pulmonary arterial blood to supply the airways has been implicated as a causative factor (35,44). Groups that perform revascularization of the bronchial arteries have reported better healing and a reduction in the rate of late stricture complications (3,40,46,47). However, this procedure adds technical complexity to the operation, and more data are needed if we are to determine how this added procedure might affect overall outcome. Attempts at identifying other causes of airway complications, such as lung preservation and ischemic times, anastomotic techniques, corticosteroid dosing, and rejection, have led to inconclusive results (43). An association has been noted between Aspergillus colonization or infection and airway necrosis (44), whether Aspergillus infection results from or causes the necrosis is unclear. The management of airway complications depends on the degree and location of the problem. A thin, web-like lesion may be amenable to laser therapy or bronchial balloon dilatation. However, most lesions require placement of an airway stent via bronchoscopy (43,44,47,48).

Complications of Native Lung Hyperinflation Both SLT and BLT have been successfully used to treat patients with emphysema. Although there is no difference in the three-year survival rates associated with SLT and BLT, SLT has been preferred for older, highrisk patients because it is associated with a lower rate of perioperative complications (49). Radiographs of patients who have undergone SLT for emphysema frequently demonstrate mediastinal shift and ipsilateral diaphragmatic flattening, findings suggestive of native lung hyperinflation (NLH) (50–52). Symptomatic NLH has been reported to cause hemodynamic instability and respiratory dysfunction requiring

291

independent lung ventilation (50). However, most studies have shown that NLH is not responsible for severe graft dysfunction or substantial increases in mortality rates (49–52). When it occurs, symptomatic NLH has been successfully treated with lung-volume reduction procedures on the native lung (53).

Renal Complications Renal dysfunction among recipients of lung transplants has several causes. It may be a consequence of renal hypoperfusion during surgery, resulting from either hemodynamic instability or hypovolemia. Alternatively, it may occur as the result of acute renal tubular necrosis, often caused by immunosuppressive medications such as calcineurin inhibitors (tacrolimus or cyclosporine) and lympholytic agents. Close monitoring of calcineurin inhibitor levels and appropriate dose adjustments are essential. Drug interactions are common and can exert additive nephrotoxic effects. Renal function declines most rapidly during the first six postoperative months. The clinical picture of chronic nephropathy is characterized by elevations in the serum creatinine concentration, decreases in creatinine clearance, disproportionate azotemia, hyperkalemia, proteinuria, decreases in sodium excretion, hypertension, and fluid retention. Diuresis is attempted first; if it is unsuccessful, hemodialysis is usually required.

Gastrointestinal Complications Gastrointestinal problems have been reported to occur among as many as 42% of patients undergoing lung and heart–lung transplantation (54,55). Diarrhea, ileus, gastroparesis, ulcers, and ischemic bowel occur during the early posttransplantation period (55,56), whereas problems related to diverticulitis and posttransplantation lymphoproliferative disorders (PTLD) occur among long-term survivors (57). Symptomatic gastroparesis is a serious complication and has been reported to occur in as many as 25% of cases (58). Symptoms include early satiety with nausea, vomiting, and abdominal complaints; these symptoms are associated with prolonged gastric emptying. The pathophysiology of symptomatic gastroparesis is not completely understood, but inadvertent thermal or traumatic injury to the vagus nerve has been implicated (58). Direct toxic effects of immunosuppressive medicines have also been implicated (55,58), although gastroparesis has not been reported to occur among recipients of kidney and liver transplants who receive the same medications (58). Treatment may include the temporary use of jejunal feeding tubes and cholinergic stimulants, but subtotal gastrectomy or gastric bypass may occasionally be required (56). Posttransplantation diarrhea is common and may be related to infection with agents such as Clostridium difficile, to colitis caused by cytomegalovirus (CMV), or to immunosuppressive medications

292

Qi and Fertel

such as mycophenolate mofetil (MMF). Differentiating immunosuppression-related disease from infectionrelated disease may be difficult. The presence of fever, inflammatory cells in the stool, moderate leukocytosis, and abnormal results on endoscopy, colonoscopy, or computed tomography suggest an infectious process (55). Reduction of the immunosuppressive regimen may be required.

Bone Complications Osteoporosis can contribute substantially to morbidity after lung transplantation by limiting rehabilitation potential and impairing quality of life. As many as 35% to 40% of transplant recipients with osteoporosis can experience vertebral fractures; avascular necrosis of weight-bearing joints seems to occur less frequently (33). Most fractures are diagnosed by routine radiographs; however, more sophisticated methods are required for assessing bone mass and bone mineral content. The bone density scan has been used most often as a reliable assessment of osteoporosis. Withdrawing corticosteroids from the treatment regimen of lung transplant recipients substantially slows the progression of osteoporosis. Regular active exercise, appropriate exposure to sunshine, oral calcium supplementation, and the administration of calcitonin, testosterone, estrogen, vitamin D, and biphosphonates have been successful in preventing osteoporosis among recipients of lung transplants.

Rejection Acute Rejection Acute rejection has been reported to occur as early as days and as late as years after lung transplantation. However, most episodes of acute rejection occur within the first three months after transplantation; the incidence declines thereafter (59,60). Substantial HLA mismatching, particularly at the HLA-DR and HLA-B foci, may be important risk factors (60,61). The clinical signs are nonspecific and may include fever, dyspnea, and impaired gas exchange. New infiltrates may be observed on radiographs within the first four to six weeks after transplantation; however, after that, changes in the results of chest radiography are uncommon (62). A decline in the forced expiratory volume in one second (FEV1) is common; a 10% decrease in FEV1 should signal the need for further diagnostic testing to rule out infection or rejection (63,64). When clinical findings suggest acute rejection, bronchoscopy with transbronchial biopsy is indicated and is the key to diagnosis (65,66). The histologic classification of rejection was instituted in 1990 and revised in 1995 (29). Acute rejection is based on the intensity of perivascular mononuclear cell infiltrates and the extent of their extension into adjacent alveolar septae. Classification starts with Grade 0, normal pulmonary parenchyma, and Grade 1, minimal acute rejection with no obvious

perivascular mononuclear infiltrates around blood vessels at low magnification. The classification extends to Grade 4, severe acute rejection with diffuse perivascular, interstitial, and airspace mononuclear cells, often associated with intra-alveolar macrophages, hyaline membranes, neutrophilic infiltrates, and hemorrhage (29). Infection can coexist with rejection and must be excluded before the immunosuppressive regimen is augmented. Controversy exists about the need for routine surveillance bronchoscopy for patients with no clinical symptoms, and whose condition is physiologically stable. However, histologic rejection, usually minimal to mild, has been observed in as many as 39% of biopsy specimens (65,66). Discrepancies between clinical impression and histologic diagnosis appear to be greatest during the first six months after transplantation (67). Standard treatment of acute rejection is 10 to 15 mg/kg of intravenously administered methylprednisolone daily for three successive days, followed by an increase in the maintenance dose of prednisone and subsequent tapering of the dose over one to two weeks. Most patients respond well to this treatment; follow-up biopsy is recommended. Recurrent or persistent acute rejection may be treated by repeated steroid boluses, by changing the immunosuppressive regimen from cyclosporine to tacrolimus, or from azathioprine to MMF, or by both. Antithymocyte globulin (ATG) and OKT3 monoclonal antibody have been used (68,69). Recurrent or refractory rejection is the primary risk factor for the subsequent development of bronchiolitis obliterans (70–75).

Chronic Rejection Chronic rejection, characterized clinically by progressive allograft dysfunction and histologically by bronchiolitis obliterans, is the key factor limiting the long-term survival of recipients of lung transplants. Bronchiolitis obliterans is unusual during the first six months after transplantation but affects 60% to 70% of patients who survive for five years or longer (73–75). The clinical presentation includes progressive dyspnea and airflow limitation. Histologic studies show dense eosinophilic plaques in the submucosa of small airways; these plaques result in luminal narrowing (63). Chest radiographs may be nondiagnostic, but high-resolution computed tomography may show a reduction in graft volume, bronchiectasis, air space disease, atelectasis, and nodular opacities (76). The diagnosis of bronchiolitis obliterans is often missed when the results of transbronchial biopsy are reviewed because of the characteristic patchy distribution of the disease and because of sampling error (77). As a result, the clinical criteria of bronchiolitis obliterans syndrome (BOS) have been developed on the basis of spirometric parameters (73). BOS is characterized by a decline in FEV1 that is present for at least one month and is not caused by infection or

Chapter 24: Complications of Lung Transplantation

acute rejection (73). Multiple treatment methods that augment the immunosuppressive regimen have been attempted (72) and may slow the decline in lung function. However, to date none of these methods have been proved to completely stop the progression of the disease (67,72,73,75).

Complications Related to Infection Infectious complications are an important cause of morbidity and mortality among recipients of lung transplants (11,61,76). Their incidence appears to be much higher among recipients of lungs than among recipients of other solid organs (63), perhaps because the lung is the only allograft that is exposed to the environment. Associated factors such as lung denervation (which impairs cough reflex), lymphatic disruption, and impaired mucociliary clearance also play a role. Furthermore, infections can be transmitted from the donor or from the native lung or proximal airways of the recipient; they may also develop as a result of the decrease in immune defenses related to immunosuppression. The lung is placed at an increased risk by airway instrumentation and is constantly at risk of aspiration. Anastomotic narrowing and airway mucosal edema may impair adequate clearance of organisms. Most infections among recipients of lung transplants occur within the lung allograft (35,77), although infections in the native lung have occurred among patients who have undergone SLT (78,79). Bacterial pathogens, frequently gram-negative organisms, predominate during the early posttransplantation period (33,76,77,80). Most centers empirically administer broad-spectrum antibiotics perioperatively and continue such administration until the specific organism involved has been determined by cultures from donor and recipient. Patients with septic lung diseases such as cystic fibrosis will require broadspectrum antibiotics to cover the pathogens they carried before transplantation. Patients among whom BOS develops later are at an increased risk of colonization and infection with bacterial pathogens, predominantly Pseudomonas species (80). Recipients of lung transplants frequently become colonized with Candida and Aspergillus species, but it is clinical infection with Aspergillus, which is associated with increased morbidity and mortality rates (33,77,80). Aspergillosis is acquired by inhaling spores and therefore occurs far more frequently among recipients of lung transplants (as many as 48%) than among recipients of other organs (81–85). The clinical presentation of Aspergillosis includes colonization, ulcerative tracheobronchitis, pseudomembranous tracheobronchitis, and invasive pneumonia (81–83). Aspergillosis has also been associated with ulcerative lesions that occur primarily at the anastomotic site. Scarring associated with the healing of these lesions results in stricture formation (82,85). Aspergillosis has been strongly associated with endobronchial

293

abnormalities and narrowing of airway anastomoses (82). Treatment includes the administration of 200 to 400 mg of itraconazole daily, with or without the addition of nebulized amphotericin B (86,87,89). Systemic amphotericin B and its lipid formulations have been the mainstay of therapy, but their use has been limited by toxic effects (84). Caspofungin, a new class of antifungal medications, has good activity against Aspergillus and offers an additional treatment option (88,89). CMV is second to bacterial organisms as a cause of infection among recipients of lung transplants and has been frequently associated with the development of BOS (33,73,90,91). The prevalence of CMV infection in the United States is high; results of serologic testing for CMV indicate that more than half of the adults in the United States have been infected with this virus sometime during their lives. The risk of CMV disease after transplantation depends on the CMV status of the donor and the recipient. The risk is greatest when the recipient is seronegative and the donor is seropositive; CMV disease occurs among as many as 90% of patients in this group. When both the donor and the recipient are seronegative, the incidence of CMV disease is less than 15% (33,90–92). Seropositive recipients can experience reactivation of disease, particularly when their immunosuppressive regimen is augmented. Infection will usually occur between two weeks and three months after transplantation. The most common presentation of CMV disease is pneumonitis; however, gastroenteritis, hepatitis, colitis, and bone marrow depression have also been documented. The diagnosis of CMV infection requires identification of the virus by culture, shell-vial assay, or polymerase chain reaction amplification of viral DNA from blood, urine, or bronchioalveolar lavage fluid. Definitive diagnosis of the disease requires histologic evidence of the characteristic cells in cytology or biopsy specimens (33). Ganciclovir and its derivative, valganciclovir, are the drugs of choice for the prophylaxis and treatment of CMV disease, but guidelines for their use appear to be center specific. Ongoing studies are needed to determine the best prophylactic and treatment regimens for this disease.

COMPLICATIONS OF IMMUNOSUPPRESSION Standard immunosuppression for recipients of lung transplants is a triple-drug regimen that includes a calcineurin inhibitor (cyclosporine or tacrolimus), an inhibitor of purine biosynthesis (azathioprine or MMF) and corticosteroids. Acute rejection necessitates therapy with high doses of corticosteroids, whereas refractory acute rejection often requires lympholytic agents [rabbit ATG or horse antilymphocyte globulin (ALG)] or murine monoclonal antibody against the human CD3 T-cell antigen (OKT3). Serious toxic effects are associated with all immunosuppressive

294

Qi and Fertel

agents, and all patients must be diligently monitored for side effects and drug interactions. Both cyclosporine and tacrolimus can cause nephrotoxicity, neurotoxicity (including headache and tremors), and hypertension. Tacrolimus more commonly causes hyperglycemia, whereas cyclosporine more commonly causes hirsutism and gingival hyperplasia. Azathioprine can cause bone marrow depression and toxicity to the pancreas and liver, whereas MMF more commonly causes gastrointestinal side effects, particularly diarrhea. Complications related to steroid use are well known and include Cushing’s syndrome, hyperglycemia, hyperlipidemia, osteoporosis, peptic ulcer disease, and cataracts. OKT3 can cause a cytokinerelease syndrome with hypotension and pulmonary edema, whereas ATG and ALG can produce a serum sickness reaction. All immunosuppressive agents enhance the risk of opportunistic infections and malignancy such as PTLD (33,93).

LIVING-DONOR LUNG TRANSPLANTATION The limited number of cadaveric lung donors prompted Starnes et al. to initiate living-donor lung transplantation in 1993 (94). By December 2000, 139 such procedures had been performed (95). Two donors are required, one to provide a right lower lobe and the other to provide a left lower lobe. Donors must be aged between 18 and 55, be in good health, and have undergone no thoracic procedures on the side from which the lung is to be donated. Donors are matched on blood type and should be taller than the recipient. They must pass through rigorous psychosocial screening (96). Although for adults, the outcomes of living-donor lung transplantation are similar to those of cadaveric lung transplantation (97), living-donor lung transplantation has been associated with improved graft function and fewer cases of BOS among children (98). Studies of complication rates have therefore focused on donors. In one group, 61.3% of donors (38 of 62 patients) suffered postoperative complications; the most severe were pleural effusions requiring drainage, bronchial stump fistulas, phrenic nerve injury, atrial flutter necessitating ablation, and bronchial stricture (99). Clearly, more data are needed for evaluating morbidity rates among the donors before living-donor lung transplantation is universally accepted as an alternative to cadaveric lung transplantation.

CONCLUSION Lung transplantation is continuing to evolve as an acceptable option for patients with end-stage lung disease. The complication of anastomotic dehiscence, once thought to be the limiting factor in lung transplantation, has been overcome and has given way to problems related to IRI, hyperacute and acute rejection, airway dysfunction, infection, and bronchiolitis

obliterans. However, continued improvements in immunosuppressive regimens, antibiotic choices, crossmatching, and organ preservation techniques, as well as research focused on vascular endothelial injury and immune tolerance, should help reduce the complications of lung transplantation in the future.

REFERENCES 1. Hardy JD, Webb SR, Dalton ML Jr, Walker GR Jr. Lung homotransplantation in man. JAMA 1963; 186: 1065–1074. 2. Veith FJ. Lung transplantation. Surg Clin North Am 1978; 58:357–364. 3. Daly RC, McGregor CG. Surgical issues in lung transplantation: options, donor selection, graft preservation, and airway healing. Mayo Clin Proc 1997; 72: 79–84. 4. Reitz BA, Wallwork JL, Hunt SA, et al. Heart-lung transplantation: successful therapy for patients with pulmonary vascular disease. N Engl J Med 1982; 306: 557–564. 5. Toronto Lung Transplant Group. Unilateral lung transplantation for pulmonary fibrosis. N Engl J Med 1986; 314:1140–1145. 6. Patterson GA, Todd TR, Cooper JD, Pearson FG, Winton TL, Maurer J. Airway complications after double lung transplantation. Toronto Lung Transplant Group. J Thorac Cardiovasc Surg 1990; 99:14–21. 7. Daly RC, Tadjkarimi S, Khaghani A, Banner NR, Yacoub MH. Successful double-lung transplantation with direct bronchial artery revascularization. Ann Thorac Surg 1993; 56:885–892. 8. Cooper JD, Patterson GA, Grossman R, Maurer JR. Double-lung transplant for advanced chronic obstructive lung disease. Am Rev Respir Dis 1989; 139(2): 303–307. 9. Pasque MR, Cooper JD, Kaiser LR, Haydock DA, Triantafillou A, Trulock RJ. Improved technique for bilateral lung transplantation: rationale and initial clinical experience. Ann Thorac Surg 1990; 49:785–791. 10. Hosenpud JD, Bennett LE, Keck BM, Boucek MM, Novick RJ. The Registry of the International Society for Heart and Lung Transplantation: seventeenth official report-2000. J Heart Lung Transplant 2000; 19:909–931. 11. Maurer JR, Frost AE, Estenne M, Higenbottam T, Glanville AR. International guidelines for the selection of lung transplant candidates. The International Society for Heart and Lung Transplantation, the American Thoracic Society, the American Society of Transplant Physicians, the European Respiratory Society. J Heart Lung Transplant 1998; 17:703–709. 12. Detterbeck FC, Egan TM, Mill MR. Lung transplantation after previous thoracic surgical procedures. Ann Thorac Surg 1995; 60:139–143. 13. Dusmet M, Winton TL, Kesten S, Maurer J. Previous intrapleural procedures do not adversely affect lung transplantation. J Heart Lung Transplant 1996; 15: 249–254. 14. Schafers HJ, Wagner TO, Demertzis S, et al. Preoperative corticosteroids. A contraindication to lung transplantation? Chest 1992; 102:1522–1525.

Chapter 24: Complications of Lung Transplantation

15. Baz MA, Palmer SM, Staples ED, Greer DG, Tapson VF, Davis DD. Lung transplantation after long-term mechanical ventilation: results and 1-year follow-up. Chest 2001; 119:224–227. 16. Davis RD Jr., Pasque MK. Pulmonary transplantation. Ann Surg 1995; 221:14–28. 17. Tyan DB, Li VA, Czer L, Trento A, Jordan SC. Intravenous immunoglobulin suppression of HLA alloantibody in highly sensitized transplant candidates and transplantation with a histoincompatible organ. Transplantation 1994; 57:553–562. 18. Halldorsson AO, Kronon MT, Allen BS, Rahman S, Wang T. Lowering reperfusion pressure reduces the injury after pulmonary ischemia. Ann Thorac Surg 2000; 69:198–204. 19. Date H, Matsumura A, Manchester JK, Cooper JM, Lowry OH, Cooper JD. Changes in alveolar oxygen and carbon dioxide concentration and oxygen consumption during lung preservation. The maintenance of aerobic metabolism during lung preservation. J Thorac Cardiovasc Surg 1993; 105:492–501. 20. Wang LS, Nakamoto K, Hsieh CM, Miyoshi S, Cooper JD. Influence of temperature of flushing solution on lung preservation. Ann Thorac Surg 1993; 55:711–715. 21. Meyers BF, Sundt TM III, Henry S, et al. Selective use of extracorporeal membrane oxygenation is warranted after lung transplantation. J Thorac Cardiovasc Surg 2000; 120:20–26. 22. Zenati M, Pham SM, Keenan RJ, Griffith BP. Extracorporeal membrane oxygenation for lung transplant recipients with primary severe donor lung dysfunction. Transplant Int 1996; 9:227–230. 23. Vlasselaers D, Verleden GM, Meyns B, et al. Femoral venoarterial extracorporeal membrane oxygenation for severe reimplantation response after lung transplantation. Chest 2000; 118:559–561. 24. Choi JK, Kearns J, Palevsky HI, et al. Hyperacute rejection of a pulmonary allograft. Immediate clinical and pathologic findings. Am J Respir Crit Care Med 1999; 160:1015–1018. 25. Christie JD, Bavaria JE, Palevsky HI, et al. Primary graft failure following lung transplantation. Chest 1998; 114:51–60. 26. Frost AE, Jammal CT, Cagle PT. Hyperacute rejection following lung transplantation. Chest 1996; 110:559–562. 27. Zander DS, Baz MA, Visner GA, et al. Analysis of early deaths after isolated lung transplantation. Chest 2001; 120:225–232. 28. King RC, Binns OA, Rodriguez F, et al. Reperfusion injury significantly impacts clinical outcome after pulmonary transplantation. Ann Thorac Surg 2000; 69:1681–1685. 29. Yousem SA, Berry GJ, Cagle PT, et al. Revision of the 1990 working formulation for the classification of pulmonary allograft rejection: Lung Rejection Study Group. J Heart Lung Transplant 1996; 15:1–15. 30. Anderson DC, Glazer HS, Semenkovich JW, et al. Lung transplant edema: chest radiography after lung transplantation—the first 10 days. Radiology 1995; 195:275–281. 31. Haydock DA, Trulock EP, Kaiser LR, Knight SR, Pasque MK, Cooper JD. Management of dysfunction in the transplanted lung: experience with 7 clinical cases. Washington University Lung Transplant Group. Ann Thorac Surg 1992; 53:635–641.

295

32. Paradis IL, Duncan SR, Dauber JH, Yousem S, Hardesty R, Griffity B. Distinguishing between infection, rejection, and the adult respiratory distress syndrome after human lung transplantation. J Heart Lung Transplant 1992; 11:S232–S236. 33. Trulock EP. Lung transplantation. Am J Respir Crit Care Med 1997; 155:789–818. 34. Chapparro C, Chamberlain D, Maurer J, De Hoyos A, Winton T, Kesten S. Acute lung injury in lung allografts. J Heart Lung Transplant 1995; 14:267–273. 35. Khan SU, Salloum J, O’Donovan PB, et al. Acute pulmonary edema after lung transplantation: the pulmonary reimplantation response. Chest 1999; 116: 187–194. 36. Aeba R, Griffith BP, Kormos RL, et al. Effect of cardiopulmonary bypass on early graft dysfunction in clinical lung transplantation. Ann Thorac Surg 1994; 57:715–722. 37. Date H, Triantafillou AN, Trulock EP, Pohl MS, Cooper JD, Patterson GA. Inhaled nitric oxide reduces human lung allograft dysfunction. J Thorac Cardiovasc Surg 1996; 111:913–919. 38. Glassman LR, Keenan RJ, Fabrizio MC, et al. Extracorporeal membrane oxygenation as an adjunct treatment for primary graft failure in adult lung transplant recipients. J Thorac Cardiovasc Surg 1995; 110:723–727. 39. Whyte RI, Deeb GM, McCurry KR, Anderson HL III, Bolling SF, Bartlett RH. Extracorporeal life support after heart or lung transplantation. Ann Thorac Surg 1994; 58:754–759. 40. Kshettry VR, Kroshus TJ, Hertz MI, Hunter DW, Shumway SJ, Bolman RM III. Early and late airway complications after lung transplantation: incidence and management. Ann Thorac Surg 1997; 63:1576–1583. 41. Date H, Trulock EP, Arcidi JM, Sundaresan S, Cooper JD, Patterson GA. Impaired airway healing after lung transplantation. An analysis of 348 bronchial anastomoses. J Thorac Cardiovasc Surg 1995; 110:1424–1433. 42. Kirk AJ, Conacher ID, Corris PA, Ashcroft T, Dark JH. Successful surgical management of bronchial dehiscence after single-lung transplantation. Ann Thorac Surg 1990; 49:147–149. 43. Shennib H, Massard G. Airway complications in lung transplantation. Ann Thorac Surg 1994; 57:506–511. 44. Herrera JM, McNeil KD, Higgens RS, et al. Airway complications after lung transplantation: treatment and long-term outcome. Ann Thorac Surg 2001; 71: 989–994. 45. Daly RC, McGregor CG. Routine immediate direct bronchial artery revascularization for single-lung transplantation. Ann Thorac Surg 1994; 57:1446–1452. 46. Couraud L, Baudet E, Martigne C, et al. Bronchial revascularization in double-lung transplantation: a series of 8 patients. Bordeaux Lung and Heart-Lung Transplant Group. Ann Thorac Surg 1992; 53:88–94. 47. Norgaard MA, Olsen PS, Svendsen UG, Pettersson G. Revascularization of the bronchial arteries in lung transplantation: an overview. Ann Thorac Surg 1996; 62:1215–1221. 48. Susanto I, Peters JI, Levine SM, Sako EY, Anzueto A, Bryan CL. Use of balloon-expandable metallic stents in the management of bronchial stenosis and bronchomalacia after lung transplantation. Chest 1998; 114: 1330–1335.

296

Qi and Fertel

49. Trulock EP III. Lung transplantation for COPD. Chest 1998; 113:269S–276S. 50. Mal H, Brugiere O, Sleiman C, et al. Morbidity and mortality related to the native lung in single lung transplantation for emphysema. J Heart Lung Transplant 2000; 19:220–223. 51. Weill D, Torres F, Hodges TN, Olmos JJ, Zamora MR. Acute native lung hyperinflation is not associated with poor outcomes after single lung transplant for emphysema. J Heart Lung Transplant 1999; 18:1080–1087. 52. Yonan NA, el-Gamel A, Egan J, Kakadellis J, Rahman A, Deiraniya AK. Single lung transplantation for emphysema: predictors for native lung hyperinflation. J Heart Lung Transplant 1998; 17:192–201. 53. Kroshus TJ, Bolman RM III, Kshettry VR. Unilateral volume reduction after single-lung transplantation for emphysema. Ann Thorac Surg 1996; 62:363–368. 54. Berkowitz N, Schulman LL, McGregor C, Markowitz D. Gastroparesis after lung transplantation. Potential role in postoperative respiratory complications. Chest 1995; 108:1602–1607. 55. Smith PC, Slaughter MS, Petty MG, Shumway SJ, Kshettry VR, Bolman RM III. Abdominal complications after lung transplantation. J Heart Lung Transplant 1995; 14:44–51. 56. Hoekstra HJ, Hawkins K, de Boer WJ, Rottier K, van der Bij W. Gastrointestinal complications in lung transplant survivors that require surgical intervention. Br J Surg 2001; 88:433–438. 57. Arcasoy SM, Kotloff RM. Lung transplantation. N Engl J Med 1999; 340:1081–1091. 58. Akindipe OA, Faul JL, Vierra MA, Triadafilopoulos G, Theodore J. The surgical management of severe gastroparesis in heart/lung transplant recipient. Chest 2000; 117:907–910. 59. Bando K, Paradis IL, Komatsu K, et al. Analysis of time-dependent risks for infection, rejection, and death after pulmonary transplantation. J Thorac Cardiovasc Surg 1995; 109:49–59. 60. Wisser W, Wekerle T, Zlabinger G, et al. Influence of human leukocyte antigen matching on long-term outcome after lung transplantation. J Heart Lung Transplant 1996; 15:1209–1216. 61. Schulman LL, Weinberg AD, McGregor C, Galantowicz ME, Suciu-Foca NM, Itescu S. Mismatches at the HLADR and HLA-B loci are risk factors for acute rejection after lung transplantation. Am J Respir Crit Care Med 1998; 157:1833–1837. 62. Millet B, Higenbottam TW, Flower CD, Stewart S, Wallwork J. The radiographic appearances of infection and acute rejection of the lung after heart-lung transplantation. Am Rev Respir Dis 1989; 140:62–67. 63. Otulana BA, Higenbottam T, Ferrari L, Scott J, Igboaka G, Wallwork J. The use of home spirometry in detecting acute lung rejection and infection following heart-lung transplantation. Chest 1990; 97:353–357. 64. Becker FS, Martinez FJ, Brunsting LA, Deeb GM, Flint A, Lynch JP III. Limitations of spirometry in detecting rejection after single-lung transplantation. Am J Respir Crit Care Med 1994; 150:159–166. 65. Guilinger RA, Paradis IL, Dauber JH, et al. The importance of bronchoscopy with transbronchial lung biopsy and bronchoalveolar lavage in the management of lung transplant recipients. Am J Respir Crit Care Med 1995; 152:2037–2043.

66. Sibley RK, Berry GJ, Tazelaar HD, et al. The role of transbronchial lung biopsies in the management of lung transplant recipients. J Heart Lung Transplant 1993; 12:308–324. 67. Fertel DP, Qi XS, Pham SM. Treatment strategies for obliterate bronchiolitis. Curr Opinion in Organ Transplant 2001; 6:231–238. 68. Shennib H, Massard G, Reynaud M, Noirclerc M. Efficacy of OKT3 therapy for acute rejection in isolated lung transplantation. J Heart Lung Transplant 1994; 13:514–519. 69. Shennib H, Mercado M, Nguyen D, et al. Successful treatment of steroid-resistant double-lung allograft rejection with Orthoclone OKT3. Am Rev Respir Dis 1991; 144:224–226. 70. Bando K, Paradis IL, Similo S, et al. Obliterative bronchiolitis after lung and heart-lung transplantation: an analysis of risk factors and management. J Thorac Cardiovasc Surg 1995; 110:4–14. 71. Scott JP, Higenbottam TW, Sharples L, et al. Risk factors for obliterative bronchiolitis in heart-lung transplant recipients. Transplantation 1991; 51:813–817. 72. Reichenspurner H, Girgis RE, Robbins RC, et al. Stanford experience with obliterative bronchiolitis after lung and heart-lung transplantation. Ann Thorac Surg 1996; 62:1467–1473. 73. Heng D, Sharples LD, McNeil K, Stewart S, Wreghitt T, Wallwork J. Bronchiolitis obliterans syndrome: incidence, natural history, prognosis, and risk factors. J Heart Lung Transplant 1998; 17:1255–1263. 74. Girgis RE, Tu I, Berry GJ, et al. Risk factors for the development of obliterative bronchiolitis after lung transplantation. J Heart Lung Transplant 1996; 15:1200–1208. 75. Boehler A, Estenne M. Obliterative bronchiolitis after lung transplantation. Curr Opin Pulm Med 2000; 6:133–139. 76. Soyer P, Devine N, Frachon I, et al. Computed tomography of complications of lung transplantation. Eur Radiol 1997; 7:847–853. 77. Kramer MR, Stoehr C, Whang JL, et al. The diagnosis of obliterative bronchiolitis after heart-lung transplantation: low yield of transbronchial lung biopsy. J Heart Lung Transplant 1993; 12:675–681. 78. Cooper JD, Billingham M, Egan T, et al. A working formulation for the standardization of nomenclature and for clinical staging of chronic dysfunction in lung allografts. International Society for Heart and Lung Transplantation. J Heart Lung Transplant 1993; 12:713–716. 79. Horvath J, Dummer S, Loyd J, Walker B, Merrill WH, Frist WH. Infection in the transplanted and native lung after single lung transplantation. Chest 1993; 104:681–685. 80. Flume PA, Egan TM, Paradowski LJ, Detterbeck FC, Thompson JT, Yankaskas JR. Infectious complications of lung transplantation. Impact of cystic fibrosis. Am J Respir Crit Care Med 1994; 149:1601–1607. 81. Yeldandi V, Laghi F, McCabe MA, et al. Aspergillus and lung transplantation. J Heart Lung Transplant 1995; 14:883–890. 82. Nathan SD, Shorr AF, Schmidt ME, Burton NA. Aspergillus and endobronchial abnormalities in lung transplant recipients. Chest 2000; 118:403–407. 83. Westney GE, Kesten S, De Hoyos A, Chapparro C, Winton T, Maurer JR. Aspergillus infection in single and double lung transplant recipients. Transplantation 1996; 61:915–919.

Chapter 24: Complications of Lung Transplantation

84. Nunley DR, Ohori P, Grgurich WF, et al. Pulmonary aspergillosis in cystic fibrosis lung transplant recipients. Chest 1998; 114:1321–1329. 85. Kramer MR, Denning DW, Marshall SE, et al. Ulcerative tracheobronchitis after lung transplantation. A new form of invasive aspergillosis. Ann Rev Respir Dis 1991; 144:552–526. 86. Boettcher H, Bewig B, Hirt SW, Moller F, Cremer J. Topical amphotericin B application in severe bronchial aspergillosis after lung transplantation: report of experiences in 3 cases. J Heart Lung Transplant 2000; 19:1224–1227. 87. Monforte V, Roman A, Gavalda J, et al. Nebulized amphotericin B prophylaxis for Aspergillus infection in lung transplantation: study of risk factors. J Heart Lung Transplant 2001; 20:1274–1281. 88. Andriole VT. The 1998 Garrod lecture. Current and future antifungal therapy: new targets for antifungal agents. J Antimicrob Chemother 1999; 44:151–162. 89. Wood DE, Raghu G. Lung transplantation. Part II. Postoperative management and results. West J Med 1997; 166:45–55. 90. Keenan RJ, Lega ME, Dummer JS, et al. Cytomegalovirus serologic status and postoperative infection correlated with risk of developing chronic rejection after pulmonary transplantation. Transplantation 1991; 51:433–438. 91. Duncan SR, Paradis IL, Yousem SA, et al. Sequelae of cytomegalovirus pulmonary infections in lung allograft recipients. Am Rev Respir Dis 1992; 146:1419–1425.

297

92. Griffith BP, Bando K, Armitage JM, et al. Lung transplantation at the University of Pittsburgh. Clin Transpl 1992; 13:149–159. 93. Hausen B, Morris RE. Review of immunosuppression for lung transplantation. Novel drugs, new uses for conventional immunosuppressants, and alternative strategies. Clin Chest Med 1997; 18:353–366. 94. Starnes VA, Barr ML, Cohen RG. Lobar transplantation. Indications, techniques, and outcome. J Thorac Cardiovasc Surg 1994; 108:403–411. 95. DeMeo DL, Ginns LC. Clinical status of lung transplantation. Transplantation 2001; 72:1713–1724. 96. Barr ML, Baker CJ, Schenkel FA, et al. Living donor lung transplantation: selection, technique, and outcome. Transplant Proc 2001; 33:3527–3532. 97. Starnes VA, Woo MS, MacLaughlin EF, et al. Comparison of outcomes between living donor and cadaveric lung transplantation in children. Ann Thorac Surg 1999; 68:2279–2284. 98. Woo MS, MacLaughlin EF, Horn MV, Szmuszkovicz JR, Barr ML, Starnes VA. Bronchiolitis obliterans is not the primary cause of death in pediatric living donor lobar lung transplant recipients. J Heart Lung Transplant 2001; 20:491–496. 99. Battafarano RJ, Anderson RC, Meyers BF, et al. Perioperative complications after living donor lobectomy. J Thorac Cardiovasc Surg 2000; 120:909–915.

PART V Cardiovascular Surgery Complications

25 Complications After Cardiopulmonary Resuscitation and Cardiac Arrest Abhijit S. Pathak, Amy J. Goldberg, and Robert F. Buckman, Jr. Department of Surgery, Temple University School of Medicine, Philadelphia, Pennsylvania, U.S.A.

Cardiopulmonary resuscitation (CPR) is typically performed in one of three locations: the emergency department, the operating room, or the intensive care unit (ICU). The surgeon’s initial response to cardiac arrest depends on the presumed cause of the arrest, on whether the arrest was witnessed, and on the patient’s clinical status (e.g., whether the arrest occurred intraoperatively while the patient’s abdomen or chest was open). Whenever cardiac arrest occurs, the well-known airway, breathing, circulation (ABC) algorithm must be emphasized. Furthermore, because the surgeon is most likely to encounter cardiac arrest in a hospital environment, it is likely that adequate monitoring will be in place and that support personnel competent to initiate CPR will be available. Only rarely will surgeons be involved in treating a patient who suffers sudden cardiac death. With a few exceptions, such as trauma, sudden massive pulmonary thromboembolism, or massive myocardial infarction, most episodes of cardiac arrest with which surgeons are involved are associated with serious multiorgan dysfunction.

CAUSES OF AND FACTORS UNDERLYING CARDIAC ARREST If cardiac arrest is to be successfully managed, certain underlying factors must be recognized and corrected. In general, cardiac compromise or arrest can be broadly categorized as due to nontraumatic or traumatic causes. Nontraumatic arrest can be further categorized as resulting from respiratory factors, cardiac factors, neurologic factors, or metabolic or electrolyte disturbances. Table 1 outlines the causes of nontraumatic arrest, and Table 2 outlines the causes of traumatic arrest.

PATHOPHYSIOLOGY OF CARDIAC ARREST Understanding the pathophysiology of the precipitating event will help guide and direct the management

of cardiac arrest during the resuscitation phase and the postresuscitation period. Cardiac arrest results in cessation of blood flow; however, the vulnerability of organs to ischemic injury differs. The central nervous system, particularly the brain, is the most susceptible to such injury; the most vulnerable areas of the brain are the cerebral cortex, the hippocampus, and the cerebellum. Irreversible brain damage can be expected after five minutes of normothermic cardiac arrest. If the blood flow to the brain is not restored within ten minutes, restoration of neurologic function rarely occurs. The heart is the second most vulnerable organ to ischemia; the endocardium is more sensitive than the epicardium. The kidneys, gastrointestinal tract, and musculoskeletal system are much more tolerant to the disruption of blood blow and can tolerate long periods of normothermic ischemia (up to one hour) without permanent damage, if adequate reperfusion is reestablished. Cessation of cardiac function results from one of three causes: ventricular fibrillation (VF) or pulseless ventricular tachycardia (VT); ventricular asystole; or pulseless electrical activity (PEA) (1).

Ventricular Fibrillation or Pulseless Ventricular Tachycardia VF or pulseless VT usually results from a primary cardiac event such as acute myocardial infarction or ischemia. The presence of antecedent multifocal premature ventricular contractions may serve as a warning sign for these serious conditions. Certain electrolyte disturbances, such as hypokalemia, hypomagnesemia, or hypocalcemia, may complicate or contribute to this scenario (Table 1). In cases of traumatic cardiac arrest, ventricular irritability may suggest air embolism or cardiac compression caused by a tension pneumothorax or pericardial tamponade. Electrocution with alternating current in the range of 100 mA to 1 A can also cause VF (Table 2).

300

Pathak et al.

Table 1 Factors Causing or Contributing to Nontraumatic Cardiac Arrest Respiratory factors Airway obstruction CNS injury Foreign body Infection Trauma Tumor Insufficient ventilation CNS injury Neuromuscular disease Drugs Hypoxemia or pulmonary dysfunction COPD Asthma Pulmonary edema Venous thromboembolism Pneumonia Cardiovascular factors Cardiac factors Acute coronary occlusion Coronary artery disease Drugs Reduced cardiac output Cardiomyopathy Valvular or structural abnormalities Tension pneumothorax Pericardial tamponade Pulmonary embolism Circulatory factors Hypovolemia or hemorrhage Vasodilatory shock Sepsis Neurogenic shock

Neurologic factors Central depression Stroke Anesthesia Drugs or toxins Metabolic or electrolyte factors Acidosis Alkalosis Hypokalemia or hyperkalemia Hypomagnesemia or hypermagnesemia Hypocalcemia Hypothermia Drugs or toxins Beta-blockers Calcium channel blockers Digoxin Antiarrhythmics Tricyclic antidepressants Carbon monoxide Cyanide Cocaine

Abbreviations: CNS, central nervous system; COPD, chronic obstructive pulmonary disease.

Ventricular Asystole Asystole usually results from cardiac arrest while the heart was in diastole. Asystole may be the final outcome of a process beginning with bradycardia in patients with hypoxemia caused by respiratory failure, by a vasovagal event, or by a metabolic disturbance such

as hyperkalemia. Furthermore, asystole may be the result of exsanguination: in this event, the progression from tachycardia to bradycardia and PEA finally degenerates to asystole. Electrocution with alternating current of more than 10 A can also result in ventricular asystole.

Pulseless Electrical Activity PEA occurs when a heart rhythm is present but cardiac output is absent. Common causes of PEA include severe hypoxia, hypovolemia, hypothermia, acidosis, tension pneumothorax, and pericardial tamponade.

PHYSIOLOGY OF STANDARD CLOSED-CHEST CARDIOPULMONARY RESUSCITATION Cardiac Pump Model Kouwenhoven et al. (2) first suggested that CPR works as a cardiac pump by squeezing the heart between the sternum and the spine. Each chest compression results in systole, with the left ventricle compressed and blood propelled forward. Because the cardiac valves operate in only one direction, prograde flow into the arterial circulation is guaranteed. The relaxation phase of CPR, which allows the sternum to return to its normal position, corresponds to diastole, during which intracardiac pressures fall, the atrioventricular valves open, and venous return occurs.

Thoracic Pump Model Most physiologists favor the thoracic pump model of CPR (3). In this model, forward flow is generated by an arteriovenous pressure gradient that is established by chest compression. External compression of the chest generates an increase in thoracic pressure that is transmitted throughout the thorax, including the heart, the aorta, and the great veins. According to this model, the mitral and tricuspid valves are incompetent and no significant atrioventricular or ventriculoaortic pressure gradients are present.

Perfusion During Cardiopulmonary Resuscitation Table 2 Factors Causing or Contributing to Traumatic Cardiac Arrest Exsanguination Hypovolemia or hemorrhage Cardiac injury Pericardial tamponade Tension pneumothorax Air embolism Airway obstruction Extrinsic compression CNS depression Direct injury Intraluminal or oropharyngeal bleeding Hemic drowning Drowning or near-drowning Severe brain injury Spinal cord injury Electrocution Hypothermia Abbreviation: CNS, central nervous system.

Closed-chest CPR results in only limited perfusion of vital organs. Cardiac output is believed to be no more than 25% of normal and cerebral blood flow is believed to be only about 15% of normal (3,4). Furthermore, coronary blood flow during standard CPR may be as low as 5% of normal (4).

MANAGEMENT OF CARDIAC ARREST The American Heart Association has established standards for basic life support, advanced cardiac life support, and CPR (5,6). These standards and guidelines are readily available and will not be discussed here. The management of cardiac arrest begins with the ABC algorithm mentioned above. However, there is one instance in which this orderly rule may be circumvented: when patients who are being monitored

Chapter 25:

Complications After Cardiopulmonary Resuscitation and Cardiac Arrest

301

experience VF, immediate defibrillation takes precedence over airway management, pharmacotherapy, and CPR because hypoxemia and prolonged fibrillation have not yet developed. As stated above, closed-chest compressions provide only a fraction of the normal cardiac output, even under optimal conditions. If standard closed-chest CPR fails to provide effective circulation, the surgeon may consider using open-chest cardiac massage. Under certain circumstances, resuscitative thoracotomy and open-chest cardiac massage may be the only means of effective resuscitation after nontraumatic cardiac arrest. Because most physiologists subscribe to the thoracic pump model of CPR, effective circulation may not be provided when the transmission of pressure throughout the thorax is impossible, such as when the abdomen or chest is open. Especially for trauma victims, the situation may be compounded by hypovolemia or an unstable chest wall, which renders conventional CPR ineffective. These conditions are most commonly encountered in the operating room, during laparotomy or thoracotomy, and in the ICU, where patients may have an open abdomen because a laparotomy has been performed to treat injuries. In such circumstances, open cardiac massage via resuscitative thoracotomy or a transdiaphragmatic route (if done during laparotomy) may be the only hope for survival. Direct cardiac massage through a minimally invasive approach rather than an open thoracotomy has been reported (7,8). This procedure, called minimally invasive directed cardiac massage, involves inserting a heart-contracting padded baseplate, connected to a handle, through a 2- to 3-cm incision in the left parasternal area. The handle remains outside the chest, and the baseplate is positioned directly on the ventricles with the pericardium intact. Manual decompression of the device compresses the heart and causes an artificial systole. In a swine model of cardiac arrest (7), this technique provided coronary and cerebral perfusion similar to that achieved using standard open-chest cardiac massage. A European pilot study (8) using this technique in the prehospital setting demonstrated promising results. The management scheme described above is appropriate in cases of nontraumatic cardiac arrest. In cases of trauma, resuscitative thoracotomy is vital for patients who are in extremis or whose condition deteriorates so that cardiac arrest appears imminent. For trauma victims, thoracotomy may be necessary to relieve pericardial tamponade, control thoracic bleeding, control air embolism, allow open cardiac massage, provide temporary aortic occlusion so as to maximize cerebral and coronary perfusion, and limit infradiaphragmatic bleeding (9).

et al. (10). In most instances, the patient is placed supine with the left arm raised above the head. For men, a standard left anterolateral thoracotomy incision is begun just lateral to the left border of the sternum and carried in a straight line laterally to a point just below the nipple (Figs. 1 and 2). For women, the incision is placed in the inframammary crease, and the breast is held under cephalad traction. The chest is entered through the fifth intercostal space. During most emergency thoracotomy procedures, the pericardium is opened with an incision that begins anterior to the phrenic nerve and proceeds longitudinally so as to avoid transecting the nerve. The pericardium must be opened widely so that the heart can be delivered through the incision (Fig. 3). An inadequate pericardiotomy may impede effective cardiac output during open massage or result in cardiac arrest because inflow is occluded. Manual cardiac massage should be performed with both hands. In this technique, the hands should be slightly cupped and placed on the anterior and posterior surfaces of the heart. The ventricles are cyclically compressed as shown in Figure 4. The fingertips should be flat against the epicardial surface. Most studies have indicated that the optimal rate of manual cardiac massage is approximately 60 beats/min. Aortic cross clamping is generally performed at the outset of cardiac massage; its purpose is to help maximize cerebral and coronary flow. The first step in clamping the aorta is to locate the descending vessel by anteriorly retracting the left lung with the left hand, a maneuver that is facilitated if the lung is deflated. Direct visualization of the aorta is ideal but can be very difficult to achieve; blind dissection may be necessary. The surgeon should run the fingers over the anterior spine until the space between the spine and

TECHNIQUE OF RESUSCITATIVE THORACOTOMY AND OPEN CARDIAC MASSAGE

Figure 1 The standard skin- and soft-tissue incision for a left anterolateral resuscitative thoracotomy extends in a straight line from the left border of the sternum to the posterior axillary line. The incision may be extended into the right chest at the same level or one intercostal space higher.

The technique of resuscitative thoracotomy and open cardiac massage has been reviewed by Buckman

302

Pathak et al.

Figure 2 Once the plane of the ribs and intercostal muscles has been reached, the intercostal muscle is thinned with a knife, but the pleura is entered with a finger or blunt-tipped scissors to avoid iatrogenic laceration of the heart or the lungs. The intercostal space is then opened widely with scissors.

aorta can be palpated; this space should be entered by blunt dissection with the surgeon’s fingers. The pleura is opened just anterior and posterior to the aorta at the site of intended occlusion. The jaws of the clamp are then passed through the created apertures to securely

Figure 4 Bimanual cardiac massage is performed via a left anterolateral thoracotomy. The fingertips must be kept flat on the cardiac surface to avoid iatrogenic cardiac penetration. Excessive traction on the heart should be avoided because it can result in obstruction of venous inflow.

grasp the aorta, as shown in Figure 5. Complete occlusion of the aorta must be ensured.

COMPLICATIONS OF CARDIOPULMONARY RESUSCITATION Standard Closed-Chest Cardiopulmonary Resuscitation Most complications after closed-chest CPR are related to thoracic wall damage and include rib or sternal fractures and costochondral separation. These injuries can lead to more severe problems such as pneumothorax, hemothorax, cardiac injury, and even aortic injury. The thorax is not the only cavity at risk of damage from CPR. Compression of the liver against the xiphoid process can damage that organ (11). In addition, gastric injury, splenic injury (12), or fat embolism may occur. It is important to remember that an endotracheal tube may become dislodged or malpositioned by closed-chest compression. Thus, chest radiographs should be obtained as soon as possible after resuscitation. The true incidence of complications after CPR may not be accurately estimated or known because the primary event that caused the cardiac arrest usually causes death.

Open-Chest Cardiac Massage and Thoracotomy

Figure 3 In cases of cardiac tamponade, the pericardium is initially nicked with a knife anterior to the phrenic nerve, as shown in the inset. The pericardium is then widely opened with scissors with care to avoid injury to the heart.

When an emergency thoracotomy is performed, technical complications can occur. Iatrogenic injuries to the lung, pericardium, heart, phrenic nerves, aorta, esophagus, and chest wall have been described (13). The standard approach for left anterolateral thoracotomy, as noted above, is through the fifth

Chapter 25:

Complications After Cardiopulmonary Resuscitation and Cardiac Arrest

303

resuscitation, the patient should be taken to the operating room for hemostasis, wound irrigation, and closure. The most serious complication of either closedchest or open-chest CPR, of course, is failure to resuscitate. The rates of successful resuscitation vary depending on the cause of cardiac arrest but are, in general, poor. Investigations continue in an attempt to improve the outcome of CPR (14).

MANAGEMENT AFTER SUCCESSFUL CARDIOPULMONARY RESUSCITATION

Figure 5 Aortic cross-clamping is carried out after anterior retraction of the left lung and penetration of the mediastinal pleura, so that the jaws of the vascular clamp securely grip the aortic adventitia.

intercostal space below the nipple of male patients and in the inframammary crease of female patients. Exposure can be limited if the skin incision is improperly placed or if the correct intercostal space is not entered. Bleeding from the chest wall, especially from the internal mammary vessels, can be troublesome, and these vessels should be ligated if they are injured. During attempts at occluding the descending thoracic aortic, damage to the aorta, the esophagus, or the intercostal branches of the aorta may occur (13). Cardiac injuries can occur during an open cardiac massage. Perforation of the right ventricle is a serious complication and is usually caused by incorrectly applied cardiac compression. This lethal complication can be avoided if compression is performed with both hands and with the digits flat rather than curled. The right ventricle is particularly susceptible to injury because it is thinner than the left ventricle and is often distended as a result of fluid resuscitation (1). If perforation occurs, it should be repaired with sutures buttressed with pledgets. As a temporary measure, skin staples can be applied to the perforation until a definitive repair can be performed. Under these circumstances, the right ventricle is usually very friable and repair can be difficult. Infection and bleeding are the two most serious concerns after emergency thoracotomy. After successful

Once spontaneous circulation has resumed, the underlying cause of arrest should be identified and treated and the adequacy of tissue perfusion should be assured and maintained. Cardiovascular and hemodynamic dysfunctions are common after cardiac arrest; such dysfunctions include hypovolemic shock, cardiogenic shock, and the systemic inflammatory response syndrome (6). Invariably, some form of cardiovascular dysfunction is present, and normal cardiac function may not return for 12 to 24 hours (6). The primary goal of management after CPR is reestablishing global and regional perfusion of tissues. Traditional end points, such as restoration of blood pressure, may not adequately reflect peripheral organ perfusion. Regional tissue malperfusion may exist particularly in the splanchnic bed, and this condition is believed to contribute to the multiple-organ dysfunction syndrome. Physicians should strongly consider the use of invasive hemodynamic monitoring, such as pulmonary artery catheterization, to guide therapy after CPR. Neurologic impairment is common among survivors of cardiac arrest; approximately 80% of patients remain comatose for various time periods (15). As many as 40% of survivors enter a persistent vegetative state and 79% die within one year (16). This generally poor outcome after cardiac arrest has prompted many physicians to attempt to develop a means of predicting outcome during the early postresuscitative phase. When coma lasts for more than six hours after CPR, the prognosis for full neurologic recovery is poor. The prognosis worsens when the coma persists for more than three days; in such cases, patients rarely survive without severe disability (16). The Glasgow Coma Scale score is the most common objective method of monitoring the neurologic status and progression of coma. A recent meta-analysis (17) concluded that somatosensory-evoked potentials are useful in predicting the outcome after severe brain injury and are particularly useful in predicting poor outcome: the false-positive rate was less than 0.5%. However, no single variable or technique has been found to be useful in predicting outcome.

CONCLUSION Despite significant advances in acute care medicine, little advancement has been made in the care of the

304

Pathak et al.

cardiac arrest patient. Little improvement to complete recovery (hospital discharge to independent living) has occurred. Perhaps a complete reevaluation of CPR, questioning every step that is currently a protocol, needs to be done to alter that uniformly poor outcome.

REFERENCES 1. Greenfield LJ. Complications after cardiopulmonary resuscitation. In: Greenfield LJ, ed. Complications in Surgery and Trauma. 2d ed. Philadelphia: Lippincott, 1984:351–358. 2. Kouwenhoven WB, Jude JR, Knickerbocker CG. Closedchest cardiac massage. JAMA 1960; 173:1064–1067. 3. Niemann JT. Cardiopulmonary resuscitation. N Engl J Med 1992; 327(15):1075–1080. 4. Ditchey RV, Winkler JV, Rhodes CA. Relative lack of coronary blood flow during closed-chest resuscitation in dogs. Circulation 1982; 66:297–302. 5. Kern KB, Halperin HR, Field J. New guidelines for cardiopulmonary resuscitation and emergency cardiac care: changes in the management of cardiac arrest. JAMA 2001; 285(10):1267–1269. 6. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 6: advanced cardiovascular life support: section 8: postresuscitation care. The American Heart Association in collaboration with the International Liaison Committee on Resuscitation. Circulation 2000; 102(suppl 8):I166–I171. 7. Buckman RF Jr, Badellino MM, Mauro LH, et al. Direct cardiac massage without major thoracotomy: feasibility and systemic blood flow. Resuscitation 1995; 29(3): 237–248.

8. Rozenberg A, Incagnoli P, Delpech P, et al. Prehospital use of minimally invasive direct cardiac massage (MIDCM): a pilot study. Resuscitation 2001; 50(3):257–262. 9. Biffl WL, Moore EE, Johnson JL. Emergency department thoracotomy. In: Moore EE, Feliciano DV, Mattox KL, eds. Trauma. 5th ed. New York: McGraw-Hill, 2004: 239–253. 10. Buckman RF, Ballard RB, Eynon CA. Resuscitative thoracotomy for trauma: critical techniques. Trauma Q 1995; 12(2):105–132. 11. Adler SN, Klein RA, Pellecchia C, Lyon DT. Massive hepatic hemorrhage associated with cardiopulmonary resuscitation. Arch Intern Med 1983; 143:813–814. 12. Fitchet A, Neal R, Bannister P. Lesson of the week: splenic trauma complicating cardiopulmonary resuscitation. BMJ 2001; 322:480–481. 13. Ivatury R. Cardiac injuries and resuscitative thoracotomy. In: Maull KI, Rodriguez A, Wiles CE, eds. Complications in Trauma and Critical Care. Philadelphia: W.B. Saunders, 1996:279–288. 14. Thel MC, O’Connor CM. Cardiopulmonary resuscitation: historical perspective to recent investigations. Am Heart J 1999; 137(1):39–48. 15. Madl C, Kramer L, Domanovits H, et al. Improved outcome prediction in unconscious cardiac arrest survivors with sensory evoked potentials compared with clinical assessment. Crit Care Med 2000; 28(3): 721–726. 16. Edgren E, Hedstrand U, Kelsey S, Sutton-Tyrell K, Safar P. Assessment of neurological prognosis in comatose survivors of cardiac arrest. BRCT I Study Group. Lancet 1994; 343:1055–1059. 17. Carter BG, Butt W. Review of the use of somatosensory evoked potentials in the prediction of outcome after severe brain injury. Crit Care Med 2001; 29(1):178–186.

26 Complications of Vascular Surgery Frank B. Pomposelli, Jr. and Allen D. Hamdan Harvard Medical School and Division of Vascular Surgery, Beth Israel Deaconess Medical Center, Boston, Massachusetts, U.S.A. Malachi G. Sheahan Division of Vascular Surgery, Louisiana State University School of Medicine, New Orleans, Louisiana, U.S.A. Guatam V. Shrikhande Department of Surgery, Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, Massachusetts, U.S.A.

4.

Advancements in surgical techniques, better instruments and graft materials, and substantial improvements in diagnostic imaging have made it possible for vascular surgeons to execute vascular surgical procedures with a high likelihood of success in most circumstances. In addition, a better understanding of how to treat patients before, during, and after vascular surgery has contributed greatly to a reduction in the morbidity and mortality rates associated with these operations. Nonetheless, procedurerelated and systemic complications continue to occur and can probably never be totally eliminated because most patients undergoing vascular surgery are elderly and the procedures are complicated by atherosclerosis. The demands of operating on the circulatory system in these patients are great, and meticulous attention to detail is required so that potential complications, which are often difficult to treat and can have catastrophic consequences, can be avoided or prevented. Indeed, the old adage that it is easier to stay out of trouble than to get out of trouble is particularly relevant to vascular surgery and, in the authors’ opinion, describes the best strategy for decreasing the likelihood of operative complications. Preventing complications depends on the following basic principles: 1. 2.

3.

A careful and comprehensive preoperative evaluation, including assessment of the severity of associated diseases that may adversely affect outcome A thorough understanding of the natural history of the disease process so that the surgeon can properly assess the risk of treatment and opt for no treatment when such a decision is appropriate High-quality preoperative imaging studies that provide an anatomic picture of the disease process and form the basis of the ultimate surgical treatment plan

5.

6.

A surgical strategy that appropriately balances the following factors: achieving the treatment goal and providing the least risk to the patient, the best chance of success, and a level of durability that will, ideally, allow the results of the procedure to last for the rest of the patient’s life An operation performed with a flawless technique so that the need for early reoperation due to technical errors can be minimized because reoperation invariably increases the risk of systemic complications and may adversely affect survival or long-term results Excellent and attentive postoperative care under the direction of the vascular surgeon

When complications do occur, early recognition and prompt treatment are crucial for diminishing their impact and maintaining a good result. This chapter will discuss complications common to all vascular surgical procedures and those specific to the most commonly performed procedures. Emphasis will be given to early recognition and treatment but also to avoidance and prevention.

ASSOCIATED DISEASES AND THEIR IMPLICATIONS FOR COMPLICATIONS OF VASCULAR SURGERY Coronary Artery Disease Approximately 80% of patients undergoing vascular surgery have coronary artery disease (CAD) (1), which is the most common cause of both early and late death after such procedures (2–4). Most surgeons believe that there is a direct relationship between the extent of CAD and the likelihood of cardiac complications. When patients have overt symptoms of CAD or have recently suffered a myocardial infarction (MI), the risk of an event during vascular surgery is high enough to warrant a thorough preoperative evaluation, including coronary angiography and angioplasty, or even

306

Pomposelli et al.

coronary bypass grafting when indicated (4). Clearly, however, many patients have clinically significant coronary disease without obvious symptoms (5). Determining which of those patients will probably suffer an adverse event related to vascular surgery has become a controversial and widely studied topic that is beyond the scope of this text. Most vascular surgeons augment the medical history and the results of clinical examination and baseline electrocardiography (ECG) with the results of some other noninvasive tests as a means of assessing the degree or severity of coronary disease. Dipyridamole thallium imaging in particular has emerged as the most widely used screening test (6,7). On the basis of a clinical evaluation and the results of noninvasive testing, patients are classified as being at low, intermediate, or high risk of cardiac complications in association with vascular surgery (8,9). Studies have demonstrated an association between a positive risk assessment and the likelihood of a cardiac event (10). No study, however, has conclusively demonstrated that preoperative interventions for treating CAD improve the outcomes associated with peripheral vascular surgery, and this lack of findings raises questions about the costeffectiveness and the usefulness of such an approach. In spite of this controversy, it is important to recognize that perioperative cardiac complications are common among patients with CAD. All patients should be monitored closely, especially during the first 48 hours after surgery, when most events are likely to occur. Many cardiac complications can be avoided, in the authors’ opinion, by judicious fluid management, especially by avoiding hypervolemia and congestive heart failure (CHF). Pulmonary artery catheters may be helpful in fluid management, especially for elderly patients, those with diabetes, and those with left ventricular dysfunction, pulmonary disease, or renal insufficiency.

Pulmonary Disease Patients undergoing vascular surgery often have chronic obstructive pulmonary disease, and many are current or former smokers. Carefully eliciting a history of chronic cough, sputum production, and shortness of breath upon minimal exertion can usually determine which patients are at high risk of complications. Occasionally, pulmonary function studies and measurements of preoperative arterial blood gases may be needed. Patients with alveolar oxygen pressure below 70, carbon dioxide partial pressure above 45, or forced expiratory volume in one second below 1 L are considered to be at especially high risk (11,12). For such patients, the chance of pulmonary complications may be diminished by methods such as pretreatment with bronchodilators, smoking cessation, or pulmonary physiotherapy (13). The presumed risk of complications can suggest modifications in surgical decisions, including such measures as the preferential use of regional anesthesia, retroperitoneal as

opposed to transabdominal approaches to intraabdominal operations, the use of extra-anatomic (axillobifemoral) bypass, or angioplasty and stenting rather than open aortofemoral reconstructions. Laparotomy for aortic aneurysm surgery can also be avoided by the use of stent grafts inserted through the femoral artery, if the patient’s anatomical structure is suitable. If unrecognized, respiratory failure can rapidly lead to death (14), but it is usually relatively shortlived and reversible if properly treated. Although pulse oximetry has greatly simplified monitoring for hypoxemia (15), a high index of clinical suspicion is important if physicians are to recognize and treat respiratory failure before a catastrophic complication ensues. Unexplained agitation in the postoperative period should be considered to be caused by hypoxia until proven otherwise. Causes of postoperative pulmonary problems include ventilatory compromise due to abdominal incisions, atelectasis, poor cough and clearance of secretions, respiratory failure due to pneumonia, fluid overload, as well as adult respiratory distress syndrome and pulmonary embolism (PE).

Renal Disease Patients with preexisting renal insufficiency are most at risk of deterioration of renal function after vascular surgery. Elevated serum creatinine concentrations and hemodialysis have been markers of poorer outcomes and more complications after many vascular surgery procedures. When the patient’s renal function is compromised, proper hydration and maintenance of adequate intravascular volume and cardiac output are fundamental if further renal parenchymal injury is to be avoided. In addition, minimizing warm renal ischemic time during surgery and taking particular care with aortic cross-clamping are other important factors in avoiding atheroembolism. Chemical agents, such as aminoglycoside antibiotics and contrast agents used for angiography, can cause renal injury. The risk of contrast-induced renal injury is higher among patients who are dehydrated and those with diabetes mellitus and multiple myeloma. The risk of renal failure can be reduced by intravenous prehydration with 0.45% normal saline (16). Myoglobinuria can cause renal failure among patients undergoing revascularization after a prolonged period of limb ischemia. Inducing a brisk diuresis with vigorous hydration and intravenous administration of mannitol, administering diuretics, and alkalinizing the urine with intravenous sodium bicarbonate can reduce the risk of renal injury (17). In patients being treated with hemodialysis, expectations for vascular surgery must be tempered by the poorer outcomes often experienced by this group. Surgical decisions should be individualized and made by balancing a careful evaluation of the patient’s life expectancy, functional status, and likely benefit with the expected rate of complications.

Chapter 26: Complications of Vascular Surgery

Coagulopathy A history of blood-clotting abnormalities, such as inherited disorders (von Willebrand’s disease and hemophilia) or acquired problems (liver disease, vitamin K deficiency, use of anticoagulants or antiplatelet agents, and uremia), should be elicited as part of the initial evaluation of all patients who may require vascular surgery. Technical errors leading to excessive bleeding can result in consumptive coagulopathy and hypothermia, which further aggravate bloodclotting problems (see below). Underlying coagulation abnormalities should be corrected with appropriate clotting factors before surgery when possible. The administration of warfarin should be discontinued 48 to 72 hours before surgery, and the administration of antiplatelet agents should be discontinued one week before the procedure. Intravenous administration of heparin should be discontinued at least four hours before vascular surgery, if possible.

COMPLICATIONS COMMON TO ALL VASCULAR PROCEDURES Early Graft Thrombosis Early graft thromboses (those that occur less than 30 days postoperatively) are usually due to errors of commission on the part of the surgeon. Hemodynamic causes of graft failure include inadequate inflow or outflow, or hypoperfusion as a result of shock or cardiac failure. Most inflow or outflow problems are a result of choosing the wrong operation. A comprehensive arteriogram is the cornerstone of most treatment plans in vascular surgery and is essential for choosing the proper locations of inflow and outflow anastomoses. In general, all arteries proximal to the inflow anastomosis should be free of significant stenosis, and the outflow target artery should have unimpeded and continuous flow to the end organ. In some circumstances, inflow can be improved by adjunctive measures such as balloon angioplasty and stenting. If the outflow artery is diseased, collateral vessels must be adequate to support graft flow; if they are not, thrombosis will occur. These surgical decisions are not always straightforward and can require considerable judgment and experience. Mechanical causes of early graft failure are innumerable, and almost all are avoidable. Poor anastomotic technique can lead to a stenotic anastomosis or intimal flaps. Injudicious clamping can cause traumatic dissection, tears, or crush injuries, particularly in diseased or calcified arteries. Conduits can get twisted when tunneled or can kink if they are too long or improperly placed. Poor hemostasis can lead to graft compression by hematomas. Wound closure sutures can perforate or constrict underlying grafts. Vein grafts are prone to all of these complications and to specific problems related to harvesting and preparation (see below). The potential for mechanical failure is substantial enough that some objective

307

method of assessing the result at the end of the procedure should be used in most circumstances. These methods include hand-held continuous-wave Doppler ultrasonography, angiography (18), angioscopy (19), and intraoperative duplex scanning (20). When early postoperative thrombosis occurs, immediate reoperation and thrombectomy with some imaging modality should be carried out to identify the underlying problem. Generally, good results can be expected with this approach when a mechanical defect can be identified and corrected. Patients with no identifiable problem should be evaluated for a hypercoagulable state (antithrombin 3 deficiency, factor V Leiden mutation, protein C or S deficiency, anticardiolipin antibodies, etc.) (21). These patients should be given anticoagulant therapy, although the long-term prognosis for graft patency is poor.

Perioperative Hemorrhage Perioperative hemorrhage can occur as a result of an uncorrected, preexisting coagulopathy (see above), but it is most commonly due to technical error. Important basic measures for preventing this complication include the use of electrocautery for dissection, meticulous hemostasis, and the avoidance of excessive heparin doses (and the reversal of excessive doses with protamine sulfate when they do occur). Dissection of any arterial structure can result in injury and substantial bleeding from adjacent veins. Knowledge of the anatomy, dissection in the plane immediately adjacent to the artery, and avoidance of circumferential dissection when possible can avoid vein injury. Arterial anastomosis should be performed with proper suture bites and graft tension. Proximal graft anastomoses should be tested for leaks before flow is restored. Atraumatic clamps should be used and should be applied carefully so that the risk of injury related to their use can be reduced. When intraoperative hemorrhage does occur, a careful systematized approach must be undertaken. It is important to notify the anesthesiologist immediately about the problem at hand. Initial application of local pressure or packs will control most of the bleeding. Once the problem has been identified, suture repair, topical hemostatic agents (oxidized cellulose, microfibrillar collagen, and thrombin-soaked gelatin sponges), or both can be used for repair. Venous injuries require particular care. Hastily clamping bleeding veins is inadvisable because this procedure can worsen the injury and exacerbate the problem. Good results can usually be achieved by applying manual pressure with good exposure and suction and using a technique of partial and advancing exposure of the injury with continuous suture repair. If blood products must be transfused in large amounts, they should be warmed. One or two units of fresh frozen plasma should be given for approximately every four to six units of blood transfused,

308

Pomposelli et al.

and a platelet pack should be given after eight units of packed cells have been transfused. Particular care must be taken when lost blood is replaced with blood retrieved from the cell saver because cell-saver blood is completely devoid of clotting factors and its administration can result in severe coagulopathy. Massive hemorrhage may also lead to disseminated intravascular coagulation (22). Good communication and cooperation between the surgeon and the anesthesiologist are crucial to avoiding exsanguination and shock during repair of a vascular injury. As a general rule, patients who experience postoperative bleeding in the recovery room are best treated by an exploratory procedure. Intra-abdominal bleeding after intracavitary procedures is particularly worrisome because there are no external signs of bleeding. Often, the first indication of postoperative hemorrhage after an abdominal or a thoracic procedure is not hypotension and shock but rather low urine output that is unresponsive to volume replacement. Wound hematomas can compromise the airway and the function of the graft, can cause compartment syndromes, and can lead to weeks of wound morbidity or infection. If the surgeon is undecided about whether to drain a hematoma, it is generally a good policy to err on the side of a return trip to the operating room.

Myocardial Infarction and Respiratory Failure Acute respiratory decompensation has already been addressed. As previously outlined, prompt intubation and institution of mechanical ventilatory support is the first treatment. Correction of the underlying problem can then be addressed. It is important to remember that agitation may be the first sign of hypoxia among elderly demented patients who are about to suffer respiratory collapse. It is inappropriate to sedate such patients until their respiratory status has been fully evaluated. Chest pain, shortness of breath, or both are often the first indications of coronary ischemia and myocardial compromise. The only symptom among patients with diabetes may be nausea or diaphoresis in the absence of chest pain. When coronary ischemia is suspected, ECG should be performed and then followed by electrocardiographic monitoring. Treatment with oxygen, intravenous morphine, and nitrates to control chest pain should be initiated. Tachyarrhythmias should be aggressively treated with beta-blockers, calcium channel blockers, and diuretics as needed. Creatine phosphokinase concentrations, isoenzyme activity, and troponin levels should be determined every eight hours for 24 hours. Patients whose hemodynamic condition is unstable or confusing should be monitored with a pulmonary artery catheter, and those whose condition does not rapidly respond to therapy may require urgent coronary arteriography and surgical intervention.

Acute Renal Failure The most common cause of oliguria in the early postoperative period is hypovolemia due to third spacing. Low urine output that does not respond as expected to volume replacement can be due to several factors, including continued bleeding, deteriorating myocardial function, mechanical problems with the Foley catheter or urinary system (ureteral injury), or acute renal failure (Table 1). When acute renal failure is suspected, a pulmonary artery catheter should be placed to help optimize fluid status and eliminate potential prerenal causes. Urinalysis should be performed to detect characteristic tubular casts. The administration of all potentially nephrotoxic drugs should be discontinued. In some cases, perfusion scans or even arteriography may be indicated if compromised arterial flow to the kidney is suspected. An increase in the serum creatinine concentration is usually seen within 24 hours after the vascular procedure. Patients with nonoliguric renal failure should be treated with fluid replacement adequate to maintain urine flow. Anuric patients may benefit from one large dose of a loop diuretic in an attempt to convert the renal failure to a nonoliguric state, for which the prognosis is better (23). Electrolyte levels, especially serum potassium concentration, acid–base balance, and volume status, must be carefully monitored. When hemodialysis is required, a temporary indwelling double-lumen venous access catheter should be placed. The cause of acute renal failure should be investigated and, when possible, treated, although most causes are diagnosed after the fact and are not correctable. Treatment is therefore directed at support and maintenance of metabolic homeostasis until renal function returns. Conditions such as contrast-induced nephropathy and ischemic injury are reversible, although recovery may not occur for days or weeks. Renal failure caused by renal infarction or atheroembolism often results in an irreversible injury to the renal parenchyma. Protection of renal function is especially crucial when aortic or renal artery reconstruction is performed (24). The mortality rate associated with anuric renal failure remains high, a fact that underscores the importance of avoiding its occurrence in the first place. Table 1 Classification of Events Causing Renal Dysfunction in Vascular Surgery Prerenal

Parenchymal

Postrenal

Low cardiac output Shock due to any cause Hypovolemia

Acute tubular necrosis Renal ischemia

Foley catheter obstruction Ureteral obstruction

Nephrotoxic agents

Clot in pelvis, ureter, and bladder

Third-spacing losses Bleeding

Aminoglycosides

Dehydration

Others

Contrast agents

Abdominal compartment syndrome

Chapter 26: Complications of Vascular Surgery

Wound Infection Wound infection after vascular surgery is particularly worrisome when it occurs in wounds containing vascular grafts. The spectrum of wound infection ranges from mild cellulitis to extensive necrosis, with abscess formation and involvement of deep structures, including grafts. Contiguous involvement of a vascular graft by an infection starting in the adjacent soft tissues can result in graft exposure, disruption, and limb loss, even death. Although prosthetic grafts are most at risk, autogenous grafts can also become infected under some circumstances. The management of graft infections is discussed in more detail in the sections on aortic and lowerextremity bypass. The management of surgical wound infections depends on their severity. Appropriate antibiotics, drainage, and dressing care are the cornerstones of therapy. Complex wounds may require additional measures that may challenge the skills of even the most experienced vascular surgeon, including graft excision, repeated bypass through uninfected tissues, extensive debridement, and closure with myocutaneous flaps or free tissue transfers. Many wound infections result from poor surgical techniques. Exposure of vascular structures should be accomplished without the creation of large flaps, which may become necrotic and infected. Care should be taken to minimize the disruption of the lymph vessels during vascular exposure and to carefully ligate or cauterize those vessels that are disrupted so that the formation of lymph leaks or lymphocele can be avoided. Meticulous hemostasis is important in avoiding wound hematomas, which are perhaps the most common cause of vascular wound infections. When large hematomas do occur, they should be drained promptly. If a patient has both an infection and ischemia, as is commonly the case with lowerextremity ulcers and gangrene, the infection should be aggressively treated before a vascular graft is implanted.

309

substantially since the abandonment of silk sutures for vascular anastomosis (27). Avoiding endarterectomy at the site of anastomoses, taking proper suture bites in grafts and in the arterial wall, avoiding excessive tension of grafts, and using polypropylene or polyester sutures are important principles in preventing false aneurysms. Failure of reconstructions more than 18 months after their placement is usually due to the progression of atherosclerosis. The fact that all vascular surgery for atherosclerosis is palliative for an incurable disease mandates that patients be followed up for life so that disease progression can be detected and treated when necessary and so that the function of previously placed grafts can be preserved. Risk factor reduction, especially smoking cessation, should also be part of the overall treatment strategy recommended by the vascular surgeon to slow the progression of disease.

COMPLICATIONS OF LOWER-EXTREMITY BYPASS FOR CHRONIC ISCHEMIA During the last two decades, the technical aspects of arterial reconstruction in the lower extremity have changed tremendously. Improvements in suture materials and vein preparation techniques and the routine use of magnification, fine cardiovascular sutures, and specialized instruments have made it possible for vascular surgeons to bypass essentially any artery in the lower extremity, from the common femoral artery to the pedal and plantar arteries in the foot. Arteries as small as 1 mm in diameter can be successfully bypassed with results comparable to those achieved by more proximal arterial reconstructions. The technical challenges of performing distal arterial bypass of small vessels are substantial, and avoiding errors that might lead to graft thrombosis is especially important during these procedures. An incomplete list of the potential pitfalls of arterial reconstructive surgery in the lower extremity is presented below.

Late Complications Arterial constructions may fail as a result of neointimal hyperplasia, an incompletely understood mechanism related to the response of blood vessels to injury. This complication consists of a hypertrophic, subintimal smooth muscle proliferation leading to stenosis at the location of anastomoses, within vein grafts, on arterial surfaces after endarterectomy or angioplasty, and within previously placed stents (25). Neointimal hyperplasia most commonly occurs 6 to 18 months after arterial interventions. Anastomotic false aneurysms can occur as a result of the failure of sutures or graft materials or as a result of the loss of the structural integrity of the arterial wall at the anastomosis because of degeneration or infection (26). Many false aneurysms occur several years after implantation of a graft. The incidence of anastomotic aneurysms has decreased

Early Graft Thrombosis Technical errors of vein graft preparation are usually related to improper handling of the greater saphenous vein. When this vein is harvested, the surgeon must be careful not to avulse side branches. Branch avulsion usually requires suture repair, which can be difficult and can lead to areas of stenosis or stricture in the vein graft. During the last 20 years, the in situ saphenous vein bypass technique (28) has become quite popular. Cutting the valves and leaving the vein in its bed creates a tapered venous conduit, which more closely matches the size and configuration of the inflow and outflow artery. The favorable size match makes the use of this vein very appealing, especially for tibial artery reconstructions, in which veins as small as 2.5 mm in diameter work extremely well. Preparing a saphenous vein for use as an in situ

310

Pomposelli et al.

conduit is fraught with many technical errors. Most problems occur when the valves are cut. Numerous valvulotomes are now commercially available and can simplify the process. Each of these instruments has advantages and disadvantages, and the use of each is associated with a steep learning curve. As a general principle, any valve-cutting technique should involve careful cutting of the leaflets of the valve without injuring the adjacent wall. Intimal injuries of the adjacent vein wall, transmural perforations, and tears or cuts in the vein that are caused by valvulotomes can lead to areas of thrombosis and stricture formation, which can cause both early and late graft failures. The authors have found that using angioscopy (29) reduces the number of valvulotome injuries related to in situ vein preparation (Fig. 1). The proximal and distal ends of in situ saphenous vein grafts should be harvested adequately so that undue tension can be avoided and so that no kinking or compression occurs when the graft is connected to the artery. An especially acute angle can occur when an in situ saphenous vein graft is tunneled from the subcutaneous bed to the popliteal space. This acute angle can lead to early graft thrombosis or late stenosis at the angulation point. For this reason, when performing femoral-popliteal reconstructions, the authors prefer to remove the vein, cut the valves, and tunnel the graft in an anatomic plane as is traditionally done with reversed vein grafts. The ‘‘nonreversed’’ vein graft lies straight and avoids acute angles but maintains the size-match advantage. Before using saphenous vein grafts, vascular surgeons commonly distend them gently, usually using a syringe distension technique with a balanced salt solution containing heparin and papaverine. Studies have shown that using excessive pressure to distend saphenous veins can cause extremely high hydrostatic pressures within the vein, and these pressures may damage endothelial cells. Many authors recommend

Figure 1 The use of angioscopy reduces the number of valvulotome injuries related to in situ vein preparation.

using chilled heparinized blood as a preparation solution and avoiding syringe distension (30). Anastomoses should be performed with great care. It is not uncommon for either inflow or outflow target arteries in lower-extremity arterial reconstructions to have some degree of atherosclerosis. Arteries may be heavily calcified, especially among patients with diabetes mellitus. Gentle occlusion clamps should be used to control flow before arteriotomies are performed. The authors prefer to use soft-jawed hydrogrip-type clamps for proximal control of larger arteries and silicone plastic vessel loop slings for control of smaller arteries. When performing arteriotomies, surgeons must avoid injuring the posterior wall of the artery when opening it with a knife or Potts scissors. When anastomoses are created, attention must be paid to avoiding intimal flaps and strictures. It is important to evert the ends of the graft and the artery when sutures are placed so that a smooth flow surface can be promoted. Hemostatic control of distal calcified arteries can be quite challenging. As a general rule, heavily calcified arteries should not be clamped because clamping can crush or crack the vessel. Proximal control with tourniquets (31) or intraluminal vascular occluders is effective in these circumstances. Tunneling of grafts requires meticulous attention. It is important to be sure that grafts passing through the popliteal space rest in the space between the medial and lateral heads of the gastrocnemius muscle. Inadvertently tunneling a graft through the muscle belly of the medial head of the gastrocnemius muscle can lead to graft compression and thrombosis. Similar concerns apply when grafts pass beneath the sartorius muscle in the thigh. In aortofemoral reconstructions, grafts should pass posterior to the ureters to avoid entrapping the ureter between the iliac artery and the graft. Creating aortofemoral graft tunnels can also damage the sigmoid colon, the cecum, and the bladder. When tunnels are created in a subcutaneous plane, they must be of an adequate caliber to allow grafts to pass through them without compression or kinking. Excessive hemorrhage in the tunnel can be avoided by creating the tunnel before heparin is administered. The authors find it useful to perform the proximal anastomosis first and then to pass the fully distended graft through the tunnels under arterial pressure so that potential kinking and twisting can be avoided. This procedure also allows the surgeon to observe blood flow from the distal end of the graft before performing the distal anastomosis. Tunneling devices are commercially available and are often quite helpful, but they are not absolutely necessary if the previously described principles are adhered to before grafts are tunneled. Judgment errors have already been discussed. The proper selection of inflow and outflow arteries should be made only on the basis of a high-quality, comprehensive arteriogram. The ability to image the

Chapter 26: Complications of Vascular Surgery

entire arterial circulation of the lower extremity, from the infrarenal aorta to the toes, has been greatly enhanced in the last 15 years by the routine use of intraarterial digital subtraction angiography (32). Currently available equipment provides images with resolution quality as good as or superior to that obtained by conventional cut-film techniques. Recently, it has been suggested that magnetic resonance arteriography (MRA) provides images of the distal infrageniculate arterial circulation that are of superior quality to those obtained by other imaging methods (33). Our experience has not supported this concept. We continue to rely mostly on intra-arterial digital subtraction contrast angiography, although we have used MRA imaging selectively in our practice when the use of contrast agents was strongly contraindicated. When early graft failure does occur, it usually does so within the first 12 to 24 hours after the surgical procedure has been completed. Managing an acutely thrombosed graft requires a systematized approach. As a general rule, all patients should be treated with heparin and immediate reexploration. The authors usually expose the distal anastomosis first. The pattern of thrombosis will often provide some clues about the cause of the problem. If thrombus is present at the terminal end of the vein graft but most of the graft is still pulsating, a technical problem will usually be found at the distal anastomosis. The hood should be opened and the thrombus carefully removed. Any intimal flaps, strictures, or stenoses can then be identified and treated appropriately. If the distal portion of the vein graft is not thrombosed, the wound must be opened completely and the vein graft examined in its entirety. Thrombosis generally starts at the point of some abnormality in the vein graft, such as a kink, twist, stricture, extrinsic compression point, uncut valve, injury from the valvulotome, etc. If the entire vein graft is clotted, the problem may be either at the anastomosis or in the vein itself. A thrombectomy should be performed, flow reestablished, and an angiogram obtained to disclose the problem. Vein graft thrombectomy should be performed with great care. A thrombectomy catheter can be used, although often the fresh nonadherent thrombus can be milked out from either the proximal or the distal end of the vein graft. This procedure avoids the passage of balloon catheters that will severely damage the endothelium and compromise long-term graft patency. When a mechanical defect in a graft has been discovered and corrected, the long-term prognosis for graft survival is reasonably good. When graft thrombosis occurs with no identifiable cause but outflow is considered reasonable, a hypercoagulable state must be suspected. Patients with this complication should be treated with anticoagulants and evaluated for thrombophilia (21). In general, even when anticoagulation is administered, the long-term prognosis for graft patency, whether or not mechanical or technical defects have been discovered, is poor (34).

311

Hemorrhage Hemorrhage in lower-extremity arterial reconstructions is usually caused by avoidable technical errors. Poor hemostasis or excessive anticoagulation can lead to the formation of wound hematomas, which may not be apparent until several hours after surgery. As previously stated, all large wound hematomas in the lower extremity should be drained in the operating room. Leaving large hematomas in the lower extremity will result in a draining wound, which is susceptible to secondary infection. Wound morbidity related to hematomas can plague the patient and the surgeon for many months and can ruin an otherwise good result. Improper ligation of vein branches on saphenous vein grafts can lead to loss of the ligature and bleeding through a side branch. Side branch bleeding is usually profuse and dramatic and requires immediate surgery. Poorly performed anastomoses can also cause bleeding between suture bites because of tears in the arterial wall or breakage of sutures that have been improperly tied. Polypropylene suture material is particularly susceptible to fracture and breakage when insufficient care is taken in tying sutures or when the suture material is crushed or clamped by metal clamps or forceps.

Cardiac Complications More than 80% of patients who undergo lowerextremity arterial reconstructive surgery will have CAD, and 20% to 30% of these patients will have severe disease (1). Many patients may have few or no symptoms because of poor physical condition and inactivity or because of diabetes. It is often impractical or impossible to completely evaluate and correct the underlying coronary artery occlusive disease among patients with severe limb-threatening arterial ischemia of the lower extremity. Many patients will therefore undergo lower-extremity surgery with uncorrected coronary occlusive disease and must be monitored closely for signs of coronary ischemia. Continuous ECG monitoring and close nursing supervision are necessary in the early postoperative period, when most cardiac complications occur (24–48 hours after surgery). A common cause of cardiac decompensation in a patient undergoing lower-extremity reconstruction is hypervolemia due to excessive replacement of intravenous fluid. Patients undergoing lower-extremity bypass do not have large third-spacing requirements. On average, approximately 1.5 to 2 L of fluid will be necessary to adequately replace insensible losses during a 3.5- to 4-hour surgical procedure. After surgery, fluid requirements rarely exceed 1 to 2 L/day. Most patients will require monitoring of central venous pressure, and those patients who are known to have compromised myocardial function will require measurement of pulmonary artery pressures. Patients with diabetes may exhibit atypical signs of coronary ischemia. Diaphoresis, nausea, and vomiting in the absence of chest pain may be the first signs of an impending MI.

312

Pomposelli et al.

Surgeons must maintain a high index of suspicion and order ECG promptly when patients exhibit these symptoms. Patients who show signs of fluid overload should undergo aggressive diuresis with a loop diuretic such as furosemide. As a general rule, patients should gain no more than 1 to 2 kg of body weight after lower-extremity arterial reconstruction. After the first 48 hours, it is prudent to use mild diuresis to return patients to their preoperative body weight within four to five days after surgery. When blood products are needed, they should be administered slowly so that sudden fluid overload can be avoided. Fluid overload can also be avoided by administering a dose of a loop diuretic after one or two units of packed red blood cells have been given to patients with compromised ventricular function.

Limb Swelling

Wound Complications

Graft infection after lower-extremity arterial reconstruction can cause graft thrombosis or catastrophic hemorrhage. Proper sterile technique, adequate treatment of preexisting infection, and avoidance of wound complications are the most important principles of prevention. Wound infections involving underlying saphenous vein grafts can be treated successfully without removing the graft, even when it is exposed. Adequate debridement and drainage of infected tissues, as well as dressing care designed to maintain a moist environment over the graft, will occasionally allow healing by formation of granulation tissue and by secondary intention. Many surgeons will use a muscle flap to cover the exposed portion of the graft (36,37). Infections caused by gram-negative organisms, particularly Pseudomonas species (38), can result in necrosis of the vein graft wall, with disruption and hemorrhage. Such grafts must be ligated. It is occasionally possible to save an exposed prosthetic graft, provided that there is minimal surrounding infection, that most of the graft is incorporated into surrounding tissues, and that the anastomosis is not involved. Most of these grafts, however, will need to be removed (39). All patients with exposed grafts of any type are at risk of bleeding and must be followed up very closely.

The skin in areas of proposed surgical incisions should be free of superficial infection. Spreading cellulitis and infection should be adequately controlled before arterial reconstruction is performed. Prophylactic antibiotics with activity against staphylococcal and streptococcal organisms should be administered before surgery. Diabetic patients with multimicrobial foot infections should be treated with broad-spectrum antibiotics. Surgical incisions should be made directly over the femoral artery, the saphenous vein, and other structures; large flaps, which predispose patients to seroma, hematoma, and wound edge necrosis, should be avoided. Lymph nodes and lymph vessels that are encountered in the groin should be reflected medially so that disruption can be avoided; if these structures are disrupted, they should be carefully ligated or electrocauterized so that spillage of lymph, which may be contaminated with bacteria in patients with distal foot wounds, can be avoided. It is not unusual for patients to have some small amount of lymphatic fluid leakage from wounds caused by lower-extremity surgery, particularly groin wounds, for a few days after surgery. Likewise, small contained lymphoceles may be evident, particularly during the first postoperative visit; most will resolve spontaneously. Small lymph leaks can be contained with appropriate dry sterile dressings until they close. If any signs of infection are present, the drainage should be cultured and appropriate antibiotics should be given. Persistent lymph leaks that do not close during the first few weeks after surgery and large lymphoceles are probably best treated by reexploration and ligation, although this procedure can sometimes prove difficult. Large lymphoceles can sometimes be managed with repeated percutaneous aspiration. If lymphoceles and seromas are aspirated, sterile technique should be used so that contamination can be avoided. Infected hematomas and fluid collections require prompt incision and appropriate packing, as well as dressing care so that open wounds can heal by secondary intention.

Limb edema of the extremity after arterial reconstruction is very common. Virtually all patients will experience some degree of leg swelling. Edema usually resolves within two to three months, but can persist for many months and may be permanent for some patients. Many factors can cause edema of the lower extremity after arterial reconstruction, including a combination of lymphatic insufficiency, endothelium dysfunction, and increased perfusion pressure (35). Most leg edemas can be treated by encouraging leg elevation and reassuring the patient that the condition is temporary. For persistent edema, a graduated compression stocking can be helpful.

Infection

Late Graft Failure The time course of graft failure will usually give some indication of the underlying cause of the failure. When grafts fail within 30 days after surgery, the cause is usually a technical error, inadequate inflow or outflow, or a hypercoagulable state. When grafts fail more than 30 days but less than 18 months after surgery, the cause is usually neointimal hyperplasia. Neointimal hyperplasia often occurs in saphenous vein grafts and in other autogenous vein grafts as a result of intimal injuries at the time of graft preparation, or in arm vein grafts in areas of previous venipuncture. In the authors’ experience, the results achieved with arm vein grafts have improved with the use of angioscopy (40) to identify and discard arm veins that are of poor

Chapter 26: Complications of Vascular Surgery

quality because of previous venipuncture. Such veins quickly form strictures because of intimal hyperplasia, and grafts using these veins usually fail. Intimal hyperplasia in vein grafts is often focal and can cause graft thrombosis before the onset of symptoms or before ankle pressures are reduced. Graft patency has improved with the use of duplex ultrasonography to detect areas of stenosis, which may cause graft failure (41). Findings of reduced blood flow velocity or focal areas of stenosis with blood flow velocities 2.5 to 3.5 times the baseline values should be investigated with arteriography so that the area of stenosis can be identified. Many such areas can be repaired with vein patch angioplasties or interposition grafting. Intimal hyperplasia can sometimes be diffuse in nature, involving long segments or even the entire vein graft. Such vein grafts are unsalvageable and should be replaced. When arterial reconstructions fail more than 18 months after surgery, the cause is usually progression of the disease in either the inflow tract or the outflow tract. When an arterial reconstruction fails after a period of normal function, the most important factor in determining a course of action is whether ischemic symptoms return. Approximately 10% to 20% of arterial reconstructions will fail late without the return of symptoms. In such cases, patients should be followed up expectantly without another arterial reconstruction. When severe symptoms do occur, a secondary bypass procedure will be necessary and an arteriogram should be obtained. Secondary arterial reconstructions can often be challenging because previously dissected arteries that are now in scar tissue will need to be exposed because usable saphenous vein is absent, and because outflow will have to be extended to more distal and smaller target arteries. The results of secondary arterial reconstructions are generally inferior to those obtained with primary procedures, and this fact underscores the need to do it right the first time.

COMPLICATIONS OF LIMB AMPUTATION Amputations are an inevitable and important part of the treatment of many patients with vascular diseases. Amputation should be performed with the same care and attention to detail that the surgeon devotes to arterial reconstructions. To do otherwise does the patient a great disservice. Amputations are usually performed when some other procedure, usually a lower-extremity bypass, has led to a complication or has failed, or as treatment for gangrene or infection. An amputation can relieve pain, improve the quality of life, and end a long course of recurrent problems. Amputations should not be assigned to junior staff. Surgeons should be sensitive to patients’ fears about all amputations. A recommendation for amputation should be given to the patient in an unhurried and compassionate manner. No amputation is complete until the patient has been rehabilitated to the fullest extent possible. The surgeon’s burden of responsibility does not end until this goal has been achieved.

313

Lack of Healing Amputations fail to heal when they are performed in areas of ischemia. Necrosis is usually not evident for several days. A variety of tests have been used to predict the chance of healing, including segmental pressure measurements, pulse volume recordings, xenon flow studies, laser Doppler flow studies, and transcutaneous oxygen measurements (42). No test is completely reliable, and may predict failure at a level where healing can still occur. The authors continue to rely on physical examination and clinical judgment to determine the level of amputation (43). This practice results in the need to revise some failed amputations but avoids amputating a limb too proximally on the basis of erroneous test results (44).

Infection Patients who undergo amputation often experience active tissue loss and infection. As many as 10% of stumps will become infected, but most infections will respond to antibiotics and will not require operative incision and drainage. When infections are severe, more proximal revisions may be required (45). When patients require below-knee amputation for severe foot sepsis, performing the procedure in two stages can reduce the risk of stump infection. In the first stage, the foot is removed at the ankle (guillotine amputation) so that the infection can be controlled. After the infection has resolved, a formal below-knee amputation is performed.

Venous Thromboembolism Patients are often bedridden or limited in their physical activity for a period of time before and after amputation. Deep venous thrombosis (DVT) occurs in as many as 12.5% of patients. A high index of suspicion is necessary for diagnosing this complication, and appropriate prophylaxis with subcutaneously administered heparin may prevent its occurrence (46).

Myocardial Infarction and Death The mortality rate associated with major amputations is 7% to 15%, 5 to 10 times higher than that associated with lower-extremity revascularization (47). Advanced CAD is the most common cause of this complication, although uncontrolled sepsis and PE also contribute to its occurrence.

COMPLICATIONS OF CAROTID ENDARTERECTOMY Carotid endarterectomy (CEA) is one of the most studied and scrutinized vascular surgery procedures. Although innumerable studies of this procedure have been performed, including several large, randomized prospective trials, controversy still exists about its proper role in the prevention of ischemic stroke, especially in patients without symptoms. All studies

314

Pomposelli et al.

emphasize the importance of a low complication rate, especially a low rate of perioperative stroke, if endarterectomy is to be of benefit in preventing ischemic stroke.

Stroke The intolerance of the brain to even brief periods of loss of blood flow is the fundamental issue that surgeons must keep in mind when performing CEA. A mishap can turn an otherwise enjoyable and usually straightforward technical exercise into a surgical nightmare, the end result of which is a devastating neurologic injury or death. More than any vascular surgery procedure, CEA requires vigilance and a clear understanding of how to stay out of trouble. The most feared complication is stroke, which can occur for a number of reasons.

Diagnostic Arteriography At one time, all patients scheduled for CEA routinely underwent diagnostic arteriography, and this test is still the gold standard for determining the degree of stenosis in the internal carotid artery. The risk of stroke associated with arteriography itself ranges from 0.5% to 1.5% (48). In recent years, arteriography has been used much more selectively because of the accuracy of duplex ultrasonography (49) and MRA in determining the severity of stenosis. Currently, fewer than 5% of patients at our institution undergo diagnostic arteriography before CEA.

Perioperative Stroke Intraoperative strokes (Table 2) are caused by technical errors, dislodgement of atheroma during dissection, or clamping ischemia (50). The most common error is poor management of the end point of the endarterectomy in the internal carotid artery. The end point should be visualized and care should be taken to ensure that all loose debris has been removed so that an intimal flap is not created upon restoration of blood flow (51). In the authors’ experience, tacking sutures should be rarely necessary if an effort has been made to completely remove the atheroma and to be sure that the end point is well adhered; however, such sutures may occasionally be needed. Arterial closures should be performed with care so that the lumen is not narrowed at the end point because such narrowing may cause thrombosis. Small arteries (less than 4 mm in diameter) are best closed with a patch. Many vascular surgeons, including the authors, now routinely use patch closure. Arterial dissection should be performed gently, especially around the atheroma-filled bulb. Table 2 Causes of Stroke After Carotid Endarterectomy, from Most Common to Least Common Technical errors Plaque embolism Clamping ischemia Cerebral hemorrhage

Adhering to the principle that the patient is dissected away from the artery and not the artery away from the patient will minimize the chance of plaque dislodgement. Clamping ischemia is uncommon and can be avoided by proper cerebral monitoring, including regional anesthesia with neurological assessment or general anesthesia with EEG monitoring, measurement of internal carotid backpressure, and other techniques. Patients with clamping ischemia should be treated with insertion of an indwelling shunt. An alternative approach favored by the authors is to use shunts for all patients and to insert the shunts while patients are under general anesthesia. Perioperative neurologic deficits usually occur within the first 12 to 24 hours after surgery, after an initial period of normalcy (52). A standardized, systematic approach must be carried out when such deficits occur. The patient should be placed under general endotracheal anesthesia, and the wound should be opened and the artery examined by palpation and hand-held Doppler ultrasonography. If there is no pulse, the artery should be opened and cleared of thrombus, and the cause of the problem should be identified and repaired. Patch closure should always be used, and arteriography is usually performed after the procedure. If the artery has a pulse, an angiogram is performed while the patient is on the operating table, and the artery is opened if a defect is identified. In more than 70% of cases, a technical error of some type will be identified. When such treatment is performed promptly and a correctable technical error is identified and corrected, the prognosis is favorable, and many patients recover fully. Stroke occurs in 1% to 5% of patients (according to single-institution studies), is mainly dependent on the severity of presenting symptoms (symptomatic or asymptomatic disease preoperatively), and may be influenced by the presence of contralateral occlusion of the internal carotid arteries. Nonetheless, most strokes are caused by surgical errors and are therefore preventable. The American Heart Association consensus statement suggests that the perioperative stroke rate associated with CEA should be less than 3% for patients without symptomatic disease and less than 6% for patients with symptomatic disease. Surgeons unable to achieve these outcomes should not perform CEA.

Cerebral Hemorrhage Cerebral hemorrhage, which is an uncommon cause of stroke and carries a mortality rate of 50%, can occur several days after CEA and is often heralded by a severe headache. The causes of cerebral hemorrhage are unclear, but relief of critical stenosis and postoperative hypertension are common features (53). When patients have suffered ischemic stroke, when computed tomography (CT) scans show large infarcts, and when severe neurologic deficits are present, delaying CEA for four to six weeks is advisable so that conversion of an ischemic infarct into a hemorrhagic infarct can be avoided.

Chapter 26: Complications of Vascular Surgery

315

Neck Hematoma Neck hematomas occur in as many as 5% of patients (54). Most result from technical errors associated with poor hemostasis or arterial closure. Airway compromise and death can occur. Tracheal deviation with large hematomas can make intubation difficult and tracheostomy impossible. Smaller hematomas are painful and unsightly, and they delay recovery. Avoiding severe postoperative hypertension is an important principle in preventing hematoma. As a general rule, the authors evacuate all but the smallest neck hematomas. Patients with rapidly expanding hematomas should be immediately intubated before they are transported to the operating room. Opening the wound and evacuating the clot at the bedside may be necessary to facilitate attempts at intubation.

Figure 2 Vein patch repair for carotid endarterectomy.

Nerve Injury The vagus, hypoglossal, glossopharyngeal, and mandibular branches of the facial nerves and the spinal accessory nerves can be injured during dissection. Injury rates are higher in association with operations for recurrent stenosis (55). Nerve injuries can be prevented by knowing the anatomy, avoiding electrocautery in areas adjacent to nerves, and exposing vein branches circumferentially before ligature. Most injuries resolve, assuming the nerve has not been transected, but they can cause many months of morbidity. Bilateral injury to the vagus nerve can lead to airway compromise.

Myocardial Infarction A high percentage of patients with carotid disease will also have underlying CAD, but the rate of MI is usually less than 2% (51).

Patch Disruption or Infection Patch disruption is rare but can occur with vein patch closure and may be more common with saphenous vein harvested from the ankle (as opposed to the groin). Most surgeons now use prosthetic patches, which rarely become infected. We have encountered only four patch infections in our last 1000 consecutive CEAs. Treatment involves removal of the infected synthetic patch and repair with a vein patch (Fig. 2). All patients should receive a dose of antibiotics before undergoing CEA. Meticulous sterile technique and avoidance of hematoma formation are preventative measures.

Restenosis Restenosis occurs in 10% to 45% of cases, especially among women, in patients with arteries less than 4 mm in diameter, and in smokers. Most cases of restenosis are due to neointimal hyperplasia and occur in the first 6 to 18 months after surgery. Symptomatic restenosis is uncommon and the risk of stroke is low. Most lesions occlude less than 70% of the artery

and some may occasionally regress. Patch closure may decrease the incidence of restenosis, especially in patients with small arteries and among women (56). Restenosis occurring more than two years after endarterectomy is usually due to progression of atherosclerosis. The precise indications for reoperation are unknown, and the complication rates associated with a second procedure are higher than those associated with the original CEA. Most surgeons will operate when patients exhibit symptoms or when critical asymptomatic lesions are at risk of thrombosis (57). Angioplasty with stenting is currently being investigated as a treatment for restenosis after CEA.

Hyperperfusion Syndrome Hyperperfusion syndrome, which occurs in fewer than 1% of patients, is manifested by cerebral edema, headache, and seizures. It usually occurs 3 to 11 days after CEA. Characteristics that predispose patients to this complication are critical contralateral disease and preoperative and perioperative hypertension. The cause of this syndrome is thought to be loss of autoregulation of cerebral blood flow; this loss leads to cerebral edema. Treatment consists of controlling seizures, which can be difficult, and correcting hypertension. If hyperperfusion syndrome is mistaken for thrombotic infarct and is treated with intravenous administration of heparin, fatal intracerebral hemorrhage can occur.

COMPLICATIONS OF SURGERY TO THE AORTA AND ITS BRANCHES Open Aortic Aneurysm Repair and Aortofemoral Bypass for Occlusive Disease Since Dubost performed the first successful abdominal aneurysm resection in 1951, substantial improvements have been made in anesthesia, critical care, and

316

Pomposelli et al.

surgical technology. These developments have led to a decline in operative mortality rates, from 14–20% in early reports to 2–5% today (58). Some large series have even found perioperative death rates below 1% (59). Because the natural history of abdominal aneurysms involves progressive dilation, surgical repair is now being advocated for patients with smaller aneurysms and for older, higher-risk patients than patients with those that could have been treated in the past. In this section, we will consider many of the commonly occurring complications associated with abdominal aneurysm surgery in conjunction with the complications associated with aortofemoral bypass because the clinical outcomes of these procedures are similar.

Hemorrhage The aorta is a large artery surrounded by veins. An old surgical pearl states that the safest place in aneurysm surgery is inside the aneurysm itself. Limiting dissection to the space immediately surrounding the aorta is an important principle in avoiding many problems with bleeding. Intraoperative bleeding usually results from injury to an adjacent venous structure, injury to the aorta itself, or inadequate hemostatic integrity of the anastomosis. Venous injury can be avoided by minimal dissection of the aneurysm and the iliac arteries. Encircling the aorta with tapes is both unnecessary and dangerous. Aortic tissue at the neck of the aneurysm is often thickened due to thrombus or atheroma, or it may be thin walled and friable. The aortas of patients with occlusive disease may be heavily calcified. Aortic clamping should be performed carefully, with appropriately sized, softjawed clamps. It is often prudent to lower the systolic blood pressure before the clamp is applied. If suprarenal clamping is necessary, the renal arteries should be clamped first so that emboli from the aorta do not reach the renal arteries. The left renal vein can be ligated to improve exposure as long as ligation is performed to the right of the left adrenal and gonadal veins, which are important collateral vessels. Transecting the aorta is rarely necessary in treating aneurysmal disease. The graft inclusion technique works well in most circumstances and avoids the hazard of injuring posterior structures when the aorta is divided. If anastomoses are erroneously placed in the proximal aneurysm rather than in its neck, bleeding may occur upon restoration of flow, or anastomotic or juxta-anastomotic aneurysms (60) may develop later. The aorta should be exposed to the level of the left renal vein, which crosses the anterior aorta at the approximate level of the renal arteries. In approximately 10% of cases, the left renal vein crosses the aorta posteriorly; in such cases, the vein may be injured during cross-clamping if its aberrant location is not recognized. The proximal anastomosis should be placed as close to the renal arteries as possible to ensure that all of the aneurysm has been excluded

from blood flow. Injury to adjacent structures and anastomotic bleeding may also be avoided by large suture bites, a running suture using the graft inclusion technique, and grafts of the proper size. Plates of calcium should not be incorporated into suture bites because they will prevent a snug approximation of the graft to the artery. These structures should be carefully removed. Occasionally, sutures may require reinforcement with felt pledgets, either in a running closure or as a series of interrupted sutures when the neck of the aneurysm is very thin and friable. Different principles apply when aortic occlusive disease is present. Because the aorta is not aneurysmal, the anastomosis can be placed at the most distal disease-free point in the aorta. Both end-to-end and end-to-side anastomoses work well. The authors generally prefer end-to-end grafts unless the patient is a sexually potent man with disease confined principally to the external iliac arteries. In such cases, an endto-side graft will preserve flow to the pelvis, and this blood flow, it is hoped, will maintain erectile function. Surgical procedures can be challenging when the arteries are heavily calcified. Clamping the aorta can cause cracking of the arterial wall and catastrophic hemorrhage. When aortic calcification is severe, the most prudent option may be to abort the planned aortic procedure and perform an axillary-bifemoral bypass instead. Renal and visceral artery bypass may be performed in conjunction with aortic surgery or as a separate procedure. Blood flow into renal reconstructions usually comes from the infrarenal aorta, but when the aorta is diseased, extra-anatomic bypass from the hepatic or splenic artery (61) may be used and is less hazardous. Retrograde mesenteric bypasses can be created from the infrarenal aorta or iliac arteries. If these grafts are improperly positioned, they can easily kink or twist when the bowel is returned to its normal position. Antegrade bypasses, created from the supraceliac aorta (62), are not as likely to kink as retrograde grafts but are technically more demanding to create, are prone to the hazards associated with clamping the aorta in the distal thorax (paraplegia and visceral emboli), and are tunneled posterior to the pancreas, a procedure that can cause troublesome bleeding. Proper patient selection and the experience of the surgeon are important factors in avoiding the complications associated with renal and visceral artery bypass. Hemodynamic instability accompanied by low hemoglobin levels and poor urine output during the postoperative period commonly indicates continued blood loss. Most postsurgical bleedings occur at the sites of anastomoses, through the graft interstices, or in the periaortic retroperitoneal tissues. Reversal of heparin at the conclusion of the procedure can generally limit such bleeding. Postsurgical bleeding may be self-limiting, but unresponsive hypotension combined with signs of hemorrhage mandates surgical intervention. Postoperative bleeding is more likely after

Chapter 26: Complications of Vascular Surgery

intervention for ruptured aneurysms because of the development of coagulopathy as a result of preoperative hypothermia and the transfusion of large volumes of blood. Using intraoperative autotransfusion to treat patients who have suffered an acute ruptured aneurysm reduces the requirement for homologous red blood cell transfusions and is associated with a lower mortality rate for hospitalized patients (63).

Limb Ischemia and Thrombosis of the Graft Limb Acute lower-limb ischemia immediately after aortic surgery, the so-called ‘‘trash foot’’ syndrome (Fig. 3), is usually due to emboli from either the graft or the native vessels. Distal small-vessel thrombosis is another possible culprit. This phenomenon is best prevented by early mobilization and clamping of the iliac arteries and copious irrigation of the graft and anastomoses with heparin saline solution (64). Thrombosis of one or both limbs of an aortic graft will usually occur during the first two years after the repair. Early thrombosis (less than 30 days after graft placement) is normally due to a technical error; the most common cause is an intimal flap. Treatment requires thrombectomy and repair of the technical problem. Late thrombosis at the distal anastomosis is probably the result of neointimal hyperplasia or recurrent atherosclerosis. Treatment options include femorofemoral bypass, reconstruction of the distal anastomosis with graft interposition, and patch angioplasty.

Myocardial Infarction Acute MI is the principal cause of hospital death after elective aortic repair. One large series demonstrated that cardiac problems accounted for 39.8% of all major postoperative complications (58). Risk factors for acute MI include a history of CHF, a prior Q-wave MI, rupture of an abdominal aortic aneurysm,

317

preoperative hypotension, excessive bleeding, and prolonged cross-clamping time. Many studies have attempted to determine the benefit of extensive preoperative cardiac screening (thallium scanning, echocardiography, and cardiac catheterization), but the most predictive findings are those obtained by a detailed history and a complete physical examination (65,66).

Ischemic Colitis and Gastrointestinal Complications Clinically evident ischemic colitis complicates 1% to 2% of cases of elective aortic repair and as many as 35% of cases with ruptured aneurysms (67). Patients usually experience bloody diarrhea one to seven days after the surgical procedure. Leukocytosis, abdominal distension, and abdominal pain should also alert the surgeon to the possibility of ischemic colitis. Sigmoidoscopy is the initial diagnostic test of choice, although it will not differentiate mucosal ischemia from full-thickness infarction. Approximately 50% of cases will be self-limiting, although some will progress to stricture formation. If transmural bowel infarction is present, surgical resection with colostomy and creation of a mucous fistula should be expediently performed. Creating two stomas will allow the surgeon to monitor further bowel necrosis, if it occurs. In an attempt to prevent ischemic colitis, many surgeons reimplant the inferior mesenteric artery (IMA) after aneurysm resection if that vessel does not demonstrate brisk back-bleeding. This technique, however, has not been shown to decrease the incidence of postoperative ischemic colitis (68). Other gastrointestinal complications of abdominal aortic surgery include paralytic ileus, Clostridium difficile enterocolitis, mechanical obstruction, and acute cholecystitis.

Acute Renal Failure Defined as a 20% increase in the blood urea nitrogen or serum creatinine concentration, acute renal failure complicates 3% to 7% of elective infrarenal aortic procedures and approximately 75% of aneurysmal ruptures. The mechanism of injury after rupture is decreased renal perfusion leading to acute tubular necrosis. In elective cases, atheroemboli are most commonly responsible for the renal failure. The hospital mortality rate is substantially increased for patients who require dialysis (69), but 75% of the survivors will regain renal function. Although in the past the preoperative administration of low-dose dopamine was advocated for preventing acute renal failure, this agent is no longer believed to be beneficial. The best preventative measure is careful monitoring of the fluid status in an attempt to maintain ideal renal perfusion.

Impotence Figure 3 Acute lower extremity ischemia immediately after aortic surgery – ‘‘trash foot syndrome’’.

Although commonly overlooked, impotence is one of the most common forms of morbidity associated with aortic surgery. Although nearly half of all men

318

Pomposelli et al.

undergoing aortic surgery are impotent to begin with, of the remainder, as many as 75% will experience loss of potency postoperatively (70). Preventing this complication requires avoiding the sympathetic plexus, which travels along the inferior abdominal aorta on the left, in close proximity to the IMA, and using an end-to-side graft configuration when appropriate, as previously described.

a hospital mortality rate of 11%, with 64% primary patency and 100% secondary patency 60 months after placement with an axillofemoral technique (73). When an extra-anatomic bypass is not feasible, reconstruction of the aorta with autogenous superficial femoropopliteal vein (74) or cryopreserved arterial homograft (75) may be possible.

Aortoduodenal Fistula Paraplegia Paraplegia, a rare but devastating complication, occurs in association with fewer than 1% of infrarenal aortic procedures and 5% to 25% of thoracoabdominal aortic aneurysm repairs. The clinical presentation ranges from anterior or posterior arterial ischemic syndromes to a complete transverse myelopathy. Once the symptoms appear, the possibility of improvement is remote. Paraplegia in association with infrarenal aortic procedures is due to pelvic devascularization and can usually be prevented by ensuring the perfusion of at least one hypogastric artery. Paraplegia in association with thoracoabdominal aortic surgery is usually due to spinal cord ischemia. Many factors, both anatomic and physiologic, may cause this complication. The precise method of prevention is unknown, but several methods have been suggested, including reimplantation of intercostal vessels, reduction of cerebrospinal fluid pressure, epidural cooling of the spinal cord, the use of various pharmacologic ‘‘neuroprotective’’ agents, and partial cardiopulmonary bypass (71).

Incisional Hernia The infrarenal aorta can be approached through either a retroperitoneal or a transperitoneal (midline) incision. The retroperitoneal approach is beneficial when patients are morbidly obese, when they have undergone multiple abdominal surgical procedures, and when exposure to the suprarenal aorta is necessary. The incision required for the retroperitoneal approach carries a 3% to 7% risk of hernia; however, a much more frequent problem is a persistent flank bulge. Occurring in 10% to 30% of patients, this bulge is related to intercostal nerve injury and can be prevented by avoiding extension into the 11th intercostal space (72). A midline incision is appropriate when access to the right renal artery or the intraperitoneal organs is necessary. This approach is associated with a 5% incidence of hernia, a rate similar to that associated with midline incisions for other surgical procedures.

Graft Infection The risk of graft infection is permanent and increases with time. The pathogenesis is usually bacterial seeding of the graft through a hematogenous route. Colonic bacterial translocation during surgery may contribute to the condition. The usual treatment involves removing the entire graft and performing an extra-anatomic bypass. One recent study reported

Often discussed but rarely seen, aortoduodenal fistula is a life-threatening complication of late graft infection (76). Months to years after surgery, patients may experience the so-called ‘‘herald’’ hemorrhage. Surgeons should maintain a high index of suspicion for this complication whenever a patient with a gastrointestinal hemorrhage has a history of aortic surgery. The first step in the diagnosis of this complication is an emergent esophagogastroduodenoscopy, which should demonstrate the fistula if one is present. Equally important, the procedure may reveal other more common causes of gastrointestinal hemorrhage, such as peptic ulcer disease or diffuse hemorrhagic gastritis. Aortoduodenal fistula usually occurs between the proximal suture line and the duodenum. Aortosigmoid fistulas have also been described.

Other Complications Additional types of morbidity that have been associated with aortic surgery include chyloperitoneum, ureteral obstruction, gluteal infarction, and obturator nerve injury.

Complications of Endovascular Aneurysm Repair The placement of aortic stent grafts to repair or exclude infrarenal aortic aneurysms was first attempted in 1991 but did not become widespread until the Food and Drug Administration approved two devices in October 1999. Stent grafts gave vascular surgeons an alternative to conventional open repair, particularly for treating high-risk or elderly patients who would probably be expected to poorly tolerate laparotomy and aortic clamping. In addition because endovascular aneurysm repair (EAR) is associated with less pain and shorter recovery times than other procedures, it is a beneficial alternative for all patients. Initial results have demonstrated the technical feasibility of EAR, as well as its advantages and shortcomings. As is true for all new surgical procedures, EAR is associated with a number of complications, some specific to the new procedure and some common to both the new and the old procedures.

Patient Selection, Improper Positioning, and Device Failure The most important determinant of the success of aortic stent grafts is the anatomy of the aneurysm and the iliac arteries. Proper proximal fixation

Chapter 26: Complications of Vascular Surgery

319

requires a suitable length of infrarenal aorta (at least 1 cm) that is free of thrombus and is not heavily calcified or too severely angulated (77). Aortic necks larger than 26 mm in diameter are too large for adequate sealing of the graft to the aortic wall. Similar criteria are important to assure proper distal fixation; in addition, the iliac arteries must be large enough to allow safe passage of the introducer (see below). During the procedure, proper proximal fixation based on the results of intraoperative arteriography, fluoroscopy, and road-mapping is crucial. If the stent graft is deployed too proximally, one or both renal arteries can be inadvertently covered. If placed too distally, it will not attach and could slip into the aneurysmal sac. Both occurrences require immediate conversion to open repair. Some devices are fixed to the aortic wall by hooks; others are fixed by the radial force of the stent. Device failure can result from hook fracture or continued expansion of the aortic neck, and such failure is perhaps the greatest concern associated with the concept of stent grafting. Because neither causative factor can be prevented, careful lifelong follow-up is mandatory for patients treated with stent grafts (see below).

Endoleak The stent graft is deployed from a remote site, usually the femoral artery, and is connected to the aortic wall by a metallic stent connected to a graft proximal to the aneurysm and distal to the iliac arteries. The aneurysm is thus excluded from blood flow; the blood moves through the device into the distal circulation. The blood in the aneurysmal sac clots and the aneurysm eventually shrinks. Persistence of blood flow in the space between the stent graft and the aortic wall is called an endoleak (78). There are four types of endoleak (79). Type I endoleaks occur when the attachment between the graft and the aorta or iliac arteries is faulty (Fig. 4). The aneurysm is not excluded from blood flow, remains pressurized, and can still rupture. Endoleaks of this type must be corrected when they are discovered. Type II leaks are caused by retrograde bleeding into the aneurysmal sac from branch vessels, usually the inferior mesenteric or lumbar arteries. Many of these endoleaks will clot within days or weeks. Type III leaks are specifically associated with modular stent grafts, which are composed of multiple pieces joined together at the time of implantation. Junction points can separate, thereby allowing blood flow into the aneurysmal sac. These endoleaks must be treated so that further expansion and possible rupture can be avoided. Type IV leaks result from defects or tears in the fabric of the graft, and most are self-limiting. Type II endoleaks are the most common; they occur in association with 20% to 30% of procedures. Their clinical significance is not fully understood; although they are generally considered benign, they have been associated with a continued increase in

Figure 4 Types of endoleak. Obtained with permission from Comprehensive Vascular and Endovascular Surgery, Elsevier 2004.

the size of the aneurysm and with rupture in some cases (80). These endoleaks should be treated when they are associated with aneurysmal expansion. The treatment of endoleaks is complex and beyond the purview of this chapter. Most can be corrected with additional endovascular techniques, but some may require open surgical repair or replacement of the stent with a conventional aortic graft (80). Lifelong follow-up is necessary for patients who have experienced endoleak complications; the function of the aortic stent graft must be evaluated with CT scans every six to twelve months.

Rupture of an Aneurysm or Vessel The stent graft is housed in an introducer sheath, which is inserted through a femoral artery cutdown and must traverse the iliac arteries. Such sheaths can be as large as 8 mm in diameter. Exerting undue force during insertion can cause arterial rupture or perforation, which results in substantial and even fatal hemorrhage. Patients with small or diseased iliac arteries are poor candidates for stent graft repair. Recent reports have described cases of aneurysmal rupture after seemingly successful placement of stent grafts (81). Most of these cases involved either short and angulated proximal aneurysm necks, which were poorly suited to stent graft placement, or a known

320

Pomposelli et al.

endoleak, with continued expansion of the aneurysm (82). Some ruptures have occurred in grafts thought to be in proper position and without obvious endoleaks or other defects. These problems underscore the fact that our understanding of the endovascular treatment of aneurysms is still incomplete and that proper patient selection and careful follow-up are crucial.

Graft Limb Thrombosis Graft limbs may occlude early, usually because of technical errors such as kinking or twisting the limb, or because of attachment site problems such as dissection or stenosis (82). When defects are seen on an angiogram obtained after the procedure has been completed, angioplasty, placement of an additional stent, or both can correct most problems. Late limb thrombosis is usually caused by changes in the shape of the aneurysm as it shrinks. Most aneurysms will shorten and will also decrease in diameter. When this happens, iliac limbs may kink, especially in the externally reinforced type of graft, which is relatively inflexible. The initial sign of this kinking may be severe limb ischemia; in such cases, an urgent femorofemoral bypass procedure is necessary. Patients without critical symptoms can often be treated with thrombolysis and an additional endovascular procedure, such as angioplasty and stenting.

Limb Ischemia Due to Emboli A thrombus or an atheroma may be dislodged by the wire sheath and by the passage of the device. Clamping the distal femoral arteries and initiating full anticoagulation before passing any devices can reduce the incidence of this problem.

Claudication and Paralysis Aneurysms of the common iliac artery are present in 50% to 60% of patients with aortic aneurysms. When the common iliac arteries are too large to allow the graft to be sealed properly, terminating the graft in the external iliac artery may be necessary; in such cases, one or two of the internal iliac arteries are covered by the stent graft. The internal iliac artery is often embolized with coils so as to prevent back-bleeding and the resultant large endoleak. It has become apparent that such embolization is not a benign maneuver: 20% to 40% of patients will experience substantial claudication in the buttocks, a condition that usually improves but does not resolve completely (83). Lowerextremity paralysis and ischemic colitis are recognized complications of bilateral internal iliac occlusion.

VENOUS DISEASES Complications of Varicose Vein Surgery For decades, the stripping and ligation method has been touted as a safe and effective means of dealing

with venous incompetence in the lower extremity. Postoperative hospital stays are usually brief, and many procedures are even performed on an outpatient basis. Although functional improvement occurs in most cases, reports of patient satisfaction are not consistently positive. Most dissatisfaction is related to the procedure’s failure to achieve the desired cosmetic result. Minor complications are fairly common and include bruising, pain, and slight numbness. Major complications are rare but include injury to the femoral vein or artery, PE, DVT, compartment syndrome, and lymphatic injury. Lymphedema and groin lymphatic fistula seem to occur most frequently after reexploration of the groin for recurrent varicosities. The incidence of major complications is less than 1% and that of minor complications is 17% (84).

Complications of Sclerotherapy The use of injectable irritating agents is an increasingly popular method of treating varicose veins and telangiectasia. Some of the more severe complications of this therapy are due to erroneous extravenous infusion of the sclerosing agent. Intradermal injection can result in ulcers or tissue necrosis, which may require wide debridement. Inadvertent intra-arterial injection is often associated with attempted injection of the agent into the lesser saphenous vein. The consequences of erroneous infusion can be severe enough to require limb amputation. Other untoward outcomes, such as allergic reaction, DVT, and PE, are rare (85).

Complications of Vena Cava Filter Placement Traditionally, vena cava filters were used to treat patients at high risk of PE for whom anticoagulation was either contraindicated or had failed to prevent thrombus formation. Many authors advocate using these filters to treat all patients at high risk of PE (86). In general, filters can be placed expediently in the operating room or even in the intensive care unit (87). Complications that may occur during filter placement include pneumothorax, PE, bleeding at the insertion site, and guidewire mishaps such as entrapment of the wire in the filter device or migration of the filter device (88). Surgeons should also recognize that the incidence of intra-atrial shunts is high among patients with chronic thromboembolic disease and that such shunts can lead to paradoxical cerebral embolism (89). Late complications include filter migration, filter fracture, and vena cava occlusion, which occurs in 6% to 7% of cases reported in most long-term series (88).

COMPLICATIONS OF SURGERY FOR ACUTE ARTERIAL INSUFFICIENCY Complications of Balloon Embolectomy For patients with acute lower-extremity ischemia due to embolism, balloon catheter embolectomy is still the treatment of choice. Limb salvage procedures are

Chapter 26: Complications of Vascular Surgery

successful in 75% to 90% of cases. Still, complications are common and sometimes severe. Mortality rates range from 10% to 20%. Local mechanical effects of the balloon on the vessel wall can include injury, perforation, or rupture. Wall injury leading to myointimal hyperplasia may result in stricture or, less commonly, in diffuse arterial narrowing. Preventing arterial injury requires using meticulous surgical technique, checking the balloon for leaks or eccentricity, and using the smallest effective catheter. Other risks associated with balloon embolectomy are groin hematoma at the insertion site and ischemia–reperfusion injury.

Complications of Ischemia and Reperfusion Haimovici first described the ischemia–reperfusion syndrome in 1960 (90). The pathological insult occurs when an acutely ischemic limb is suddenly reperfused with warm blood. Acute muscle edema leads to elevated compartment pressures, which in turn lead to neuromuscular dysfunction. The syndrome is mediated by the release of potassium, myoglobin, free radicals, cytokines, and tumor necrosis factor-a from the injured muscle. Acute myoglobinuria can result in renal failure. Tumor necrosis factor-a is believed to lead to pulmonary injury and even to death in severe cases. The extent of the initial muscle injury is determined by ischemia time, limb temperature, and muscle location and type. Many treatment options exist for ischemia– reperfusion syndrome. Fluid support and aggressive monitoring of the patient’s electrolyte levels, especially the potassium concentration, are essential. Perioperative administration of hypertonic mannitol may be protective. Controlled limb reperfusion with a mixture of blood and a crystalloid solution and with limb hemodialysis has been advocated to control the metabolic derangements that lead to ischemia–reperfusion syndrome. Prophylactic fasciotomy is indicated when ischemia is prolonged (for more than six hours) (91).

Complications of Fasciotomy When performed in a timely and expedient manner, a fasciotomy can literally save both life and limb. In light of this fact, complications arising from fasciotomy, although frequent, are relatively minor. Wound pain, altered skin sensation, pruritus, skin discoloration, and substantial scarring are the most commonly reported complaints. Infections are rare but can cause tibial osteomyelitis; excellent wound care is required to avoid this complication. The authors prefer to close all fasciotomies, usually with a split-thickness skin graft, as soon as possible. Patient education and reassurance provide adequate therapy in most cases. Bleeding and nerve injury are less common than infection and should be recognized and treated early.

Complications of Surgery for Hemodialysis Surgical creation of hemodialysis access is a leading cause of morbidity among patients with end-stage

321

renal disease. Complications associated with these procedures lead to frequent hospital admissions for the patient, large costs to the community, and a great deal of frustration for both the doctor and patient. Therefore, every effort should be made to minimize these untoward outcomes.

Temporary Devices Providing emergent access for dialysis when patients have not yet undergone placement of permanent access or when the access has not yet matured often requires the insertion of a venous catheter. These catheters must be placed in a large vein, such as the subclavian vein or the femoral vein, so that the high flow rates necessary for hemodialysis can be accommodated. Temporary dialysis catheters are subject to the same complications that plague all central lines, including pneumothorax, hemothorax, hematoma, infection, and thrombosis.

Arteriovenous Fistulas and Grafts Because it is associated with higher rates of patency and a lower incidence of complications, creating an arteriovenous fistula (AVF) is the procedure of choice for obtaining permanent access for hemodialysis. When creating an AVF is not possible because of anatomic constraints or poor vein quality, an arteriovenous graft (AVG) of polytetrafluoroethylene is the primary alternative. The problems most frequently encountered with each procedure will be discussed below. Lack of Maturation of Arteriovenous Fistula

Although the long-term patency rate of AVFs is higher than that of AVGs, their initial failure rate is also higher. This lack of development is usually due to small-diameter veins (less than 2 mm). In some series, this complication has been reported to occur in as many as 15% to 25% of cases. Once technical problems have been ruled out as a cause of the fistula’s failure, little can be done to salvage the site. The surgeon may then elect to place a fistula at an alternative site, such as the antecubital fossa if the wrist was the primary site, or to construct an AVG. If risk factors for lack of maturation are present (patients with diabetes, women, and older patients), using the upper arm as the initial site may yield improved results. Ischemic Steal

A complication associated with both AVGs and AVFs is the vascular steal phenomenon, which can lead to hand ischemia (Fig. 5). This complication occurs most commonly when the brachial artery is used. The size of the arterial anastomosis is crucial; the diagnosis is correlated with a wrist–brachial index of less than 0.75, which improves with digital occlusion of the fistula. Treatment options include fistula ligation, arterial banding, and graft lengthening (92).

322

Pomposelli et al.

Venous Hypertension

Figure 5 Vascular steal phenomenon leading to hand ischemia.

Early and Late Thrombosis

The leading cause of fistula loss is thrombosis, which can be classified as either early (less than six weeks after placement of the fistula) or late. Early thrombosis is often due to a technical problem, but it can also occur when the diameter of the native vessels is small—especially the vessels on the venous end. Late thrombosis can be caused by multiple factors, including stenosis (which reduces the flow of blood through the graft), coagulation abnormalities, and cardiac dysfunction. Factors that confer a negative prognosis for graft survival are diabetes, advanced age, female sex, hypoalbuminemia, and hyperfibrinogenemia. The stenosis forms at the venous anastomosis, in the graft itself, or in the central venous system. Approximately 90% of stenoses occur at the venous anastomosis. Thrombectomy, graft revision, or both can often restore function. Thrombolysis with endovascular therapy has been reported to achieve patency rates equivalent to those achieved by thrombectomy or graft revision but at a substantially greater cost and with a high rate of technical failure, which requires surgical intervention (93). Infection and False Aneurysm

Infection, although rare in association with autogenous fistulas, complicates 3% to 8% of all AVGs. The causative organism is predominantly Staphylococcus aureus (94). The ideal treatment is complete removal of the graft; however, when access is limited, some success has been achieved by partial removal of the graft or by incision and drainage with administration of intravenous antibiotics. The formation of a pseudoaneurysm, which is fairly common in association with AVGs, results from continued puncture of the graft with large-bore needles. When puncture wounds become excessively large or numerous, surgical revision is indicated.

Venous hypertension is a rare complication most often seen when side-to-side Brescia–Cimino fistulas are created. The hypertension results from arterialization of the venous system proximal to the fistula. If the venous valves are incompetent, retrograde flow can develop. Physical findings include distal edema, venous congestion, and discoloration that can progress to ulceration. The most common cause is subclavian vein thrombosis proximal to a patent AVF, which causes a painful swollen arm. The diagnosis of venous hypertension can be made by duplex ultrasonography, magnetic resonance venography, or contrast venography. Surgical treatment of venous hypertension caused by a Brescia–Cimino shunt is ligation of the vein distal to the fistula and conversion to a functional venous end-to-side anastomosis. Treatment of other forms of venous hypertension involves control of venous hypertension either by ligation of the fistula or by relief of the subclavian vein stenosis, usually by balloon angioplasty with or without stenting (95).

SUMMARY The practice of vascular surgery is perhaps one of the more demanding of the medical specialties. Given the significant premorbid conditions of most patients that come to the attention of the vascular surgeon, and the physiological stress of the surgical procedures, complications can be frequent and significant. While not entirely unavoidable, a high degree of expertise and attention to detail can lessen the incidence of these adverse events.

REFERENCES 1. Hertzer NR, Beven EG, Young JR, et al. Coronary artery disease in peripheral vascular patients. A classification of 1000 coronary angiograms and results of surgical management. Ann Surg 1984; 199(2):223–233. 2. L’Italien GJ, Cambria RP, Cutler BS, et al. Comparative early and late cardiac morbidity among patients requiring different vascular surgery procedures. J Vasc Surg 1995; 21(6):935–944. 3. Mangano DT. Perioperative cardiac morbidity. Anesthesiology 1990; 72(1):153–184. 4. Hertzer NR. The natural history of peripheral vascular disease. Implications for its management. Circulation 1991; 83(suppl 2):I12–I19. 5. Cohn PF. Silent myocardial ischemia. Ann Intern Med 1988; 109(4):312–317. 6. Boucher CA, Brewster DC, Darling RC, Okada RD, Strauss HW, Pohost GM. Determination of cardiac risk by dipyridamole-thallium imaging before peripheral vascular surgery. N Engl J Med 1985; 312(7):389–394. 7. Cutler BS, Leppo JA. Dipyridamole thallium 201 scintigraphy to detect coronary artery disease before abdominal aortic surgery. J Vasc Surg 1987; 5(1):91–100.

Chapter 26: Complications of Vascular Surgery

8. Eagle KA, Coley CM, Newell JB, et al. Combining clinical and thallium data optimizes preoperative assessment of cardiac risk before major vascular surgery. Ann Intern Med 1989; 110(11):859–866. 9. Eagle KA, Singer DE, Brewster DC, Darling RC, Mulley AG, Boucher CA. Dipyridamole-thallium scanning in patients undergoing vascular surgery. Optimizing preoperative evaluation of cardiac risk. JAMA 1987; 257(16):2185–2189. 10. Eagle KA, Boucher CA. Cardiac risk of noncardiac surgery. N Engl J Med 1989; 321(19):1330–1332. 11. Jayr C, Matthay MA, Goldstone J, Gold WM, WienerKronish JP. Preoperative and intraoperative factors associated with prolonged mechanical ventilation. A study in patients following major abdominal vascular surgery. Chest 1993; 103(4):1231–1236. 12. Kispert JF, Kazmers A, Roitman L. Preoperative spirometry predicts perioperative pulmonary complications after major vascular surgery. Am Surg 1992; 58(8): 491–495. 13. Bartlett RH, Gazzaniga AB, Geraghty TR. Respiratory maneuvers to prevent postoperative pulmonary complications. A critical review. JAMA 1973; 224(7): 1017–1021. 14. Shanley CJ, Bartlett RH. The management of acute respiratory failure. Curr Opin Gen Surg 1994:7–16. 15. Barker SJ, Tremper KK. Pulse oximetry: applications and limitations. Int Anesthesiol Clin 1987; 25(3):155–175. 16. Solomon R, Werner C, Mann D, D’Elia J, Silva P. Effects of saline, mannitol, and furosemide to prevent acute decreases in renal function induced by radiocontrast agents. N Engl J Med 1994; 331(21):1416–1420. 17. Eneas JF, Schoenfeld PY, Humphreys MH. The effect of infusion of mannitol-sodium bicarbonate on the clinical course of myoglobinuria. Arch Intern Med 1979; 139(7):801–805. 18. Liebman PR, Menzoian JO, Mannick JA, Lowney BW, LoGerfo FW. Intraoperative arteriography in femoropopliteal and femorotibial bypass grafts. Arch Surg 1981; 116(8):1019–1021. 19. Miller A, Stonebridge PA, Jepsen SJ, et al. Continued experience with intraoperative angioscopy for monitoring infrainguinal bypass grafting. Surgery 1991; 109(3 Pt 1):286–293. 20. Yu A, Gregory D, Morrison L, Morgan S. The role of intra-operative duplex imaging in arterial reconstructions. Am J Surg 1996; 171(5):500–501. 21. Donaldson MC, Belkin M, Whittemore AD, Mannick JA, Longtine JA, Dorfman DM. Impact of activated protein C resistance on general vascular surgical patients. J Vasc Surg 1997; 25(6):1054–1060. 22. Bick RL. Disseminated intravascular coagulation and related syndromes: a clinical review. Semin Thromb Hemost 1988; 14(4):299–338. 23. Anderson RJ, Linas SL, Berns AS, et al. Nonoliguric acute renal failure. N Engl J Med 1977; 296(20):1134–1138. 24. Miller DC, Myers BD. Pathophysiology and prevention of acute renal failure associated with thoracoabdominal or abdominal aortic surgery. J Vasc Surg 1987; 5(3): 518–523. 25. Clowes A. Pathologic intimal hyperplasia as a response to vascular injury and reconstruction. In: Rutherford R, ed. Vascular Surgery. Vol. 1. 5th ed. Philadelphia: W.B. Saunders Company, 2000:408–418.

323

26. Sedwitz MM, Hye RJ, Stabile BE. The changing epidemiology of pseudoaneurysm. Therapeutic implications. Arch Surg 1988; 123(4):473–476. 27. Moore WS, Hall AD. Late suture failure in the pathogenesis of anastomotic false aneurysms. Ann Surg 1970; 172(6):1064–1068. 28. Corson JD, Leather RP, Naraynsingh V, Shah DM, Young HL, Karmody AM. In situ femoral, popliteal and distal bypass for limb salvage. Br J Surg 1983; 70(12):744–745. 29. Miller A, Stonebridge PA, Tsoukas AI, et al. Angioscopically directed valvulotomy: a new valvulotome and technique. J Vasc Surg 1991; 13(6):813–820. 30. Adcock OT Jr, Adcock GL, Wheeler JR, Gregory RT, Snyder SO Jr, Gayle RG. Optimal techniques for harvesting and preparation of reversed autogenous vein grafts for use as arterial substitutes: a review. Surgery 1984; 96(5):886–894. 31. Bernhard VM, Boren CH, Towne JB. Pneumatic tourniquet as a substitute for vascular clamps in distal bypass surgery. Surgery 1980; 87(6):709–713. 32. Blakeman BM, Littooy FN, Baker WH. Intra-arterial digital subtraction angiography as a method to study peripheral vascular disease. J Vasc Surg 1986; 4(2): 168–173. 33. Hertz SM, Baum RA, Owen RS, Holland GA, Logan DR, Carpenter JP. Comparison of magnetic resonance angiography and contrast arteriography in peripheral arterial stenosis. Am J Surg 1993; 166(2):112–116. 34. Robinson KD, Sato DT, Gregory RT, et al. Long-term outcome after early infrainguinal graft failure. J Vasc Surg 1997; 26(3):425–437. 35. Gloviczki P, Lowell RC. Lymphatic complications of vascular surgery. In: Rutherford R, ed. Vascular Surgery. Vol. 1. 5th ed. Philadelphia: W.B. Saunders Company, 2000:781–789. 36. Meyer JP, Durham JR, Schwarcz TH, Sawchuk AP, Schuler JJ. The use of sartorius muscle rotation-transfer in the management of wound complications after infrainguinal vein bypass: a report of eight cases and description of the technique. J Vasc Surg 1989; 9(5):731–735. 37. Tukiainen E, Biancari F, Lepantalo M. Deep infection of infrapopliteal autogenous vein grafts—immediate use of muscle flaps in leg salvage. J Vasc Surg 1998; 28(4):611–616. 38. Campbell DR, Bartlett FF. Postoperative Pseudomonas urinary tract infections as a source of bacterial contamination of an autogenous vein graft. J Vasc Surg 1987; 5(3):492–494. 39. Cherry KJ Jr, Roland CF, Pairolero PC, et al. Infected femorodistal bypass: is graft removal mandatory? J Vasc Surg 1992; 15(2):295–303. 40. Stonebridge PA, Miller A, Tsoukas A, et al. Angioscopy of arm vein infrainguinal bypass grafts. Ann Vasc Surg 1991; 5(2):170–175. 41. Mills JL, Bandyk DF, Gahtan V, Esses GE. The origin of infrainguinal vein graft stenosis: a prospective study based on duplex surveillance. J Vasc Surg 1995; 21(1):16–22. 42. Keagy BA, Schwartz JA, Kotb M, Burnham SJ, Johnson G Jr. Lower extremity amputation: the control series. J Vasc Surg 1986; 4(4):321–326. 43. Fearon J, Campbell DR, Hoar CS Jr, Gibbons GW, Rowbotham JL, Wheelock FC Jr. Improved results with

324

44.

45. 46.

47.

48.

49. 50. 51.

52.

53.

54.

55. 56. 57.

58.

59.

60.

Pomposelli et al.

diabetic below-knee amputations. Arch Surg 1985; 120(7):777–780. Gibbons GW, Wheelock FC Jr, Hoar CS Jr, Rowbotham JL, Siembieda C. Predicting success of forefoot amputations in diabetics by noninvasive testing. Arch Surg 1979; 114(9):1034–1036. Moller BN, Krebs B. Antibiotic prophylaxis in lower limb amputation. Acta Orthop Scand 1985; 56(4):327–329. Yeager RA, Moneta GL, Edwards JM, Taylor LM Jr, McConnell DB, Porter JM. Deep vein thrombosis associated with lower extremity amputation. J Vasc Surg 1995; 22(5):612–615. Feinglass J, Pearce WH, Martin GJ, et al. Postoperative and late survival outcomes after major amputation: findings from the Department of Veterans Affairs National Surgical Quality Improvement Program. Surgery 2001; 130(1):21–29. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1995; 273(18): 1421–1428. Muto PM, Welch HJ, Mackey WC, O’Donnell TF. Evaluation of carotid artery stenosis: is duplex ultrasonography sufficient? J Vasc Surg 1996; 24(1):17–22. Riles TS, Imparato AM, Jacobowitz GR, et al. The cause of perioperative stroke after carotid endarterectomy. J Vasc Surg 1994; 19(2):206–214. Hamdan AD, Pomposelli FB Jr, Gibbons GW, Campbell DR, LoGerfo FW. Perioperative strokes after 1001 consecutive carotid endarterectomy procedures without an electroencephalogram: incidence, mechanism, and recovery. Arch Surg 1999; 134(4):412–415. Rockman CB, Jacobowitz GR, Lamparello PJ, et al. Immediate reexploration for the perioperative neurologic event after carotid endarterectomy: is it worthwhile? J Vasc Surg 2000; 32(6):1062–1070. Pomposelli FB, Lamparello PJ, Riles TS, Craighead CC, Giangola G, Imparato AM. Intracranial hemorrhage after carotid endarterectomy. J Vasc Surg 1988; 7(2): 248–255. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med 1991; 325(7):445–453. Schauber MD, Fontenelle LJ, Solomon JW, Hanson TL. Cranial/cervical nerve dysfunction after carotid endarterectomy. J Vasc Surg 1997; 25(3):481–487. Das MB, Hertzer NR, Ratliff NB, O’Hara PJ, Beven EG. Recurrent carotid stenosis. A five-year series of 65 reoperations. Ann Surg 1985; 202(1):28–35. O’Donnell TF Jr, Rodriguez AA, Fortunato JE, Welch HJ, Mackey WC. Management of recurrent carotid stenosis: should asymptomatic lesions be treated surgically? J Vasc Surg 1996; 24(2):207–212. Galland RB, Wolfe JH. Mortality after elective abdominal aortic aneurysm repair: not where . . . but how many and by whom. Ann R Coll Surg Engl 1998; 80(5):339–340. Whittemore AD, Clowes AW, Hechtman HB, Mannick JA. Aortic aneurysm repair. Reduced operative mortality associated with maintenance of optimal cardiac performance. Ann Surg 1980; 192(3):414–421. Hagino RT, Taylor SM, Fujitani RM, Mills JL. Proximal anastomotic failure following infrarenal aortic reconstruction: late development of true aneurysms,

61.

62.

63.

64. 65.

66.

67. 68.

69.

70. 71. 72.

73.

74.

75.

76.

77.

pseudoaneurysms, and occlusive disease. Ann Vasc Surg 1993; 7(1):8–13. Moncure AC, Brewster DC, Darling RC, Atnip RG, Newton WD, Abbott WM. Use of the splenic and hepatic arteries for renal revascularization. J Vasc Surg 1986; 3(2):196–203. Farber MA, Carlin RE, Marston WA, Owens LV, Burnham SJ, Keagy BA. Distal thoracic aorta as inflow for the treatment of chronic mesenteric ischemia. J Vasc Surg 2001; 33(2):281–287. Marty-Ane CH, Alric P, Picot MC, Picard E, Colson P, Mary H. Ruptured abdominal aortic aneurysm: influence of intraoperative management on surgical outcome. J Vasc Surg 1995; 22(6):780–786. Kuhan G, Raptis S. ‘Trash foot’ following operations involving the abdominal aorta. Aust N Z J Surg 1997; 67(1):21–24. D’Angelo AJ, Puppala D, Farber A, Murphy AE, Faust GR, Cohen JR. Is preoperative cardiac evaluation for abdominal aortic aneurysm repair necessary? J Vasc Surg 1997; 25(1):152–156. Klonaris CN, Bastounis EA, Xiromeritis NC, Balas PE. The predictive value of dipyridamole-thallium scintigraphy for cardiac risk assessment before major vascular surgery. Int Angiol 1998; 17(3):171–178. Brewster DC, Franklin DP, Cambria RP, et al. Intestinal ischemia complicating abdominal aortic surgery. Surgery 1991; 109(4):447–454. Killen DA, Reed WA, Gorton ME, et al. Is routine postaneurysmectomy hemodynamic assessment of the inferior mesenteric artery circulation helpful? Ann Vasc Surg 1999; 13(5):533–538. Katz DJ, Stanley JC, Zelenock GB. Operative mortality rates for intact and ruptured abdominal aortic aneurysms in Michigan: an eleven-year statewide experience. J Vasc Surg 1994; 19(5):804–815. Lee ES, Kor DJ, Kuskowski MA, Santilli SM. Incidence of erectile dysfunction after open abdominal aortic aneurysm repair. Ann Vasc Surg 2000; 14(1):13–19. Cambria R. Thoracoabdominal aortic aneurysms. In: Rutherford R, ed. Vascular Surgery. Vol. 2 5th ed.. Philadelphia: W.B. Saunders Company, 2000:1303–1325. Gardner GP, Josephs LG, Rosca M, Rich J, Woodson J, Menzoian JO. The retroperitoneal incision. An evaluation of postoperative flank ‘bulge’. Arch Surg 1994; 129(7):753–756. Seeger JM, Pretus HA, Welborn MB, Ozaki CK, Flynn TC, Huber TS. Long-term outcome after treatment of aortic graft infection with staged extra-anatomic bypass grafting and aortic graft removal. J Vasc Surg 2000; 32(3):451–459. Gordon LL, Hagino RT, Jackson MR, Modrall JG, Valentine RJ, Clagett GP. Complex aortofemoral prosthetic infections: the role of autogenous superficial femoropopliteal vein reconstruction. Arch Surg 1999; 134(6):615–620. Locati P, Socrate AM, Costantini E. Surgical repair of infected peripheral graft and abdominal aortic aneurysm using arterial homograft. Ann Vasc Surg 2000; 14(2):176–180. Pegoraro M, Ferrero F, Palladino F, Baracco C, Merlo M, Bretto P. Aorto-enteric fistula as a rare complication of reconstructive surgery of the abdominal aorta. Minerva Cardioangiol 1984; 32(3):113–118. Buth J, Laheij RJ. Early complications and endoleaks after endovascular abdominal aortic aneurysm

Chapter 26: Complications of Vascular Surgery

78.

79.

80.

81. 82. 83.

84. 85.

86. 87.

repair: report of a multicenter study. J Vasc Surg 2000; 31(1 Pt 1):134–146. Schurink GW, Aarts NJ, van Bockel JH. Endoleak after stent-graft treatment of abdominal aortic aneurysm: a meta-analysis of clinical studies. Br J Surg 1999; 86(5): 581–587. Wain RA, Marin ML, Ohki T, et al. Endoleaks after endovascular graft treatment of aortic aneurysms: classification, risk factors, and outcome. J Vasc Surg 1998; 27(1):69–78. Zarins CK, White RA, Hodgson KJ, Schwarten D, Fogarty TJ. Endoleak as a predictor of outcome after endovascular aneurysm repair: AneuRx multicenter clinical trial. J Vasc Surg 2000; 32(1):90–107. Zarins CK, White RA, Fogarty TJ. Aneurysm rupture after endovascular repair using the AneuRx stent graft. J Vasc Surg 2000; 31(5):960–970. Zarins CK, White RA, Moll FL, et al. The AneuRx stent graft: four-year results and worldwide experience 2000. J Vasc Surg 2001; 33(suppl 2):S135–S145. Lee WA, O’Dorisio J, Wolf YG, Hill BB, Fogarty TJ, Zarins CK. Outcome after unilateral hypogastric artery occlusion during endovascular aneurysm repair. J Vasc Surg 2001; 33(5):921–926. Mildner A, Hilbe G. Complications in surgery of varicose veins. Zentralbl Chir 2001; 126(7):543–545. Goldman M. Complications and adverse sequelae of sclerotherapy. In: Goldman MP, Bergan JJ, eds. Sclerotherapy: Treatment of Varicose and Telangiectatic Leg Veins. 3rd ed. St. Louis: Mosby, 2001:191–240. Proctor MC. Indications for filter placement. Semin Vasc Surg 2000; 13(3):194–198. Ebaugh JL, Chiou AC, Morasch MD, Matsumura JS, Pearce WH. Bedside vena cava filter placement guided

88. 89.

90.

91. 92. 93.

94.

95.

325

with intravascular ultrasound. J Vasc Surg 2001; 34(1): 21–26. Greenfield LJ, Proctor MC. Filter complications and their management. Semin Vasc Surg 2000; 13(3): 213–216. Krieter DH, Rumpf KW, Muller GA, Werner GS. Paradoxical cerebral embolism during fibrinolytic therapy in deep venous thrombosis of the leg and pulmonary embolism. Dtsch Med Wochenschr 1994; 119(22):825. Haimovici H. Arterial embolism with acute massive ischemic myopathy and myoglobinuria: evaluation of a hitherto unreported syndrome with report of two cases. Surgery 1960; 47:739–747. Patman RD, Thompson JE. Fasciotomy in peripheral vascular surgery. Report of 164 patients. Arch Surg 1970; 101(6):663–672. Haimov M, Schanzer H, Skladani M. Pathogenesis and management of upper-extremity ischemia following angioaccess surgery. Blood Purif 1996; 14(5):350–354. Brooks JL, Sigley RD, May KJ Jr, Mack RM. Transluminal angioplasty versus surgical repair for stenosis of hemodialysis grafts. A randomized study. Am J Surg 1987; 153(6):530–531. Mennes PA, Gilula LA, Anderson CB, Etheredge EE, Weerts C, Harter HR. Complications associated with arteriovenous fistulas in patients undergoing chronic hemodialysis. Arch Intern Med 1978; 138(7): 1117–1121. Bhatia DS, Money SR, Ochsner JL, et al. Comparison of surgical bypass and percutaneous balloon dilatation with primary stent placement in the treatment of central venous obstruction in the dialysis patient: one-year follow-up. Ann Vasc Surg 1996; 10(5):452–455.

27 Acute Complications of Cardiovascular Surgery and Trauma Riyad C. Karmy-Jones Division of Cardiothoracic Surgery, University of Washington, and Thoracic Surgery, Harborview Medical Center, Seattle, Washington, U.S.A. Edward M. Boyle Division of Cardiothoracic Surgery, University of Washington, Seattle, Washington, U.S.A. John C. Mullen Department of Cardiac Sciences, University of Alberta, Edmonton, Canada

The spectrum of cardiac surgery has changed substantially in the past decade. The complexity of cases has increased, with increases in the number of repeated procedures and the number of older and sicker patients referred for surgery. At the same time, surgical practice has changed to include minimal access approaches, off-pump cardiopulmonary bypass techniques, bypass modifications to alleviate inflammatory responses, biochemical investigations to ameliorate ischemia–reperfusion injury, and improved artificial valves (such as stentless valves). These changes have substantial implications for the long-term outcome of patients undergoing surgical interventions, and a detailed discussion of these implications is beyond the scope of this chapter. However, the prevention and treatment of acute complications remain important, particularly as the surgical population becomes older and more likely to have comorbid conditions (1). Bojar noted that after coronary bypass procedures, approximately 10% of patients will experience a complication that will require treatment, prolonged hospitalization, or both (Table 1) (1). The incidence of traumatic cardiovascular injury continues to increase (2,3). Advances in prehospital care mean that more severely injured patients will arrive at the hospital alive, although in extremis, and that more of them will survive. At the same time, newer diagnostic and therapeutic approaches have changed the management of these traumatic injuries, e.g., the use of endovascular stent grafts for the nonoperative treatment of patients with blunt aortic injury (4). This chapter will focus on the immediate and early postoperative complications that may be encountered and should be anticipated in association with treatment of traumatic cardiovascular injuries. The primary emphasis is on the technical aspects and acute management of cardiovascular diseases.

TECHNICAL COMPLICATIONS RELATED TO CARDIOPULMONARY BYPASS Iatrogenic Arterial Dissection Iatrogenic dissection of the ascending aorta occurs in as few as 0.12% of cases in which cardiopulmonary bypass is used (5). Factors predisposing patients to undergo iatrogenic dissection include hypertension and connective tissue disorders, but the primary risk factor is calcification of the aorta (6). Dissection can occur from cannulation, clamping, or cardioplegia, or because of the creation of proximal anastomotic sites (7). In two-thirds of cases, dissection of the ascending aorta occurs acutely, with a rapidly expanding hematoma, bleeding, and, eventually, decreased return from the venous pump, which reflects blood loss (8). In the remaining third of cases, dissection occurs postoperatively. The overall mortality rate associated with acute dissection of the ascending aorta is nearly 30% (8). When the dissection is discovered intraoperatively, the Table 1 Complications After Coronary Bypass Procedures Complications Arrhythmias Atrial Ventricular Infectious complications Leg wound Sternal wound Myocardial infarction Respiratory failure Reoperation for bleeding Stroke Gastrointestinal complications Renal failure Source: From Ref. 1.

Average incidence (%) 30 5 5 3 5 5 3 2 2 2

328

Karmy-Jones et al.

mortality rate is approximately 20%; if it is recognized postoperatively, the mortality rate increases to 50% (7). Management of this condition requires removal of the cause (if it occurs at the site of arterial cannulation, the cannula must be replaced at another site), deepening hypothermia, and repair. Repair may be simple, using gelatin–resorcin–formalin glue or pledgeted sutures, but aortic tube grafts may also be required (6). The symptoms of postoperative dissection of the ascending aorta are similar to those of acute aortic dissection, including tamponade (6). The primary methods of preventing this complication are to ensure that the surgical procedure avoids areas of aortic calcification that have been defined by palpation and echocardiography and to carefully control blood pressure during manipulation of the aorta. Aortic dissection may occur more commonly after cannulation of the femoral artery (9). Manifestations can include high pressures, which reflect obstruction, or acute rupture of the vessel. More common complications after cannulation of the femoral artery, however, are occlusion or emboli, which can cause postoperative limb ischemia. As is true of the ascending aorta, palpation of the femoral artery at the time of surgery may indicate that one site is safer than another. Limited arterial dissections may be medically managed, but extensive dissections require surgical repair.

Rupture of the Coronary Sinus Rupture of the coronary sinus occurs during retrograde cardioplegia in fewer than 1% of cases. It manifests itself as vigorous retrocardiac bleeding into the pericardial sac (10,11). Management depends upon how large the rupture is, when it occurred, and whether antegrade cardioplegia is possible. If antegrade cardioplegia is possible, reasonable treatment consists of cooling, antegrade cardioplegia, and subsequent cardioplegia through successive vein grafts, followed by repair at the end of the case. If antegrade cardioplegia is not a reliable option (e.g., in the case of reoperation using a patent mammary artery for repair, or because of severe obstructive lesions in the coronary arteries), then one of two treatments may be used: rapid cooling and bypass with the patient under fibrillatory arrest to ensure adequate venting of the heart and with antegrade cardioplegia through successive vein grafts, or immediate repair over the catheter. Although immediate repair may initially seem to be the ‘‘best’’ approach, it must be remembered that the repair will have to withstand the pressures created by delivering cardioplegia and that the catheter tip will most likely be advanced past the tributary that drains a substantial portion of the right ventricle and atrium; thus, less protection will be provided to these areas. If this option is chosen, the first anastomosis should be to the right coronary distribution. The actual repair can be performed over the catheter, which acts as a stent. Primary repair, vein patch, and, occasionally, Gore-Tex1 (W.L. Gore & Associates,

Inc., Newark, Delaware U.S.A.) patch repairs that connect to the right atrium have been used (10,11).

Venous Air Lock Occasionally, a large amount of air will be entrained into the venous cannula. This complication most commonly occurs during operations on the right side of the heart, such as closure of a patent foramen ovale or tricuspid valve repair, when snares around the cannula loosen. If a substantial air lock occurs, the pump will stop automatically until the reservoir level has been replenished. Management includes the use of ‘‘sucker bypass’’ to refill the reservoir, purging the venous lines with saline solution, and repositioning the cannula and securing its snares.

Systemic Air Embolism One large study has documented the occurrence of systemic air embolism in 458 of 575,000 cases (12). The impact of this complication is predominantly noted postoperatively; left hemiparesis is observed as a consequence of embolization of the right carotid artery (13). Massive systemic air embolism is associated with an immediate mortality rate of 22% (14). Air embolism has many potential causes (Table 2). The use of membrane oxygenators rather than bubble oxygenators has reduced the risk of air embolism, particularly the risk of a drop in the reservoir level (15). In general, there is no substitute for carefully checking the pump circuits and connections and for having an organized plan for removing air, particularly during open heart procedures. The left atrial appendage should be invaginated, the lungs should be ventilated with restricted venous return, and residual blood should be aspirated. If the left atrium has been opened, a catheter placed across the mitral valve will prevent Table 2 Sources of Systemic Air Embolism Bypass machine Drop in reservoir level Pressurization of cardiotomy Reservoir reversal of pump-head rotation Disconnection of oxygenator Inadequate debubbling of lines Venting Excessive suction on atrial vent lines Air entering aorta retrograde via nonoccluded coronary arteries when excessive aortic venting is applied Heart Retained air in atrial appendage Retained air in trabeculae Retained air in pulmonary veins Unexpected resumption of cardiac activity when heart is open Unexpected patent foramen ovale when the patient is on cardiopulmonary bypass but heart is not arrested Other Continued use of intra-aortic balloon pump, which can suck air when arch is opened Air introduced from venous lines in setting of patent foramen ovale Entry via partial occlusion clamp when proximal anastomoses are performed

Chapter 27: Acute Complications of Cardiovascular Surgery and Trauma

ventricular ejection until deairing is complete. Aortic clamping should be maintained until all air bubbles have been removed. If air is pumped into the patient, a sequence of events must be undertaken quickly (Fig. 1). Retrograde perfusion at 20 C, with flows of 1 to 2 L/min in the average adult, can prevent injury in 61% of cases (12). Maintaining hypothermia at 20 to 22 C will reduce cerebral metabolic requirements and encourage gas solubility (13). Administering high-dose steroids, mannitol, glycerol, or some combination of the three has been shown to reduce the likelihood of injury (16,17). Increasing cerebral perfusion by augmenting flow (6.0–6.6 L/min) and blood pressure (including administering vasopressor drugs) may help purge the cerebral vasculature. Cerebral perfusion can be further enhanced by temporarily clamping the descending thoracic aorta (13). Thiopental (40 mg/kg) has been used to quiet brain activity until bypass is complete (18). Although the use of thiopental is attractive, it is associated with substantial negative effects, including myocardial depression and a delay in early return of consciousness, which would otherwise be the best indicator of a good prognosis (19). If the necessary equipment is available, hyperbaric oxygenation may be a useful adjunct (20).

Coronary Air Embolism Air may be introduced into the coronary arteries by any of the mechanisms described above, but it may also be inadvertently injected through the cardioplegia lines. When the source of the air is systemic, the right coronary artery, which is more anterior, is

329

most often involved. Heart block, right ventricular dysfunction, or both will result (13). Prevention includes all of the techniques described above. In addition, removing air from the aortic root by means of a clamp placed across the valve and using a venting needle in the anterior aorta during slow removal of the cross clamp have proved useful (21). If air introduction is recognized, purging the coronary arteries by elevating aortic root pressure or by retrograde flushing will assist in return of function (13).

Protamine Reactions Adverse reactions to protamine, which is used to reverse the effects of heparin, may be the most common intraoperative complication associated with cardiopulmonary bypass (22,23). Various reactions have been characterized and loosely categorized as ‘‘predictable’’ (nonspecific) hypotension and ‘‘idiosyncratic’’ reactions (24). Nonspecific hypotension appears to be related to direct or histamine-mediated vasodilatation rather than to a direct myocardial depressant effect (25). Slow administration of protamine over a period of 10 to 15 minutes will reduce the incidence and magnitude of the hypotension, but the condition may still occur. Treatment includes temporarily stopping protamine administration and, if the reaction is serious, administering short-acting vasoconstrictor drugs. Idiosyncratic reactions include an anaphylactic reaction mediated by immunoglobulins and an anaphylactoid reaction not mediated by immunoglobulins. Both reactions result in histamine release, which leads to bronchospasm, edema, and decreased systemic and pulmonary vascular resistance; this sequence, in turn, leads to low systemic and pulmonary pressures. These reactions may be more likely when patients have received neutral protamine Hagedorn insulin or have fish allergies. Initial management includes reinstitution of bypass if the reaction is too severe to allow time for medical intervention to take effect. Medical treatments include volume expansion, administration of epinephrine, steroids, and antihistamines (both H1 and H2 blockers), and administration of aminophylline if bronchospasm is persistent. Catastrophic pulmonary vasoconstriction differs from protamine reactions in that systemic hypotension is coupled with pulmonary hypertension and acute right heart failure. Heparin–protamine complexes are believed to cause the release of thromboxane by means of the complement cascade (26,27). Treatment includes repeated administration of heparin, reinstitution of bypass, and administration of pulmonary vasodilator drugs (28).

Atrial Cannulation During Left Heart Bypass

Figure 1 Management of air embolism.

The left atrial appendage is often extremely friable and can tear with catastrophic results. In addition, passing a cannula into the left atrium is associated with the risk of posterior atrial rupture. Furthermore, in the setting

330

Karmy-Jones et al.

of traumatic injury, hematoma can obscure landmarks and cardiac irritability can increase the risk of malignant arrhythmias (29). Accessing the left atrium extrapericardially through the pulmonary veins is associated with a marked reduction in the incidence of complications. Karmy-Jones et al. (30) noted a 37% incidence of significant complications (including ventricular fibrillation leading to arrest) when the atrial appendage was cannulated for bypass during repair of traumatic aortic rupture, as opposed to 7% when the pulmonary veins were used.

COMPLICATIONS OF REPEAT CARDIAC OPERATION The cumulative incidence of repeated coronary bypass is approximately 11% by 10 years after the initial bypass procedure. The incidence increases if internal mammary arteries (IMAs) were not used or if incomplete revascularization was performed (31). Indications for repeated operation [as opposed to percutaneous transluminal coronary angioplasty (PTCA)] include (32): 1.

2. 3.

Vein graft stenosis of more than 50%, occurring more than five years after initial surgery, particularly if large areas of myocardium [usually the left anterior descending artery (LAD)] are at risk Left main or multiple-vessel disease and decreased left ventricular function, or proximal LAD stenosis in large vessels with no patent grafts Disabling angina

The operative mortality rate associated with second coronary bypass is approximately 3% to 4%, whereas that associated with a third procedure is 7% (33,34). Independent risk factors are the age of the patient being more than 70 years and diminished ventricular function (33,34). However, encouraging results indicate that the five-year survival rate after a third coronary artery bypass operation is 84% and the 10-year survival rate is 66% (35). Technical issues related to repeated surgery include the risk of catastrophic bleeding when the sternotomy is reopened, potential damage to a patent graft (particularly the IMA), difficulty in achieving adequate myocardial protection, and atheroembolism in patent grafts. Because of these issues, the incidence of perioperative infarction is two to three times greater for repeated procedures than for primary surgery (1). Most repeated operations will require sternotomy. Occasionally, minimally invasive approaches can be used, for example, if only the LAD needs to be grafted and the IMA was not used originally (32). Preparation includes evaluating potential conduits (right internal mammary and radial arteries, available arm and leg veins, gastroepiploic artery, etc.). When sternotomy is performed during a repeated procedure, the initial concern is avoiding tears of the right ventricle. The density of adhesions between the sternal incision and the surface of the heart (particularly the right ventricle and aorta) may be estimated by viewing the lateral chest radiograph. If little or no

retrosternal space is seen, the bypass lines should be brought into the operative field. In some instances, cannulating the femoral or axillary artery may be advisable before the sternotomy is begun. The axillary artery is less atherosclerotic than the femoral artery and provides antegrade rather than retrograde flow; these factors may reduce the risk of embolization (36). Defibrillator paddles should be placed before the procedure is started. The sternal wires are left intact posteriorly, and the approach starts cautiously from the xiphoid level, with elevation provided by pulling up on the costal margin. As the dissection proceeds, an oscillating saw is used to open the sternum carefully by steps. If adhesions are particularly dense, a small right anterolateral thoracotomy may allow the surgeon to place a hand under the sternum (32). Should catastrophic bleeding occur, bypass can be initiated by instituting ‘‘sucker’’ bypass, draining venous blood by pump suckers, and returning flow by the femoral cannula. If the left IMA (LIMA) is patent, exposure may be assisted by entering the left pleural space at the level of the diaphragm and dissecting superiorly. The pericardium can be divided to the left of the LAD, creating a flap of tissue containing the LIMA if it is patent. Once conduits have been prepared, cannulation of the right atrium can be performed, but care must be taken not to manipulate patent grafts. Retrograde cardioplegia will allow adequate and uniform protection in most cases, with the added advantage of not creating embolization in partially patent grafts. Because of the theoretical disadvantage of decreased right ventricular protection, and because performing distal anastomoses can be difficult when continuous warm cardioplegia is used, many surgeons prefer to use intermittent retrograde and antegrade cardioplegia, although the latter can be used less frequently if it is clear that retrograde cardioplegia results in adequate cooling and arrest (32). In addition, patent IMA grafts must be clamped during arrest to prevent rewarming of the heart. As revascularization proceeds, antegrade cardioplegia can be provided through each of the newly anastomosed vein grafts. Injury to a patent LIMA occurs in 5% to 8% of cases. In nearly 60% of these cases, the graft can be preserved by using a variety of techniques, ranging from repair to repeated bypass with a vein graft (37). The perioperative infarction rate is increased (40%) with higher overall mortality (8%). The risk can be reduced during the first bypass procedure by positioning the LIMA graft in the left chest away from the sternal table (37).

POSTOPERATIVE MYOCARDIAL INFARCTION AND CORONARY SPASM The incidence of postoperative myocardial infarction varies depending on the circumstances in which the surgery occurred. The incidence is 2% to 3% among patients with stable angina, 5% to 10% among those

Chapter 27: Acute Complications of Cardiovascular Surgery and Trauma

with unstable angina, and 30% to 50% among those undergoing emergency bypass after a failed PTCA procedure (1). Others have reported the occurrence of myocardial ischemia within six hours of operation in as many as 40% of cases and of infarction in 5% to 25% of cases (38,39). The diagnosis can be suggested by new-onset low cardiac output, electrocardiography (ECG) changes, creatinine phosphokinase (CPK)-2–myocardial band (MB) activity greater than 50 U/L, or elevated levels of the cardiac isoenzymes troponin T or troponin I. Patients at the highest risk of infarction and death are those whose condition does not respond to preoperative intra-aortic balloon pump (IABP) support (40). For these patients, outcomes are related to timing, with mortality rates as low as 3% if bypass can be instituted within four hours of the acute event (41,42). In addition, coronary atheroembolism, particularly in repeated operations, remains a risk. Finally, the inflammatory cascade initiated by cardiopulmonary bypass may play a role in the occurrence of postoperative myocardial infarction. Aldea (43) noted that the perioperative infarction rate decreased from 3.96% to 0.99% when heparinbonded circuits were used. For patients with severe ischemia at the start of the operation, an IABP should be used, bypass and hypothermia should be instituted early to decrease metabolic demands, and warm-blood cardioplegia should be delivered to areas of the myocardium as a resuscitative tool (44). Postoperative management should be simplified if complete revascularization was performed because the issues of coronary perfusion pressure are not so prevalent. Management is directed at decreasing workload by using an IABP and, if necessary, by administering pressor drugs with some vasodilating properties. Coronary artery spasm occurs after myocardial revascularization in only approximately 0.1% of cases (45). This complication is believed to be due to a combination of injury to coronary arteries during operation and production of thromboxane A2 during bypass (46). Coronary artery spasm manifests itself as acute ventricular dysfunction coupled with diffuse ECG changes, often with catastrophic cardiovascular collapse. It usually occurs shortly after bypass has been terminated and carries a high mortality rate (46,47). Initial management requires administration of heparinization, reinstitution of cardiopulmonary bypass, and rapid assessment of graft patency. Injecting nitroglycerin directly into the grafts will quickly reverse the condition, but sublingual or systemic administration of calcium channel blockers (e.g., diltiazem 5–25 mg/hr intravenously) is also effective (48). Administration of intravenous calcium channel blockers should be maintained after the operation to prevent the recurrence of coronary artery spasm, although there are as yet no clear guidelines for safely discontinuing these drugs. Administering calcium channel blockers indefinitely is a reasonable treatment for patients who have experienced coronary artery spasm.

331

POSTOPERATIVE BLEEDING AND TAMPONADE Persistent hemorrhage will necessitate a return to the operating room for 2% to 5% of patients (48). Patients at increased risk for this complication are primarily those receiving preoperative antiplatelet or thrombolytic agents and those who have undergone a repeated procedure, but multiple factors can play a role (Table 3). Postoperative bleeding should always be considered surgically related until proven otherwise (49). This bleeding will generally manifest itself as persistent drainage from the mediastinal tubes or as an increase in pleural effusion. Surgical treatment is indicated if bleeding exceeds 400 mL/hr in any hour, if it exceeds 200 mL/hr for two to four hours, or if there is any hemodynamic instability. Nonspecific interventions include avoiding hypertension; some surgeons recommend the use of high levels of positive end-expiratory pressure (PEEP) (up to 20 cmH2O) to ‘‘tamponade’’ bleeding. However, the prophylactic use of PEEP does not appear to reduce the incidence of repeated operation and can potentially complicate the postoperative course by aggravating cardiac dysfunction through physiologic tamponade or by putting pressure on an internal mammary graft (48). Apart from surgical bleeding, multiple coagulation disorders may be quickly assessed by using a thromboelastogram; however, the results of this procedure do not reliably predict the likelihood of bleeding complications preoperatively (50,51). The most common coagulopathic cause of bleeding is platelet dysfunction (48). Inciting factors include hypothermia, contact with inert surfaces during bypass, and platelet consumption (43). The occurrence of platelet dysfunction can be combated with combinations of rewarming, administering D-amino-D-arginine vasopressin (DDAVP) at a dose of 0.3 mg/kg, and transfusing 6 to 10 platelet packs to maintain a platelet

Table 3 Causes of Postoperative Bleeding Preoperative factors Reoperation Preoperative anticoagulation Antiplatelet agents Thrombolytic agents Cyanotic heart disease Liver dysfunction Coagulation disorders Von Willebrand’s disease Uremia Intraoperative factors Inadequate surgical hemostasis (most common cause) Long pump run Core cooling to 65 yr) Insulin-requiring diabetes mellitus with end-organ complications Current or recent diverticulitis Active infection Recent or unresolved pulmonary infarction Cachexia Severe osteoporosis Psychosocial instability and lack of adequate support systems Active substance abuse

transplantation, heterotopic cardiac transplantation is more technically complex, requires a longer ischemic time, and is associated with a greater potential for thrombus formation in the dilated, poorly functioning native heart and with a higher incidence of mitral regurgitation in the donor heart. The procedure is contraindicated for recipients with right-sided heart failure and tricuspid insufficiency.

INTRAOPERATIVE COMPLICATIONS Immediate Primary Graft Failure Immediate primary graft failure accounts for approximately 5% of the deaths that occur immediately after cardiac transplantation. Aside from technical errors (which are rare) such as inadvertent occlusion of the coronary arteries or veins with sutures, the main cause of primary graft failure is ischemia-reperfusion injury. The current myocardial preservation technique, which involves hypothermic cardiac arrest (with various cardioplegic solutions) and static hypothermic storage, yields an acceptable total ischemic time of approximately five to six hours. An ischemic time of more than five hours is associated with poor survival rates (19). Primary graft failure is more likely to occur when donors are older than 45 years or have left-ventricular hypertrophy, when ischemic time exceeds five hours, or when the donor-torecipient weight ratio is less than 0.7 (20,21).

Hyperacute Rejection Hyperacute rejection occurs within minutes to hours after transplantation and results when preformed antibodies in the recipient are directed against donor human leukocyte antigens (HLAs). Hyperacute rejection may also occur when the donor and recipient

blood groups are incompatible (22,23) or when antibodies against donor tissue antigens are present on the vascular endothelial cells (24). When this complication occurs, the graft looks hemorrhagic and edematous, with minimal ventricular contractility, shortly after being reperfused. The patient usually shows signs of disseminated intravascular coagulation. To date, because there is no effective treatment for hyperacute rejection, prevention of the complication is emphasized. Before transplantation, each candidate is tested for the presence of preformed antibodies against a T-lymphocyte panel that represents most HLAs [panel-reactive antibodies (PRA)]. When test results show that the recipient reacts to more than 10% of the T-lymphocytes on the panel (PRA > 10%), most transplant centers prospectively crossmatch donor lymphocytes with recipient serum before transplantation. Because of the dire consequences of hyperacute rejection, positive results from leukocyte crossmatching are generally considered a contraindication to heart transplantation, even though these results are not an accurate predictor of hyperacute rejection (24,25).

EARLY POSTOPERATIVE COMPLICATIONS In the early postoperative period, recipients of heart transplants are given treatment similar to that given to any patient recovering from cardiac surgery, except for some specific treatment for problems related to the use of immunosuppressive agents. Recipients should be weaned from mechanical ventilatory support and inotropic drugs as quickly as possible. Early mobilization and physical therapy are essential.

Mediastinal Hemorrhage The management of hemorrhage in heart transplant recipients is similar to the management of this complication in any patient who has undergone open-heart surgery and cardiopulmonary bypass. After cardiopulmonary bypass, approximately 10% to 20% of patients require transfusion of blood components; fewer than 3% require reexploration. Problems such as inadequate heparin reversal, heparin rebound, hypothermia, quantitative or qualitative platelet defects, depletion of coagulation factors, and fibrinolysis should be addressed. Mediastinal exploration will be necessary when bleeding occurs at a rate of 500 cc/hr for one hour, 400 cc/hr for two hours, or 300 cc/hr for three hours, or when there is evidence of cardiac tamponade (26). The importance of early reexploration for mediastinal bleeding cannot be overstated because the morbidity and mortality rates associated with early reexploration are lower than those associated with delayed reexploration (27,28).

Ventricular Dysfunction Because of reperfusion injury, the function of the transplanted heart is usually depressed in the

Chapter 28: Complications of Cardiac Transplantation

immediate postoperative period. During the first four to five days after transplantation, the normal inverse relationship between heart rate and stroke volume is not seen (29). Stroke volume remains fixed because of a restrictive hemodynamic pattern. Recipients of cardiac transplants usually require chronotropic support for the first several days after transplantation because of sinoatrial node dysfunction (30). Isoproterenol or dobutamine is commonly used in the immediate posttransplant period to increase inotropy and chronotropy, reduce pulmonary vascular resistance, and improve ventricular diastolic relaxation. Aminophylline also improves heart rate in cardiac transplant recipients (31). Severe pulmonary hypertension may require the concomitant use of inotropic agents and pulmonary vasodilators such as milrinone, nitroglycerine, or inhaled nitric oxide (32–34). Cardiac function and sinoatrial node activity will usually return to normal within seven days. Because the transplanted heart is completely denervated, circulating catecholamines are responsible for the increased chronotropic and inotropic responses to exercise (35,36).

Right-Ventricular Failure Right-ventricular failure is a frequent cause of early morbidity and mortality after cardiac transplantation. This complication usually results from ischemiareperfusion injury or from high pulmonary vascular resistance. In recipients with elevated pulmonary vascular pressures, the normal donor right ventricle may be unable to meet the sudden demand of a high afterload. Long ischemic time and a donor-torecipient size discrepancy (donor body weight lower than recipient body weight) exacerbate this problem. The goal of medical therapy is to increase rightventricular inotropy, with drugs such as type III phosphodiesterase inhibitors (milrinone, amrinone), isoproterenol, and dobutamine, and to reduce pulmonary vascular resistance, with drugs such as nitroglycerin, prostacyclin, and inhaled nitric oxide. In cases of severe right-heart failure, ventricular-assist device or extracorporeal membrane oxygenation support is needed.

Left-Ventricular Failure Left-ventricular failure is a rare complication that usually occurs because of prolonged ischemic time and poor myocardial preservation. Occasionally, preexisting coronary artery disease or embolization of air or particulate matter to the coronary arteries may cause ventricular failure. As is true for rightventricular failure, left-ventricular failure should be treated by increasing ventricular inotropy and reducing ventricular afterload by medical or mechanical means (e.g., intraaortic counterpulsation). For severe cases (approximately 5% of all cases), a ventricularassist device should be used.

359

Electrophysiologic Dysfunction Arrhythmia is common during the early postoperative period and occurs in as many as 70% of patients receiving cardiac transplants (37). Sinus bradycardia, junctional rhythm, and junctional escape rhythms are the most common arrhythmias. These conditions are usually due to injury to the sinus node during surgery but may also herald the onset of acute rejection. Chronic use of beta-blockers and amiodarone before transplantation are associated with prolonged sinus bradycardia, whereas the administration of aminophylline accelerates recovery from sinus bradycardia (31,38). The bicaval technique is reportedly associated with a lower incidence of atrial arrhythmias and sinus node dysfunction than the biatrial approach (39,40). Normal sinus node activity usually returns several weeks after transplantation; therefore, implantation of a permanent pacemaker should be delayed for at least four weeks and after an adequate trial of aminophylline (41,42). Permanent perioperative right-bundle branch block occurs frequently after cardiac transplantation. However, premature atrial contractions, atrial flutter, and atrial fibrillation usually herald the onset of acute rejection. Premature ventricular contractions occur in more than half of patients during the early posttransplant period and are less frequently associated with acute rejection. However, complex ventricular ectopy can be associated with chronic allograft vasculopathy and is a risk factor for sudden death (30,43). Because the transplanted heart is denervated, it will not respond to cardiac drugs in the same way the normal heart does. Therefore, it is important to understand the effects of cardiac drugs on the transplanted heart before therapeutic intervention is considered. The following general rules (44) should be observed: (i) the transplanted heart is not affected by drugs that act via the autonomic nervous system; (ii) cardiac receptors on the transplanted heart are more responsive to beta-agonists than those on the native heart; and (iii) drugs with negative inotropic or chronotropic effects are not well tolerated by heart transplant recipients because these effects may not be offset by the autonomically mediated increase in myocardial contractility and sympathetic tones (30,45). Digoxin exerts its electrophysiologic effect primarily on the autonomically mediated function of sinoatrial and atrioventricular nodes, and therefore this drug has very little therapeutic value in treating supraventricular tachycardia or atrial fibrillation in the denervated transplanted heart. Similarly, atropine, edrophonium, and vagal maneuvers will have no effect on the transplanted heart. Because of their direct electrophysiologic effects, quinidine and procainamide are effective treatments for atrial and ventricular arrhythmias in transplant recipients. The transplanted heart is especially sensitive to adenosine; therefore, this drug must be used carefully and at low doses (one-fifth to one-third the usual dose), if at all (46). Aminophylline

360

Qi et al.

can correct sinus bradycardia in transplant recipients by blocking extracellular A1 adenosine receptors and has been used to reverse rejection-induced bradycardia or to treat the bradycardia that may occur in the immediate postoperative period (31,38).

Acute Renal Failure Cardiac transplant recipients are quite susceptible to acute renal failure. The high incidence of this complication after heart transplantation results from low renal reserve because of hypertension and prolonged congestive heart failure, the nonpulsatile flow produced by cardiopulmonary bypass during transplantation, and toxicity associated with immunosuppressive agents. Nephrotoxicity is one of the common side effects of calcineurin inhibitors (e.g., cyclosporine and tacrolimus) (47) and has both acute and chronic components. The acute component is the primary cause of acute renal failure in the early postoperative period. Measures that can prevent this toxicity include infusion of a low dose (a renal dose) of dopamine, intravenous infusion of furosemide, hydration, and minimal use of vasoconstrictive agents (48). The incidence of acute renal failure in the early postoperative period may be reduced by using immunosuppressive agents that are not nephrotoxic (e.g., antithymocyte antibodies) and by delaying the administration of calcineurin inhibitors until adequate renal function has returned. The chronic component of nephrotoxicity results from the long-term use of calcineurin inhibitors and is associated with a gradual increase in the serum creatinine concentration and a decrease in renal function. The severity of renal dysfunction may be reduced by administering a lower dose of cyclosporine or tacrolimus in combination with other immunosuppressive agents. Because nonsteroidal anti-inflammatory agents (e.g., ketorolac tromethamine or ibuprofen) exacerbate the nephrotoxicity of calcineurin inhibitors, heart transplant recipients should not use these agents for pain control, especially in the perioperative period when the incidence of acute renal failure is high.

Stroke The risk of stroke for cardiac transplant recipients is similar to that of any patient undergoing cardiopulmonary bypass (approximately 2.5%). Stroke may result from low perfusion during cardiopulmonary bypass in a patient with preexisting compromised cerebral circulation (carotid or intracerebral arterial disease) or from embolization of air or particulate matter (fragments of pericardial fat or myocardial tissues, or cholesterol debris from the diseased aorta). Thrombus at the atrial suture line in the transplanted heart is a common source of emboli in the early postoperative period. To minimize thrombus at the suture line, surgeons must take care to evert the atrial

anastomosis and approximate the endocardial layers of the donor and recipient vessels.

Acute Rejection Despite advances in immunosuppression during the past five decades, acute allograft rejection is still a serious problem after cardiac transplantation. Approximately 70% of recipients will experience an episode of moderate acute rejection (grade 3A) during their lifetime. Acute rejection is mediated by T-lymphocytes and commonly occurs during the first six months after transplantation. Moderate acute rejection may be present without evidence of hemodynamic instability or echocardiographically documented decrease in ventricular function. Therefore, recipients of heart transplants require routinely performed surveillance endomyocardial biopsy for histologic determination of rejection. The standardized grading system for cardiac rejection produces a score ranging from 0 (no rejection) to 4 (severe rejection); treatment of rejection is in part determined by this grade (49,50). Moderate acute rejection (grade 3A or higher) is treated with boluses of methylprednisolone (1 g/day for adults, and 10–20 mg/kg/ day for children, for three days); for mild acute rejection, augmenting the level of baseline immunosuppression will usually suffice (grade 1B–2). Lympholytic therapy with antithymocyte globulin or antilymphocyte antibody is reserved for cases of steroid-resistant rejection.

Accelerated Acute Rejection Accelerated acute rejection usually occurs 24 to 72 hours after transplantation and is mediated by antibodies to donor HLA. It occurs in recipients who have been sensitized to donor antigens but whose level of antidonor antibodies at the time of transplantation is too low to trigger hyperacute rejection. After transplantation, the immune system of a sensitized patient is reexposed to the sensitizing antigens from the donor, resulting in rapid production of high antibody titer to the donor (anamnestic response). The high level of donor-specific antibody will fix complement, and results in deposition of antibody and complement in the vessels of the graft, causing graft dysfunction. Accelerated acute rejection should be suspected on the basis of clinical findings, including hemodynamic instability, worsening cardiac function, previous history of blood transfusion, or multiple pregnancies. The diagnosis is confirmed by a positive crossmatch between the donor’s leukocytes and the recipient’s serum (obtained after transplantation), by a rising level of antibody that is specific against donor HLAs, and by the presence of antibodies and complement in endomyocardial biopsy specimens from coronary arteries. Early recognition of this type of rejection is crucial because aggressive therapy with plasmapheresis, high-dose steroids, and antithymocyte globulin may save the graft (51,52).

Chapter 28: Complications of Cardiac Transplantation

INFECTIOUS COMPLICATIONS Bacterial Infections Infections accounted for most of the postoperative deaths that occurred in the era before cyclosporine (53). The infection rate is now lower than it was earlier, but infection still accounts for 22% of deaths during the first postoperative year (54). After cardiac transplantation, infection is most commonly caused, in decreasing order, by bacteria, viruses, fungi, and protozoans. Of 596 infections that occurred among heart transplant recipients at Stanford University between 1980 and 1989, 42% were bacterial (53). The most common site of infection is the lung; pneumonia occurs in 24% to 40% of patients. The other sites of infection are the bloodstream and the urinary tract. The introduction of cyclosporine to the immunosuppressive regimen for cardiac transplant recipients reduced bacterial and fungal infection; however, viral infection, especially cytomegalovirus (CMV), has emerged as an important cause of morbidity (55).

Viral Infections The viral agents that commonly cause infection in heart transplant recipients include CMV, herpes simplex virus, and Epstein–Barr virus (EBV). The peak incidence of CMV infection occurs approximately six weeks after transplantation (56). Although the lung is the most common site of CMV infection, the gastrointestinal tract, the liver, and the retina may also be involved. Viral cultures demonstrate CMV infection in 80% to 90% of patients (57). Although most CMV infections are asymptomatic, 15% to 40% of infected patients may develop flu-like symptoms, enteritis, or hepatitis. CMV infection is associated with an increased risk of rejection, and this infection accelerates the development of allograft coronary artery disease (58). Because CMV may be transmitted via the graft, both donor and recipient must be screened for the presence of CMV antibodies. CMV disease is usually mild in a recipient who was seropositive for CMV before transplantation (reactivated CMV infection). However, primary CMV infection, which occurs when a seronegative recipient receives an organ or blood products that are CMV-positive, is quite serious. Therefore, when a seropositive donor heart is implanted into a seronegative recipient, prophylaxis with intravenously administered ganciclovir is recommended (58–60). In addition, if a blood transfusion is required, CMV-negative blood should be used. The herpes simplex virus is shed postoperatively by half of all transplant recipients. Of these patients, half will experience clinical stomatitis, esophagitis, or genital lesions (54). Most cases of herpes simplex infection occur within the first two to three weeks after transplantation and are treated with orally administered acyclovir. Clinically significant herpetic infections are less common because of prophylactic antiviral regimens. Herpes zoster infection (shingles) occurs in approximately 10% of heart

361

transplant patients and usually follows an uncomplicated course; it should be treated with antiviral therapy to prevent dissemination. However, varicella (chicken pox) is serious and may be fatal; therefore, aggressive antiviral therapy is indicated immediately upon diagnosis. Immunization of seronegative candidates is recommended before transplantation. Infection with the EBV is common among transplant recipients. EBV infection is associated with the development of posttransplant lymphoproliferative disease (PTLD), and an EBV-seronegative patient receiving a seropositive graft is at substantial risk of PTLD (61).

Other Infections Pneumocystis carinii pneumonia occurs in 3% to 10% of all cardiac transplant recipients who are not given prophylactic treatment. This infection usually occurs between the 2 and 12 month after transplantation. The common findings are fever, dyspnea, nonproductive cough, and the classic pattern of diffuse interstitial and alveolar infiltrates on chest radiographs. Seventy-five percent of cases will respond to trimethoprim–sulfamethoxazole or pentamidine. Prophylaxis with oral trimethoprim–sulfamethoxazole has markedly decreased the incidence of this infection among transplant recipients (62,63). Toxoplasma gondii infection is of particular importance for cardiac transplant recipients because this organism often targets the myocardium. This organism is usually transmitted from the donor; therefore, serologic testing should be performed on both the donor and the recipient before the transplant procedure (64). The toxoplasma antibody titers of seronegative recipients who receive a seropositive heart should be carefully checked, and these patients should receive prophylaxis with pyrimethamine and folic acid (65). Acute infection is accompanied by serologic conversion or an increase in antibody titer; however, not all patients will exhibit symptoms. The isolation of tachyzoites from tissue or body fluids is required for diagnosis (65). Autopsy evidence of toxoplasmosis has been found in the heart, lungs, pericardium, and brain of transplant recipients with primary infections (66).

POSTTRANSPLANTATION MALIGNANCY Chronic immunosuppression is associated with a 100fold increase in the risk of malignancy (67). In the era before cyclosporine, Krikorian et al. (68) reported that the actuarial probability of developing any type of malignancy (solid tumors and lymphomas) was 2.7% one year after heart transplantation and 25.6% five years after transplantation (68). More recent findings show that malignancy accounts for 18.6% of deaths that occur four years after transplantation (69). After transplantation, the most commonly occurring solid tumor is skin cancer, and the most common malignancy other than solid tumor is PTLD (67,69).

362

Qi et al.

STRATEGIES FOR AVOIDING COMPLICATIONS &

&

&

&

&

&

&

Primary graft failure is usually a result of ischemia-reperfusion injury. This complication is more likely when donors are older than 45 years or have left-ventricular hypertrophy, when ischemic time exceeds five hours, or when the donor-to-recipient weight ratio is less than 0.7. The incidence of primary graft failure can be reduced by closely matching donors’ and recipients’ weights and by minimizing ischemic time. Injury to the right main pulmonary artery at its junction with the main pulmonary artery may occur when the ascending aorta is dissected from the main pulmonary artery, especially in repeated operations. This complication can be avoided by delaying the dissection until after cardiopulmonary bypass has been initiated and the native heart has been removed. Stroke is a rare but devastating complication of heart transplantation. In addition to the usual precautions taken to prevent embolism of air and particulate matter during the operation, the endocardium should be meticulously approximated in the left-atrial suture line to avoid thrombus formation and stroke. A rare cause of immediate graft failure is the technical complication of catching the circumflex artery or the coronary sinus in the left-atrial suture line. This complication is rare, but requires special vigilance to avoid it, especially when the remnant of the left donor atrium is small, as in the case of concomitant lung retrieval from the same donor. Accelerated acute rejection usually occurs 24 to 48 hours after transplantation and results from an anamnestic response to donor antigens. This complication is heralded by increasing hemodynamic instability in a previously sensitized recipient. Plasmapheresis, aggressive immunosuppression, and inotropic or mechanical support may save the patient and the graft. Acute renal failure is common in the early postoperative period after heart transplantation. This condition usually responds to hydration, high doses of diuretics, and temporary discontinuation of nephrotoxic drugs. Furosemide is much more effective when administered by continuous infusion (up to 20 mg/hr) than by intermittent bolus. Nonsteroidal anti-inflammatory agents should not be used for pain control because they exacerbate the renal toxicity of calcineurin inhibitors. Catastrophic intra-abdominal complications may occur after heart transplantation, but because of the masking effects of immunosuppression, patients may not exhibit the usual signs and symptoms. A high index of suspicion and expedient diagnostic and therapeutic procedures will improve the outcome.

Cumulative findings convincingly suggest that PTLD is related to EBV infection. In immunosuppressed transplant recipients, EBV-infected B-cells undergo malignant transformation into polyclonal or monoclonal EBV-positive lymphomas. The first line of treatment for these B-cell lymphomas is to reduce the level of immunosuppression (61,70). Chemotherapy, radiation therapy, and gammainterferon therapy have also been used with limited success. The mortality rate associated with PTLD is high, ranging from 23% to 35%. New treatment strategies that have been developed in recent years have been promising. For bone marrow recipients, adoptive transfer from the donor of cytotoxic, HLA-matched T-cells that are specific for EBV shows encouraging results (71,72). For solid-organ recipients, immunotherapy with autologous lymphokine-activated killer cells has been tried with some success (73). Preliminary studies of the use of humanized anti-CD20 monoclonal antibody against B-cells (rituximab) to treat PTLD have produced promising findings, with an overall response rate of 69% and a one-year survival rate of 73% (74).

Future strategies should focus not only on treating established disease but also on preventing infection by using more specific immunosuppression, vaccination against EBV, and induction of transplantation tolerance.

CARDIAC ALLOGRAFT VASCULOPATHY Even with the use of potent immunosuppressive agents, the recipient’s immune system continues to slowly reject the transplanted organ. This chronic rejection manifests itself as an interstitial fibrosis and as a progressive and diffuse intimal thickening of the graft arteries. For heart transplant recipients, the predominant feature of chronic rejection is the development of an accelerated form of coronary artery disease, which is often called cardiac allograft vasculopathy (CAV). CAV is responsible for more than 50% of late deaths among heart transplant recipients (75). The histologic features of allograft vasculopathy include a diffuse, concentric hyperplasia of the intimal layer of small- and medium-sized arteries; this

Chapter 28: Complications of Cardiac Transplantation

hyperplasia very rapidly progresses to complete luminal occlusion. It is generally accepted that the intimal hyperplasia is due to the migration and uncontrolled proliferation of vascular smooth muscle cells of the media layer in response to immune-mediated and nonimmune-mediated damage to the graft vessels (76). The cause of CAV is multifactorial and involves immune injury (cellular and humoral rejection), nonimmune injury (graft ischemic injury and CMV infection), and vascular factors (lipid accumulation, thrombosis, and growth factors). Diltiazem (Cardizem1, Biovail Corp., Ontario, Canada) and ‘‘statin’’ drugs reduce the incidence of CAV (77–79). However, because of the diffuse nature of the disease, once CAV is established, there is no effective treatment except retransplantation.

GASTROINTESTINAL COMPLICATIONS Gastrointestinal complications are common after heart transplantation and occur in approximately 20% of patients (80). The causes of these complications include surgical stress (gastritis, ulceration, and perforation), side effects of medication (steroidinduced gastritis, ulceration, pancreatitis, chemical hepatitis, cholelithiasis, and diverticulitis), and opportunistic infection and malignancy due to immunosuppression (candidiasis, CMV enteritis, and PTLD). Because of the masking effect of immunosuppressive agents, transplant recipients usually do not exhibit the typical systemic and local inflammatory signs and symptoms of gastrointestinal complications. Therefore, physicians must maintain a high index of suspicion for these problems (81,82). Diagnostic procedures, such as endoscopy with biopsy of suspicious lesions, computerized tomography, or exploratory laparotomy, must be initiated expediently to prevent catastrophe (80).

OTHER COMPLICATIONS OF IMMUNOSUPPRESSION A typical immunosuppressive regimen for heart transplant recipients consists of three agents: corticosteroids (prednisone), an antimetabolite (azathioprine or mycophenolate mofetil), and a calcineurin inhibitor (either cyclosporine or tacrolimus). Lympholytic agents such as rabbit antithymocyte globulin, horse antilymphocyte globulin, and murine monoclonal antibody against the human CD3 T-cell antigen (OK3) are also used as ‘‘induction therapy’’ in the perioperative period and as treatment for steroid-resistant rejection. In addition to their immunosuppressive properties, these agents cause various other toxic effects. Serious side effects of corticosteroids are cushingoid appearance, osteoporosis, diabetes mellitus, hyperlipidemia, peptic ulcer disease, cataracts, and capillary fragility. Fortunately, now that calcineurin inhibitors are available, the use of steroids has been markedly

363

reduced and a substantial number of heart transplant recipients can be weaned off steroids six months to one year after transplantation (83). Serious side effects of cyclosporine and tacrolimus include nephrotoxicity, hypertension, seizures, tremors, and peripheral neuropathy. Hirsutism and gingival hyperplasia are unique side effects of cyclosporine, whereas alopecia, diabetes mellitus, and hypomagnesemia are more commonly associated with tacrolimus (84). Gout is also a troublesome complication among heart transplant recipients and is associated with both cyclosporine and tacrolimus. Hyperuricemia is a result of decreased renal clearance, which is exaggerated by the use of diuretics to control hypertension (85). As treatment for acute episodes of gout, corticosteroids are the agent of choice because nonsteroidal anti-inflammatory agents frequently potentiate the nephrotoxicity of calcineurin inhibitors. Serious side effects of the antimetabolite drugs are leukopenia, anemia, thrombocytopenia, and gastrointestinal disturbances. Lympholytic agents are associated with cytokine release syndrome, thrombocytopenia, an increased risk of opportunistic infection, and PTLD (86–89).

CONCLUSION Heart transplantation has matured into an effective treatment for heart failure. Refinements in donor and recipient selection, surgical techniques, perioperative care, and immunosuppression during the past four decades have markedly reduced the rate of postoperative complications. To date, most complications associated with heart transplantation have been related to the use of immunosuppressive agents. Fortunately, during the past two decades, the emergence of new agents with different toxicity profiles has enabled transplant physicians to individualize the immunosuppressive regimen (90). As a result, complications associated with the use of immunosuppression have been manageable and the quality of life after heart transplantation has improved tremendously.

REFERENCES 1. Hosenpud JD, Bennett LE, Keck BM, Boucek MM, Novick RJ. The Registry of the International Society for Heart and Lung Transplantation: Eighteenth Official Report-2001. J Heart Lung Transplant 2001; 20(8): 805–815. 2. Kasper EK, Achuff SC. Clinical evaluation of potential heart transplant recipients. In: Baumgartner WA, Reitz BA, Kasper EK, Theodore J, eds. Heart HeartLung Transplantation. 2nd ed. Philadelphia: W.B. Saunders Company, 2002:57–63. 3. Blanche C, Blanche DA, Kearney B, et al. Heart transplantation in patients seventy years of age and older: a comparative analysis of outcome. J Thorac Cardiovasc Surg 2001; 121(3):532–541.

364

Qi et al.

4. Fonarow GC. How old is too old for heart transplantation? Curr Opin Cardiol 2000; 15(2):97–103. 5. Blanche C, Kamlot A, Blanche DA, et al. Combined heart-kidney transplantation with single-donor allografts. J Thorac Cardiovasc Surg 2001; 122(3):495–500. 6. Castillo-Lugo JA, Brinker KR. An overview of combined heart and kidney transplantation. Curr Opin Cardiol 1999; 14(2):121–125. 7. Tazbir JS, Cronin DC II. Indications, evaluations, and postoperative care of the combined liver-heart transplant recipient. AACN Clin Issues 1999; 10(2):240–252. 8. Surakomol S, Olson LJ, Rastogi A, et al. Combined orthotopic heart and liver transplantation for genetic hemochromatosis. J Heart Lung Transplant 1997; 16(5):573–575. 9. Detry O, Honore P, Meurisse M, et al. Advantages of inferior vena caval flow preservation in combined transplantation of the liver and heart. Transpl Int 1997; 10(2):150–151. 10. Shumway NE, Lower RR, Stofer RC. Transplantation of the heart. Adv Surg 1966; 2:265–284. 11. Traversi E, Pozzoli M, Grande A, et al. The bicaval anastomosis technique for orthotopic heart transplantation yields better atrial function than the standard technique: an echocardiographic automatic boundary detection study. J Heart Lung Transplant 1998; 17(11):1065–1074. 12. Beniaminovitz A, Savoia MT, Oz M, et al. Improved atrial function in bicaval versus standard orthotopic techniques in cardiac transplantation. Am J Cardiol 1997; 80(12):1631–1635. 13. Blanche C, Nessim S, Quartel A, et al. Heart transplantation with bicaval and pulmonary venous anastomoses. A hemodynamic analysis of the first 117 patients. J Cardiovasc Surg (Torino) 1997; 38(6):561–566. 14. Barnard CN, Barnard MS, Cooper DK, et al. The present status of heterotopic cardiac transplantation. J Thorac Cardiovasc Surg 1981; 81(3):433–439. 15. Nakatani T, Frazier OH, Lammermeier DE, Macris MP, Radovancevic B. Heterotopic heart transplantation: a reliable option for a select group of high-risk patients. J Heart Transplant 1989; 8(1):40–47. 16. Reichenspurner H, Hildebrandt A, Boehm D, et al. Heterotopic heart transplantation in 1988—recent selective indications and outcome. J Heart Transplant 1989; 8(5):381–386. 17. Ridley PD, Khaghani A, Musumeci F, et al. Heterotopic heart transplantation and recipient heart operation in ischemic heart disease. Ann Thorac Surg 1992; 54(2):333–337. 18. Kawaguchi A, Gandjbakhch I, Pavie A, et al. Cardiac transplant recipients with preoperative pulmonary hypertension. Evolution of pulmonary hemodynamics and surgical options. Circulation 1989; 80(5 Pt 2): III90–III96. 19. Baumgartner WA. Myocardial and pulmonary protection: long-distance transport. Prog Cardiovasc Dis 1990; 33(2):85–96. 20. Anyanwu AC, Rogers CA, Murday AJ. A simple approach to risk stratification in adult heart transplantation. Eur J Cardiothorac Surg 1999; 16(4):424–428. 21. Marelli D, Laks H, Fazio D, Moore S, Moriguchi J, Kobashigawa J. The use of donor hearts with left ventricular hypertrophy. J Heart Lung Transplant 2000; 19(5):496–503.

22. Weil R III, Clarke DR, Iwaki Y, et al. Hyperacute rejection of a transplanted human heart. Transplantation 1981; 32(1):71–72. 23. Cooper DK. Clinical survey of heart transplantation between ABO blood group-incompatible recipients and donors. J Heart Transplant 1990; 9(4):376–381. 24. Trento A, Hardesty RL, Griffith BP, Zerbe T, Kormos RL, Bahnson HT. Role of the antibody to vascular endothelial cells in hyperacute rejection in patients undergoing cardiac transplantation. J Thorac Cardiovasc Surg 1988; 95(1):37–41. 25. Loh E, Bergin JD, Couper GS, Mudge GH Jr. Role of panel-reactive antibody cross-reactivity in predicting survival after orthotopic heart transplantation. J Heart Lung Transplant 1994; 13(2):194–201. 26. Bojar RM, Warner KG. Manual of Perioperative Care in Cardiac Surgery. 3rd ed. Malden, Massachsetts: Blackwell Science, 1998. 27. Unsworth-White MJ, Herriot A, Valencia O, et al. Resternotomy for bleeding after cardiac operation: a marker for increased morbidity and mortality. Ann Thorac Surg 1995; 59(3):664–667. 28. Talamonti MS, LoCicero J III, Hoyne WP, Sanders JH, Michaelis LL. Early reexploration for excessive postoperative bleeding lowers wound complication rates in open-heart surgery. Am Surg 1987; 53(2): 102–104. 29. Tischler MD, Lee RT, Plappert T, Mudge GH, St. John Sutton M, Parker JD. Serial assessment of left ventricular function and mass after orthotopic heart transplantation: a 4-year longitudinal study. J Am Coll Cardiol 1992; 19(1):60–66. 30. Young JB, Winters WL Jr., Bourge R, Uretsky BF. 24th Bethesda conference: cardiac transplantation. Task Force 4: function of the heart transplant recipient. J Am Coll Cardiol 1993; 22(1):31–41. 31. Bertolet BD, Eagle DA, Conti JB, Mills RM, Belardinelli L. Bradycardia after heart transplantation: reversal with theophylline. J Am Coll Cardiol 1996; 28(2): 396–399. 32. Kieler-Jensen N, Lundin S, Ricksten SE. Vasodilator therapy after heart transplantation: effects of inhaled nitric oxide and intravenous prostacyclin, prostaglandin E1, and sodium nitroprusside. J Heart Lung Transplant 1995; 14(3):436–443. 33. Auler Junior JO, Carmona MJ, Bocchi EA, et al. Low doses of inhaled nitric oxide in heart transplant recipients. J Heart Lung Transplant 1996; 15(5):443–450. 34. Ardehali A, Hughes K, Sadeghi A, et al. Inhaled nitric oxide for pulmonary hypertension after heart transplantation. Transplantation 2001; 72(4):638–641. 35. Pope SE, Stinson EB, Daughters GT II, Schroeder JS, Ingels NB Jr, Alderman EL. Exercise response of the denervated heart in long-term cardiac transplant recipients. Am J Cardiol 1980; 46(2):213–218. 36. Stinson EB, Griepp RB, Bieber CP, Shumway NE. Hemodynamic observations after orthotopic transplantation of the canine heart. J Thorac Cardiovasc Surg 1972; 63(3):344–352. 37. Jacquet L, Ziady G, Stein K, et al. Cardiac rhythm disturbances early after orthotopic heart transplantation: prevalence and clinical importance of the observed abnormalities. J Am Coll Cardiol 1990; 16(4): 832–837.

Chapter 28: Complications of Cardiac Transplantation

38. Ellenbogen KA, Szentpetery S, Katz MR. Reversibility of prolonged chronotropic dysfunction with theophylline following orthotopic cardiac transplantation. Am Heart J 1988; 116(1 Pt 1):202–206. 39. el Gamel A, Yonan NA, Grant S, et al. Orthotopic cardiac transplantation: a comparison of standard and bicaval techniques. J Thorac Cardiovasc Surg 1995; 109(4):721–730. 40. Trento A, Takkenberg JM, Czer LS, et al. Clinical experience with one hundred consecutive patients undergoing orthotopic heart transplantation with bicaval and pulmonary venous anastomoses. J Thorac Cardiovasc Surg 1996; 112(6):1496–1503. 41. Herre JM, Barnhart GR, Llano A. Cardiac pacemakers in the transplanted heart: short term with the biatrial anastomosis and unnecessary with the bicaval anastomosis. Curr Opin Cardiol 200; 15(2):115–120. 42. Scott CD, Dark JH, McComb JM. Sinus node function after cardiac transplantation. J Am Coll Cardiol 1994; 24(5):1334–1341. 43. Berke DK, Graham AF, Schroeder JS, Harrison DC. Arrhythmias in the denervated transplanted human heart. Circulation 1973; 48(suppl 1):III112–III115. 44. Uretsky BF. Physiology of the transplanted heart. Cardiovasc Clin 1990; 20(2):23–56. 45. Yusuf S, Theodoropoulos S, Dhalla N, et al. Influence of beta blockade on exercise capacity and heart rate response after human orthotopic and heterotopic cardiac transplantation. Am J Cardiol 1989; 64(10): 636–641. 46. Ellenbogen KA, Thames MD, DiMarco JP, Sheehan H, Lerman BB. Electrophysiological effects of adenosine in the transplanted human heart. Evidence of supersensitivity. Circulation 1990; 81(3):821–828. 47. Hunt SA. New immunosuppressive agents in clinical use: mycophenolate mofetil and tacrolimus. Cardiol Rev 2000; 8(3):180–184. 48. Taylor DO, Barr ML, Meiser BM, Pham SM, Mentzer RM, Gass AL. Suggested guidelines for the use of tacrolimus in cardiac transplant recipients. J Heart Lung Transplant 2001; 20(7):734–738. 49. Billingham ME, Cary NR, Hammond ME, et al. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection. Heart Rejection Study Group. The International Society for Heart Transplantation. J Heart Transplant 1990; 9(6):587–593. 50. Billingham ME. Endomyocardial biopsy diagnosis of acute rejection in cardiac allografts. Prog Cardiovasc Dis 1990; 33(1):11–18. 51. Madan AK, Slakey DP, Becker A, et al. Treatment of antibody-mediated accelerated rejection using plasmapheresis. J Clin Apheresis 2000; 15(3):180–183. 52. Woodle ES, Newell KA, Haas M, et al. Reversal of accelerated renal allograft rejection with FK506. Clin Transplant 1997; 11(4):251–254. 53. Pennock JL, Oyer PE, Reitz BA, et al. Cardiac transplantation in perspective for the future. Survival, complications, rehabilitation, and cost. J Thorac Cardiovasc Surg 1982; 83(2):168–177. 54. Dummer JS. Infectious complications of transplantation. Cardiovasc Clin 1990; 20(2):163–178. 55. Hofflin JM, Potasman I, Baldwin JC, Oyer PE, Stinson EB, Remington JS. Infectious complications in heart

56.

57.

58. 59.

60.

61. 62. 63. 64.

65.

66. 67. 68. 69.

70.

71.

72.

365

transplant recipients receiving cyclosporine and corticosteroids. Ann Intern Med 1987; 106(2):209–216. Kirklin JK, Naftel DC, Levine TB, et al. Cytomegalovirus after heart transplantation. Risk factors for infection and death: a multiinstitutional study. The Cardiac Transplant Research Database Group. J Heart Lung Transplant 1994; 13(3):394–404. Dummer JS, White LT, Ho M, Griffith BP, Hardesty RL, Bahnson HT. Morbidity of cytomegalovirus infection in recipients of heart or heart-lung transplants who received cyclosporine. J Infect Dis 1985; 152(6):1182–1191. Valantine HA. Role of CMV in transplant coronary artery disease and survival after heart transplantation. Transpl Infect Dis 1999; 1(suppl 1):25–30. Keay S, Petersen E, Icenogle T, et al. Ganciclovir treatment of serious cytomegalovirus infection in heart and heart-lung transplant recipients. Rev Infect Dis 1988; 10(suppl 3):S563–S572. Balfour HH Jr., Chace BA, Stapleton JT, Simmons RL, Fryd DS. A randomized, placebo-controlled trial of oral acyclovir for the prevention of cytomegalovirus disease in recipients of renal allografts. N Engl J Med 1989; 320(21):1381–1387. Nalesnik MA, Makowka L, Starzl TE. The diagnosis and treatment of posttransplant lymphoproliferative disorders. Curr Probl Surg 1988; 25:367–472. Hughes WT, Kuhn S, Chaudhary S, et al. Successful chemoprophylaxis for Pneumocystis carinii pneumonitis. N Engl J Med 1977; 297(26):1419–1426. Dummer JS. Pneumocystis carinii infections in transplant recipients. Semin Respir Infect 1990; 5(1):50–57. Hakim M, Wreghitt TG, English TA, Stovin PG, CoryPearce R, Wallwork J. Significance of donor transmitted disease in cardiac transplantation. J Heart Transplant 1985; 4(3):302–306. Luft BJ, Naot Y, Araujo FG, Stinson EB, Remington JS. Primary and reactivated toxoplasma infection in patients with cardiac transplants. Clinical spectrum and problems in diagnosis in a defined population. Ann Intern Med 1983; 99(1):27–31. Wreghitt TG, Hakim M, Gray JJ, et al. Toxoplasmosis in heart and heart and lung transplant recipients. J Clin Pathol 1989; 42(2):194–199. Penn I. Tumors after renal and cardiac transplantation. Hematol Oncol Clin North Am 1993; 7(2):431–445. Krikorian JG, Anderson JL, Bieber CP, Penn I, Stinson EB. Malignant neoplasms following cardiac transplantation. JAMA 1978; 240(7):639–643. Hosenpud JD, Bennett LE, Keck BM, Fiol B, Boucek MM, Novick RJ. The Registry of the International Society for Heart and Lung Transplantation: Sixteenth Official Report-1999. J Heart Lung Transplant 1999; 18(7): 611–626. Starzl TE, Nalesnik MA, Porter KA, et al. Reversibility of lymphomas and lymphoproliferative lesions developing under cyclosporin-steroid therapy. Lancet 1984; 1(8377):583–587. O’Reilly RJ, Small TN, Papadopoulos E, Lucas K, Lacerda J, Koulova L. Adoptive immunotherapy for Epstein–Barr virus-associated lymphoproliferative disorders complicating marrow allografts. Springer Semin Immunopathol 1998; 20(3–4):455–491. Papadopoulos EB, Ladanyi M, Emanuel D, et al. Infusions of donor leukocytes to treat Epstein–Barr virus-associated lymphoproliferative disorders after

366

73.

74.

75. 76. 77.

78.

79. 80. 81.

Qi et al.

allogeneic bone marrow transplantation. N Engl J Med 1994; 330(17):1185–1191. Nalesnik MA, Rao AS, Furukawa H, et al. Autologous lymphokine-activated killer cell therapy of Epstein– Barr virus-positive and -negative lymphoproliferative disorders arising in organ transplant recipients. Transplantation 1997; 63(9):1200–1205. Milpied N, Vasseur B, Parquet N, et al. Humanized anti-CD20 monoclonal antibody (Rituximab) in post transplant B-lymphoproliferative disorder: a retrospective analysis on 32 patients. Ann Oncol 2000; 11(suppl 1):113–116. Haverich A. Cyclosporine in heart transplantation. Transplant Proc 1992; 24(4 suppl 2):82–84. Hosenpud JD. Immune mechanisms of cardiac allograft vasculopathy: an update. Transpl Immunol 1993; 1(4): 237–249. Schroeder JS, Gao SZ, Alderman EL, et al. A preliminary study of diltiazem in the prevention of coronary artery disease in heart-transplant recipients. N Eng J Med 1993; 328(3):164–170. Wenke K, Meiser B, Thiery J, et al. Simvastatin reduces graft vessel disease and mortality after heart transplantation: a four-year randomized trial. Circulation 1997; 96(5):1398–1402. Kobashigawa JA, Katznelson S, Laks H, et al. Effect of pravastatin on outcomes after cardiac transplantation. N Eng J Med 1995; 333(10):621–627. Mueller XM, Tevaearai HT, Stumpe F, et al. Extramediastinal surgical problems in heart transplant recipients. J Am Coll Surg 1999; 189(4):380–388. Steed DL, Brown B, Reilly JJ, et al. General surgical complications in heart and heart-lung transplantation. Surgery 1985; 98(4):739–745.

82. Johnson R, Peitzman AB, Webster MW, et al. Upper gastrointestinal endoscopy after cardiac transplantation. Surgery 1988; 103(3):300–304. 83. Pham SM, Kormos RL, Hattler BG, et al. A prospective trial of tacrolimus (FK506) in clinical heart transplantation: intermediate term results. J Thorac Cardiovasc Surg 1996; 111(4):764–772. 84. Asante-Korang A, Boyle GJ, Webber SA, Miller SA, Fricker FJ. Experience of FK506 immune suppression in pediatric heart transplantation: a study of long-term adverse effects. J Heart Lung Transplant 1996; 15(4):415–422. 85. Lin HY, Rocher LL, McQuillan MA, Schmaltz S, Palella TD, Fox IH. Cyclosporine-induced hyperuricemia and gout. N Engl J Med 1989; 321(5):287–292. 86. Burk ML, Matuszewski KA. Muromonab-CD3 and antithymocyte globulin in renal transplantation. Ann Pharmacother 1997; 31(11):1370–1377. 87. Taylor DO, Kfoury AG, Pisani B, Hammond EH, Renlund DG. Antilymphocyte-antibody prophylaxis: review of the adult experience in heart transplantation. Transplant Proc 1997; 29(8A):13S–15S. 88. Zuckermann AO, Grimm M, Czerny M, et al. Improved long-term results with thymoglobuline induction therapy after cardiac transplantation: a comparison of two different rabbit-antithymocyte globulines. Transplantation 2000; 69(9):1890–1898. 89. Guttmann RD, Caudrelier P, Alberici G, Touraine JL. Pharmacokinetics, foreign protein immune response, cytokine release, and lymphocyte subsets in patients receiving thymoglobuline and immunosuppression. Transplant Proc 1997; 29(7A):24S–26S. 90. Gummert JF, Ikonen T, Morris RE. Newer immunosuppressive drugs: a review. J Am Soc Nephrol 1999; 10(6): 1366–1380.

29 Complications of Mechanical Circulatory Support Fotios M. Andreopoulos Departments of Surgery and Biomedical Engineering, University of Miami School of Medicine, Miami, Florida, U.S.A. Richard J. Kaplon Cardiac and Thoracic Surgery Medical Group, Sacramento, California, U.S.A.

During the past 10 years, various modes of mechanical circulatory support have been used clinically to successfully restore homeostasis, improve end-organ function, and serve as a bridge to transplantation or recovery. The true longterm goal, however, once the technology has adequately evolved, is to use cardiac assist devices as destination therapy. Currently available ventricular assist devices (VADs) include extracorporeal membrane oxygenation (ECMO), extracorporeal pulsatile assist devices, implantable assist devices, and the total artificial heart (TAH). Some of these devices provide either univentricular or biventricular support (Thoratec1, ABIOMED1); others provide only left ventricular (LV) assistance [HeartMate1 (Thoratec Corporation, Berkley, California, U.S.A.) and Novacor1 (WorldHeart Inc., Oakland, California, U.S.A.)]. ECMO and the TAH [CardioWestTM (SynCardia Systems Inc., Tucson, Arizona, U.S.A.)] provide total circulatory support. Current experimental devices include axial flow impeller pumps (Jarvik 20001, HeartMate II1, NASA/ DeBakey VADTM) and a fully implantable electric TAH [AbioCor1 (ABIOMED, Inc., Danvers, Massachusetts, U.S.A.)]. This chapter will briefly describe the functional characteristics of ECMO and the various VADs; it will also explain the complications associated with each.

DEVICES Extracorporeal Membrane Oxygenation ECMO is used primarily as a therapeutic option for patients suffering from severe acute respiratory distress syndrome or acute cardiogenic shock. This method was introduced in the early 1970s by Hill et al. and has been employed worldwide since the late 1980s (1). ECMO can provide cardiac and pulmonary support for short periods (days to weeks), thereby allowing time for the native heart or lungs to recover. Its basic configuration consists of a venous drainage cannula, a centrifugal pump, an oxygenator, and either a venous or an arterial return cannula. The venous–arterial (V–A)

configuration is used primarily for cardiac or cardiorespiratory support, whereas the venous–venous (V–V) mode is used solely for respiratory support (2). ECMO insertion is typically achieved by peripheral cannulation, either percutaneously or by means of an open cut-down technique; however, central cannulation from the right atrium to the aorta has also been used. Drainage to the pump is typically accomplished from the femoral vein, with return either to the femoral artery (V–A ECMO) or to the right atrium via the femoral vein (V–V ECMO) (2). Alternatively, for V–V ECMO, cannulation may be achieved through the femoral vein or the internal jugular vein (2). In this mode, femoral drainage and atrial reinfusion via the internal jugular vein is preferred to atrial drainage and femoral reinfusion because it provides greater extracorporeal flow and higher pulmonary arterial mixed venous oxygen saturation (3). The main benefits of ECMO include ease of peripheral cannula insertion and effective univentricular or biventricular cardiopulmonary support.

ABIOMED BVS 50001 The ABIOMED BVS 50001 (ABIOMED Cardiovascular Inc., Danvers, Massachusetts, U.S.A.) is an extracorporeal, pneumatically actuated VAD that can provide short-term left, right, or biventricular support. The device is a dual-chamber pump within a polycarbonate housing. Each chamber contains a seamless polyurethane bladder with a volume of approximately 100 cc. Blood drains by gravity from the native atria of the heart into the upper (atrial) chamber of the pump, which acts as a reservoir for the lower (ventricular) pumping chamber. Two trileaflet polyurethane [AngioflexTM (MicroMed Tech Inc., Houston, Texas, U.S.A.)] valves proximal and distal to the ventricular chamber ensure unidirectional flow. Once the lower chamber has filled with blood, compressed air enters the blood chamber, causing bladder collapse and return of blood to the patient (pump systole). A bedside console that provides self-regulating, pulsatile

368

Andreopoulos and Kaplon

KEY POINTS &

&

&

&

&

Bleeding is the most common complication seen with extracorporeal membrane oxygenation (ECMO) and ventricular assist device (VAD) implantation. The use of aprotinin at the time of device placement or the administration of vitamin K can help reduce early bleeding. Postoperative bleeding necessitates blood transfusion, but alloimmunization and elevated panel-reactive antibody levels can complicate organ matching at the time of transplantation. The incidence of thromboembolism and stroke is directly related to the type of device used and to the patient’s underlying medical condition. Most of the mechanical circulatory assist devices currently used require careful anticoagulation protocols. Infection is the leading cause of late mortality for patients treated with assist devices. Causes of infection include the preoperative clinical status of the patient, surgical trauma during device insertion, and exiting cannulae and drivelines. Prevention is the foundation of treatment for infection. Suppressive antibiotic therapy, pump-pocket debridement, device exchange, and explantation are common therapeutic techniques; transplantation is the treatment of choice. Right heart failure among patients with left ventricular assist devices (LVADs) is an important cause of perioperative morbidity and may lead to death. Right ventricular dysfunction can be avoided by optimizing patients’ clinical condition preoperatively and by administering phosphodiesterase inhibitors and nitric oxide during LVAD implantation. Mechanical complications are relatively common in association with ECMO support; these complications necessitate regular exchange of the oxygenator. In contrast, LVADs in general are mechanically resilient; the most common complications are controller malfunctions, inflow valve deterioration, and driveline or outflow kinking.

support controls the pump. The console operates asynchronously to the native heart by adjusting pump rate and duration of systole and diastole to compensate for changes in preload and afterload (4). A number of cannulation options are available for pump insertion (5). Left-sided inflow cannulation sites include the left atrium by means of the interatrial groove, the dome or the left atrial appendage, or the LV apex. Although left dome cannulation may be used for small hearts, the interatrial groove is most frequently used for inflow. LV apical cannulation is associated with higher flows and a lower incidence of LV thrombus formation; this is the site of choice for patients with a prosthetic mitral valve. The right atrial free wall is the site most commonly used for right atrial cannulation, although direct right ventricular (RV) cannulation has also been used successfully. Arterial return is achieved with an end-to-side anastomosis of the outflow grafts to either the ascending aorta or the pulmonary artery. The ABIOMED VAD is approved by the U.S. Food and Drug Administration (FDA) for postcardiotomy support and for short-term bridge-to-recovery therapy. Its use has been expanded to include all forms of reversible heart failure, including myocardial infarction, cardiac trauma, and RV support with an implantable LV assist device (LVAD) as bridge-to-recovery or bridge-totransplantation therapy.

Thoratec1 Ventricular Assist Devices The Thoratec VAD (Thoratec Corporation, Pleasanton, California, U.S.A.) is also an extracorporeal, pneuma-

tically actuated pump that can provide left, right, or biventricular support. This VAD’s rigid polysulfone casing lies on the patient’s abdominal wall and contains a diaphragm that separates the air chamber from the polyurethane (Thoralon1) blood sac. An external console provides alternating positive and negative pressure that assists with filling and emptying of the blood sac. MonostrutTM mechanical disc valves in the inflow and outflow conduits of the pump ensure unidirectional flow. The Thoratec VAD can be operated in several modes: fixed-rate (asynchronous) mode, synchronous (timed by the patient’s electrocardiogram) mode, or volume mode. The volume mode is recommended for maintaining physiologic blood flow, although it is asynchronous to the heart. Paracorporeal pump placement and right-heart and left-heart cannulation options are similar to those of the ABIOMED BVS 5000. Once cannulation is complete, air is removed from the pump through the arterial graft. The Thoratec VAD is activated at a slow fixed rate (40 bpm, 20% systolic ejection) with a drive pressure of 100 to 120 mmHg and a vacuum pressure of 0 to 5 mmHg. After the device is checked for leaks, the ejection pressure is gradually increased to 200 mmHg and the vacuum is set between 10 and 20 mmHg. Once the VAD is appropriately filled and emptied, the pump is placed in volume mode. When the chest is closed, full vacuum pressure ( 25 to 40 mmHg) is applied, and the ejection pressure is set at 100 mmHg above the systolic blood pressure (6–8). The paracorporeal position of the device allows for easy inspection of the blood chamber and for easy

Chapter 29: Complications of Mechanical Circulatory Support

pump replacement if necessary. The versatility of the Thoratec VAD is due to its multiple cannulation configurations, its biventricular potential, its applicability even for small patients (body surface area of less than 1.5 m2), and its capability for short-term or long-term support. Additionally, the ability to apply suction to the venous drainage enhances blood return to the pump and helps to maintain adequate flows. The FDA has approved the Thoratec VAD for bridgeto-transplant and bridge-to-recovery purposes. More widespread use of this device is limited because of its large pneumatic driver, which makes outpatient management difficult, and because of the need for aggressive anticoagulation to minimize thromboembolic complications from the mechanical valves and thrombogenic surfaces of the pump.

HeartMate1 The HeartMate (Thoratec Corporation, Berkley, California, U.S.A.) is an implantable, pulsatile electric VAD that provides LV support. It is fabricated from sintered titanium and houses a flexible, textured polyurethane diaphragm. The pump is typically implanted in the left upper quadrant and is connected to a controller and a power source by a percutaneous driveline. The blood pump is textured to enable circulating cells to adhere to the pump and form a biological lining that reduces blood–device interaction. This ‘‘pseudoneointima’’ is responsible for the low thromboembolic risk associated with this device, even without anticoagulation therapy (9). Unidirectional flow is ensured by porcine valves within the inflow and outflow conduits. The vented electric HeartMate device received FDA approval as bridge-to-transplantation therapy in 1998. This device allows patients to be discharged from the hospital and to resume their everyday activities while awaiting heart transplantation. The pump can operate in two different modes: fixed and automatic. The fixed mode allows the pump to operate at set rates ranging from 50 to 120 bpm, whereas the auto mode allows the pump rate to vary according to the physiological filling of the pump. Once the pump chamber has been 90% filled with blood, ejection is initiated. The auto mode is the most common mode of operation and allows the pump to respond to circulatory demand. The HeartMate may be implanted intraabdominally or in a preperitoneal pocket. Most surgeons experienced with HeartMate implantation favor the preperitoneal location because excessive heat and fluid loss from the abdominal cavity can be avoided, internal organ erosion and bowel obstruction are eliminated, intra-abdominal adhesions are minimized, and infections can be easily managed (10,11). A pocket is created between the peritoneum and the rectus muscle sheath. The pump’s driveline exits from the right lower quadrant through a small incision made between the costal margin and the superior iliac crest. Special effort is made to create a subcutaneous tunnel long enough to permit optimal tissue ingrowth

369

around the driveline and to minimize the risk of infection (11). Inflow to the pump is through the LV apex; return is through a Dacron1 outflow conduit anastomosed end-to-side to the ascending aorta. The inflow conduit is directed away from the interventricular septum to prevent inflow complications associated with sucking the ventricular septal muscle into the cannula. Once the inflow conduit and the arterial outflow graft have been connected, air is removed from the pump with a hand pump to minimize the likelihood of air embolism (11). The presence of air can be monitored by transesophageal echocardiography (TEE). Initially, the HeartMate is activated at a fixed rate of 50 bpm; the rate is then slowly increased as the patient is weaned from cardiopulmonary bypass. Once the patient has been completely weaned from bypass, the device can be fully activated. The results of Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH), a prospective, multicenter trial, have recently been reported (12). This study, designed to assess the effectiveness of the HeartMate VAD in comparison to the effectiveness of medical therapy for patients with end-stage heart disease, and for whom transplantation is contraindicated, found that the use of implantable VADs enhances survival and improves quality of life. Compared with patients receiving only pharmacologic therapy, patients receiving the HeartMate experienced a 48% reduction in risk of death (12).

Novacor1 Left Ventricular Assist System Like the HeartMate, the Novacor LV assist system (LVAS) (World Heart Corporation, Ottawa, Canada) is an implantable, electromechanical device designed for long-term LV support. The integrated blood pump/energy converter is implanted into a preperitoneal pocket in the left upper quadrant (13). During systole, a polyurethane pump sac is compressed by two symmetrical pusher plates that are coupled to a solenoid energy converter; this compression causes blood ejection. Again, unidirectional flow is ensured by bioprosthetic valves attached to the inflow and outflow conduits of the pump chamber. A percutaneous driveline exits the patient’s lower right quadrant and is connected to an external control unit. The Novacor device fills passively; upon mechanical actuation of the pusher plate, blood is ejected. As with the Thoratec device, there are three modes of operation: fixed, synchronous, and fill-to-empty. In the fixed mode, the pump rate is set and the device operates asynchronously to the native heart. In the synchronous mode, pump ejection is triggered by the R-wave of the native heart; even though the pump operates synchronously with the native heart, the use of this mode is limited by the dependence of the pump on an external signal. Fill-to-empty is the most common mode of operation, providing sufficient output by responding to variable physiological demands.

370

Andreopoulos and Kaplon

Implantation of the Novacor device is accomplished by a procedure similar to that used to implant the HeartMate (13). Briefly, the device itself is assembled by connecting the inflow and outflow valved conduits to the inflow and outflow grafts, respectively. The inflow conduit is tunneled through the diaphragm into the left ventricle through a ventriculotomy at the LV apex. Once the conduits and grafts have been positioned, air is removed from the pump through a needle hole in the outflow graft. The pump is started at a low fixed rate, and the patient is slowly weaned from cardiopulmonary bypass. Once weaning has been completed, the pump can be switched to the fill-to-empty mode. TEE is used to ensure adequate air removal, to assess RV function, and to diagnose aortic insufficiency or to determine the patency of the foramen ovale. Like the HeartMate, the Novacor system because of its wearable configuration, allows the patient to be discharged from the hospital and to resume life outside the hospital. Initially designed for bridge-to-transplant therapy, the Novacor VAD is now less frequently used because of reports of excessive development of thromboembolism. Recent changes to the inflow conduit appear to have alleviated that problem, and a new trial, Investigation of Non–Transplant-Eligible Patients Who Are Inotrope Dependent, is under way to investigate the use of this device as destination therapy for patients who are not candidates for transplantation.

CardioWestTM Total Artificial Heart The CardioWest TAH, previously known as the Jarvik 71 or the Symbion1 TAH, is a pneumatically driven, biventricular pulsatile pump that provides full circulatory support. It consists of two rigid polyurethane ventricles that are connected to the native atria and great vessels. Each ventricle contains a smooth, flexible polyurethane diaphragm that separates the blood from the air chambers. Compressed air, delivered through a percutaneous driveline that connects each ventricle to an external console, pressurizes the blood chamber, and causes blood ejection. Mechanical valves provide unidirectional flow. The pump has a maximum stroke volume of 70 mL and a maximum pump output of 15 L/min; however, average flow rates range from 6 to 8 L/min. Pump rate, percent systole, and driveline pressures are controlled for each ventricle; the device is operated so that the blood sacs do not fill completely but always empty completely. To ensure full ejection, the drive pressure of the console is set 30 to 40 mmHg higher than the pressure of the great vessels (14). The CardioWest TAH is implanted in the orthotopic position (15). Because this device does not fit in all patients, sizing guidelines must be followed. Criteria for optimal fit include a body surface area of at least 1.7 m2, an appropriate cardiothoracic ratio, and adequate LV diastolic dimensions, anteroposterior

distance, and combined ventricular volumes. In smaller patients, the device may not fit well, and left pulmonary vein or inferior vena cava compression may result. Device insertion requires that each ventricle be excised at the atrioventricular groove and that each great vessel be excised at the sinotubular junction. Ventricular tissue is trimmed away to create atrial cuffs, and the TAH ‘‘atrial quick connectors’’ are sewn in place. The great vessels are anastomosed to the outflow conduits, and the ventricles are connected to the atrial quick connectors. After the air has been removed from the ventricles under TEE guidance, the patient is weaned from cardiopulmonary bypass, and pump support is initiated at a slow rate. Once the chest has been closed, vacuum pressure ( 10 to 15 cmH2O) is applied to the device during diastole to assist with ventricular filling, and the pump’s mode of operation is switched to the more physiological full-eject mode. Patients who receive the CardioWest TAH must be given anticoagulation therapy to prevent thrombus formation. Patients may ambulate, but their movement is greatly restricted by the large drive console. A smaller portable drive console is being developed for use outside the hospital.

COMPLICATIONS Complications leading to significant clinical morbidity and mortality are common in association with ECMO, VAD, and TAH support. Despite advances in technology and in our understanding of the physiology of assist devices, the most problematic complications, as expected, are bleeding, thromboembolism, and infection. For patients with left-sided heart support, right heart failure is a unique problem. For all devices, mechanical failure is an underlying concern.

Bleeding With an incidence as high as 60%, bleeding is the most common complication of ECMO (16–18). Clotting factor deficiency and disseminated intravascular coagulation resulting from hepatic dysfunction are common among patients requiring this type of support. McManus et al. (19) reported that 68% of patients had clotting factor deficiencies before the start of ECMO. Furthermore, the blood–device interface causes mechanical trauma to circulating blood, with resultant activation of platelets, leukocytes, and the clotting and fibrinolytic cascades. Subsequent platelet loss and consumption of clotting factors can lead to hemorrhage, oxygenator obstruction, and system malfunction. The administration of blood products (platelets, fresh frozen plasma, or cryoprecipitate) is often necessary to counter this ECMO-associated blood dyscrasia. Unfortunately, to avoid circuit clotting, anticoagulation is necessary, but this therapy further exacerbates problematic bleeding. Typically, heparin is the anticoagulant of choice; activated

Chapter 29: Complications of Mechanical Circulatory Support

clotting times (ACT) of 200 to 220 seconds are recommended. For patients with heparin-induced thrombocytopenia, recombinant hirudin has been used successfully (20). However, for patients with specific contraindications to anticoagulation, heparin-coated circuits (Carmeda1, Medtronic Inc., Minneapolis, Minnesota, U.S.A.) may be used rather than systemic anticoagulation (21,22). As expected, these heparinbonded circuits have less effect on platelet consumption than uncoated circuits. Promising results have been obtained with selective anticoagulation given directly into the extracorporeal circuit and with new anticoagulants that have a short half-life (23). As with ECMO, bleeding is the most frequent complication associated with the implantation of VADs or TAHs. The risk of bleeding ranges from 20% to 80% depending upon the device being implanted, the type of support required (univentricular vs. biventricular), and the patient’s preexisting conditions (4,24–26). Patients who require VADs, like those who require ECMO, often have underlying clinical or subclinical hepatic dysfunction; in addition, many are already receiving anticoagulants such as heparin or warfarin for ECMO, have been treated with implanted intraaortic balloon pumps, or have coronary artery disease or underlying poor LV function. According to the Registry of the International Society of Heart and Lung Transplantation, bleeding substantially influences the survival of patients who are supported by mechanical assist devices (27). The incidence of bleeding was lower among bridge-to-transplant patients who were discharged after transplantation than among those patients who died with a VAD in place (28). Whereas late bleeding is typically associated with leaking at the connection sites and erosion of the device into anatomic structures, early bleeding results from patients’ underlying comorbid conditions, cardiopulmonary bypass–induced thrombocytopenia and platelet dysfunction, perioperative hypothermia, and the complexity of the operative technique. Early bleeding is often exacerbated by the need for anticoagulation when certain devices are used (e.g., ABIOMED, Thoratec, Novacor, CardioWest). One of the advantages of the HeartMate is its textured lining that avoids the need for early anticoagulation. The use of aprotinin, a serine protease inhibitor, has reduced the incidence of early bleeding. Although many surgeons administer aprotinin when the device is placed, some prefer to reserve the use of this agent for the time of device explantation or heart transplantation (29). The incidence of bleeding is higher for patients who are supported with biventricular devices than for those with LVADs; however, this higher incidence is probably due, at least in part, to the patients’ clinical condition before implantation. Most patients require only left-sided heart support; patients who also require right-sided heart support are usually sicker and more malnourished and have experienced more episodes of hepatic congestion and concomitant

371

coagulopathy than patients requiring LVADs. Even for patients who require only LVADs, subclinical hepatic dysfunction is associated with a higher incidence of hemorrhage. Maneuvers as simple as administering vitamin K, even when prothrombin times are normal, can often help to reduce bleeding (30). Although postoperative hemorrhage is usually not life threatening, it typically necessitates blood transfusion. Platelet use has been associated with alloimmunization and elevated panel-reactive antibody levels among patients receiving VADs; this therapy can complicate organ matching at the time of transplantation and create a need for preoperative crossmatch testing. Massad et al. (31) demonstrated that leukocytes in blood products transfused at the time of device implant increased LVAD recipients’ sensitization to human leukocyte antigens. Moazami et al. (32) showed that this sensitization resulted predominantly from platelets. In the Cleveland Clinic experience, LVAD recipients responded to 19% of T-cell panel-reactive antibodies, whereas non-LVAD transplant candidates responded to only 5% (33). In addition to the immunologic consequences of transfusion, postoperative transfusion requirements can also worsen preexistent pulmonary hypertension. For patients whose transpulmonary gradients are already elevated, this exacerbation of pulmonary hypertension can mean the difference between LVAD or both left and right ventricular assist devices (BiVAD) support and may change a relatively short hospitalization period with discharge on an LVAD to months of in-hospital care on a BiVAD. Furthermore, patients who require multiple units of blood transfusions when the VAD is placed are at increased risk of prolonged ventilatory support and long-term infection. Murphy et al. (34) demonstrated that infection rates are higher (22%) among patients undergoing cardiac surgery who receive six or more units of blood than among patients who receive no more than two units. Given that the predominant long-term cause of death for patients on LVAD support is infection, factors that increase the risk of infection, such as bleeding, must be carefully addressed (12). ECMO is associated with a substantial risk of intracranial bleeding; in one study, 19% of patients suffered outcome-determining intracranial bleeding while on ECMO support (35). The risk of intracranial hemorrhage is higher among female patients, those taking heparin, those with an elevated serum creatinine concentration ( >2.6 mg/mL), and those with thrombocytopenia. The aggressive management of anticoagulation, prevention of renal failure, and correction of thrombocytopenia may reduce the risk of intracranial hemorrhage among adults supported by ECMO.

Thromboembolism and Stroke The incidence of device-related thromboembolism ranges from 2% to 50%, depending on the type of device

372

Andreopoulos and Kaplon

used and on the patient’s underlying medical condition (4,9,25,26,33). Turbulent flow patterns, platelet damage, and large synthetic blood-contacting surfaces, combined with low blood flow, may predispose patients to the formation of clots and thrombi (36). Although the HeartMate requires no specific anticoagulation and requires only aspirin as antiplatelet therapy, careful anticoagulation protocols must be followed so as to prevent thromboembolic complications with ECMO or the ABIOMED, Thoratec, Novacor, or CardioWest devices. Patients supported by ECMO are typically treated with heparin for anticoagulation. Once postoperative bleeding has been minimized, heparin is administered to keep ACTs in the range of 200 to 220 seconds. Like patients supported by ECMO, patients supported by the ABIOMED BVS 5000 are treated early with heparin; once their condition has stabilized, antiplatelet therapy in the form of either aspirin or clopidogrel is added. Because the Thoratec and Novacor devices are usually intended for longterm support, after patients with these devices receive early anticoagulation therapy with heparin, they are maintained on a combination of warfarin and aspirin or clopidogrel. To avoid early heparin-related bleeding problems, yet still maintain a low level of anticoagulation, some surgeons use an infusion of dextran to affect rheology and avoid early thromboembolic complications (37). Once the patients are in stable condition, they receive heparin therapy with a target partial thromboplastin time of 60 to 80 seconds, or warfarin with a target international normalized ratio of 2.5 to 3.5. In addition, some patients are treated with dipyridamole. Whereas the ABIOMED, Thoratec, and Novacor devices use smooth, seamless polyurethane bloodcontacting surfaces, the HeartMate has textured interior surfaces. Sintered titanium microspheres are used on the rigid metallic components, and textured polyurethane is used on the flexible pusher-plate diaphragm. The textured surfaces entrap blood elements and form a stable and densely adherent biological lining. This pseudoneointima forms a long-term blood-contacting interface and because the fibrin-based neointima densely adheres to the rough interior surfaces of the pump, it substantially reduces the potential for embolization (9). Patients supported by the HeartMate typically receive only aspirin (325 mg/day). Despite careful anticoagulation management, the incidence of clinical thromboembolism among patients supported by VADs is substantial. In a report of their experience with the ABIOMED BVS 5000, Samuels et al. (4) noted that of 7 (16%) of 43 patients who underwent device implantation died of thromboembolic events to the central nervous system (4). In a report of their cumulative experience with 111 patients treated with the Thoratec VAD, McBride et al. (25) found thrombus in 14% of devices at the time of explantation, for a complication rate of 19% (25). Similarly, El-Banayosy et al. (26) noted a 15%

incidence of thromboembolism among their 144 patients supported by the Thoratec VAD (26). Reports indicate that the incidence of thromboembolic complications in association with the Novacor LVAS is as high as 48%; the incidence of stroke ranges from 21% to 48% (26,33,38–40). Recent changes in the inflow conduit reduced the stroke rate from 21% to 12% in one series and from 48% to 13% in another (33,40). Whereas the incidence of thromboembolic strokes is 8% among patients supported by a CardioWest TAH despite treatment with anticoagulants, the incidence of thromboembolism among patients supported by the HeartMate is typically 2% to 6% despite a lack of anticoagulation (24,41,42). In studies comparing outcomes associated with the Thoratec VAD, the Novacor LVAS, and the CardioWest TAH, Copeland et al. found that the incidence of stroke was 34% among patients supported with the Novacor ventricular assist system, 12% among those supported by the Thoratec VAD, and 8% among those supported by the TAH (41). In a similar comparison, Minami et al. (43) found that the incidence of stroke was 22% in the Thoratec group, 39% in the Novacor group, and 16% in the HeartMate group. Notably, this is the highest reported incidence of stroke associated with the HeartMate device; other reports typically document a stroke rate of 2% to 6% (9,24,42). Finally, Kasirajan et al. (33), in their experience with 205 patients, found that the incidence of thromboembolism was 43% among patients supported by the Novacor and 12% among patients supported by the electric HeartMate. In that series, the incidence of early (no more than seven days after device implantation) stroke was the same for the two devices; however, the rate of later ischemic strokes was markedly different (38% for the Novacor and 2% for the HeartMate). As noted above, these researchers found that the rate of neurologic thromboembolism decreased substantially among patients supported by the Novacor LVAS after revision of the inflow conduit (from 48% to 13%).

Infection Whereas bleeding and thromboembolism account for much of the morbidity associated with VADs, infection leading to sepsis and multisystem organ failure is the primary cause of late mortality for patients treated with these assist devices (12). Because of the large surface area of synthetic material in ECMO circuits, patients supported by these devices are subject to infections; however, the incidence of sepsis among these patients is lower than among patients supported by other devices. This difference in the incidence of sepsis is most likely due to the short duration of ECMO support (typically 90% for both) in diagnosing symptomatic, acute, proximal DVT (3). The usefulness of compression US is limited, however, when patients are obese, when their legs are edematous, or when they suffer from chronic DVT. In addition, US is not reliable in diagnosing thrombi in the iliac veins; in such cases, MR imaging or venography should be used (4). The ultrasound probe can combine gray-scale, duplex, and color Doppler imaging, although a meta-analysis of data from multicenter trials demonstrated that real-time B-mode US, duplex US, and color flow Doppler US are not sensitive enough to diagnose asymptomatic acute proximal DVT (3). This study also emphasized the importance of the competency of the technician (3). Another study, published in the orthopedic literature, used duplex US to screen patients at the time of discharge from the hospital who had undergone an orthopedic procedure; the incidence of DVT was only 1% within 90 days of the procedure, a finding suggesting once again that the sensitivity of duplex US may depend on the skill of the technician (5). Because it is difficult to distinguish recurrent DVT from chronic DVT, follow-up US performed at three months and six months after an episode of DVT should be used as a baseline for the detection of recurrent DVT. Most cases of DVT of the upper extremity are caused by indwelling venous catheters. The sensitivity and specificity of US in detecting upper-extremity DVT are similar to those of US in detecting asymptomatic and symptomatic DVT of the lower extremity. PE occurs in as many as 36% of patients with upperextremity DVT (6).

clinical conditions as pericarditis, cardiac failure, hypotension, and arterial insufficiency are absent.

Venography Venography with the use of a contrast agent allows direct visualization of the venous system of an extremity. This procedure is the most accurate diagnostic method for DVT and is the standard against which other methods are compared in prospective, randomized studies. The most reliable criterion for the diagnosis of acute DVT is the appearance of a constant intraluminal-filling defect on two or more venographic views (7). False-positive results can be obtained if poor technique causes improper filling or external compression of veins. One disadvantage of venography is its invasiveness, which may result in contrast-induced phlebitis that leads to thrombosis, allergic reactions, or contrastinduced nephropathy—a complication that occurs most frequently among patients with diabetes. Venography tends to be painful and can be technically difficult when patients are obese or when venous access is poor. In addition, venography is more time consuming and costlier than noninvasive tests. For these reasons, venography is generally not considered the standard of care.

Radioactively Labeled Fibrinogen Radioactively labeled fibrinogen has been used to diagnose DVT since the advent of portable scintillation counters. When the count in the extremity is 20% higher than that in the heart, the diagnosis of underlying thrombosis is made. Radioactively labeled fibrinogen studies are sensitive in diagnosing DVT of the calf vein but have severe limitations in other applications. Several well-designed orthopedic studies have demonstrated that the use of this method is limited by its low sensitivity (8,9). It cannot detect thrombi in pelvic veins and it cannot be used to study extremities with healing wounds, ulcers, or any other inflammatory process.

D-Dimer Assays Impedance Plethysmography IPG measures the volume response of the extremity to temporary occlusion of the venous system. The diagnosis of DVT depends on the rate of venous emptying of the extremity upon release of the tourniquet. If recording of the outflow wave is prolonged after the inflation cuff or tourniquet is released, the diagnosis of severe venous thrombosis can be made with 95% accuracy. The primary disadvantage of IPG is that it cannot diagnose DVT in the calf vein, nor can it differentiate a new thrombus from old post-thrombotic changes. The role of IPG as a therapeutic decision maker has decreased, but the procedure is still valuable in physiologic research studies. The sensitivity of IPG in evaluating asymptomatic lower extremities is low, but its accuracy in diagnosing symptomatic DVT is excellent as long as such

D-dimer assays measure the specific degradation products of fibrin when it undergoes endogenous fibrinolysis. Early trials demonstrated that the various types of D-dimer assays differed in their sensitivity and specificity (10). The enzyme-linked immunosorbent assay is most sensitive, but this assay is costly and cannot be performed rapidly. The latex agglutination tests have the lowest specificity. Recent studies have shown that D-dimer assays, which can be performed at the bedside, have a high negative predictive value. However, false-positive results are relatively common, and this problem fuels the controversy about the use of D-dimer assays as a first-line screening test. Rapid bedside assays are increasingly available, but additional outcome studies are necessary (11,12). Recently, one study recommended the use of D-dimer assays as the least expensive tests for

Chapter 30: Venous Thromboembolism

initial screening and suggested coupling this test with US if the results of the assay were positive (13).

Magnetic Resonance Imaging The most recent development in the diagnosis of DVT is the use of MR imaging, which can demonstrate occlusive and nonocclusive intraluminal thrombi. Early reports suggested that MR imaging has a sensitivity and specificity of 90% to 100% in detecting symptomatic DVT (1). The value of using MR imaging to screen patients with asymptomatic disease awaits further study. MR imaging has few promising advantages except for its impressive sensitivity and specificity. It is probably the best test for diagnosing DVT in pelvic veins, especially in the noniliac veins, which can be assessed by no other diagnostic method, including venography. The contour of the vein walls and the appearance of the perivenous tissue may be used as markers to distinguish acute DVT from chronic conditions. Another appealing feature of MR imaging is that it is noninvasive and carries no risk of renal impairment. In addition, MR imaging is unique in that it can diagnose other conditions of the extremity that cause DVT-like symptoms. Pathological conditions such as cellulitis, edema, and joint inflammation are readily demonstrated by MR imaging. The primary disadvantages of MR imaging are its cost and the possibility that it may not be available in remote rural institutions. Because MR imaging is usually not an option for patients with indwelling metallic devices, its use may be precluded for patients in the early postoperative period or those who have been treated with metal instrumentation for fractures.

Summary At present, no findings support routine screening for DVT when patients exhibit no symptoms. However, each diagnostic center should develop its own algorithm for screening patients with symptoms that suggest a diagnosis of DVT; such algorithms should be based on the institution’s available human and technologic resources.

Prophylaxis Ongoing studies and advances in chemotherapy and technology have resulted in frequent changes in the prophylaxis against DVT among postoperative and trauma patients. Many studies have shown the effectiveness of prophylaxis in preventing thromboembolism. The current effort is directed at finding the optimal method of preventing DVT, with a focus on effectiveness and safety.

Low-Dose Heparin The time-honored drug heparin is probably still the most commonly used method of prophylaxis against DVT.

381

Heparin’s mechanism of action is based on the potentiation of antithrombin, which, in turn, inhibits a cascade of procoagulation factors, especially IIa and Xa. Heparin has been shown to decrease the risk of DVT by 60% to 70%, although its effectiveness varies according to the specific procedure undergone by the patient (14). The main adverse effects of heparin therapy are bleeding (usually hematoma in the operative wound) and the rare but dangerous occurrence of heparin-induced thrombocytopenia.

Low-Molecular-Weight Heparin Low-molecular-weight heparin (LMWH) is a fractionated heparin with fewer pentasaccharide chains than other forms of heparin. Several products with slightly different chemical and kinetic properties are commercially available. Because of its reduced molecular weight, LMWH has less effect on factor IIa than do other heparins, although its inhibitory effect on factor Xa is still substantial. Theoretically, this property should increase the effectiveness of LMWH and give it a safer therapeutic profile and better pharmacodynamic properties than unfractionated heparin (UFH). After general surgery, prophylaxis with LMWH offered the same protection against DVT as prophylaxis with heparin (15). LMWH is 2 to 10 times more expensive than heparin, but its pharmacodynamic properties are better; this advantage may reduce the difference in cost between the two treatments. LMWH has been shown to be more effective and safer than UFH for patients undergoing orthopedic procedures and for those who have suffered trauma; thus, it is currently the best treatment choice for these patients (16). The risk of heparin-induced thrombocytopenia is lower with LMWH than with UHF, but LMWH should not be considered an alternative to UFH for patients with this syndrome. This problem will be solved in the future by new investigational chemotherapy agents with heparinoid compounds.

Compression Devices The fear of postoperative bleeding prompted the implementation of lower-extremity compression devices ranging from simple elastic compression stockings to sophisticated pneumatic compressive devices (PCDs). All of these devices are used in an attempt to minimize the effect of immobilization and blood stasis. After general surgery, the prophylactic effectiveness of PCDs is similar to that of low-dose heparin (14). A recent study has demonstrated no significant difference in DVT between compression devices and LMWH in trauma patients (17). The theory that sequential compression of the extremity may activate the fibrinolysis cascade has not been shown to have clinical significance, and the use of PCDs on the upper extremities of patients with trauma to the lower extremities has not proved to be beneficial.

382

Klein and Ginzburg

Summary

Complications

For trauma patients, the best prophylaxis against DVT is the use of compression devices. LMWH should be used when casts or hardware on the lower extremities preclude the use of bilateral leg compression devices. If the patient is receiving suboptimal mechanical or chemical anticoagulation, duplex US screening for DVT should be considered (14). When both anticoagulation and mechanical compression are contraindicated, as they are, for example, among patients with spinal cord transection or severe pelvic fractures, PE prophylaxis with an inferior vena cava filter (IVCF) should be considered.

The most common complication of DVT is the postphlebitic syndrome (PPS), which will affect as many as 30% of patients with DVT and will cause substantial disability in 7% of patients (20). PPS has a range of symptoms characterized by leg swelling, edema, and skin discoloration leading to ulcers. The syndrome is believed to be the result of venous hypertension caused by outflow obstruction and valvular incompetence. The risk of PPS is highest among patients with recurrent ipsilateral DVT (21). Conservative treatment with graduated compression stockings, weight reduction, and continous nursing care has been shown to reduce the symptoms of 30% of patients, but for approximately two-thirds of patients, the symptoms did not improve or even worsened (22). Another rare but feared complication of severe venous outflow obstruction is phlegmasia cerulea dolens. This devastating condition results from increased interstitial pressure that compromises tissue perfusion and leads to progressive ischemia. The mortality rate associated with this condition is estimated to be 30% to 40%, with tissue loss appearing in 50% of patients. Aggressive treatment with thrombolysis, fasciotomy, and thrombectomy is necessary once anticoagulation therapy fails (23). The recurrence of DVT despite adequate treatment depends on the risk factors of each patient. In a study (24) of recurrent DVT, the risk of recurrence approached 30% after eight years. The risk of recurrent DVT was highest among patients with malignancy and lowest among those who had suffered trauma (24).

Treatment The mainstay of DVT treatment is full anticoagulation with intravenously administered adjusted-dose UFH or subcutaneously administered LMWH. Numerous studies and meta-analyses have compared the two methods of treatment. Although most studies have shown no difference in the effectiveness of the two methods of prophylaxis, a meta-analysis of 11 randomized studies found that severe bleeding is less common with LMWH than with UFH (18). Heparin should be the initial treatment for DVT; a bolus of 80 U/kg should be followed by maintenance infusion of 18 U/kg/hr until the patient’s partial thromboplastin time is 46 to 70 seconds. Dosages of subcutaneously administered LMWH differ according to the preparations of the individual drugs. Long-term anticoagulation with oral warfarin [adjusted to achieve an international normalized ratio (INR) of 2–3] should be started on the first day after heparin therapy is initiated. After five days, the initial treatment with intravenously administered heparin or LMWH should be discontinued if the INR is higher than 2. LMWH has been used successfully as longterm therapy. Long-term anticoagulation should be continued for 6 to 12 weeks in cases of symptomatic calf-vein thrombosis, for three to six months after the first incidence of proximal DVT, and for life in cases of recurrent venous thromboembolism among patients with chronic risk factors for thromboembolism (19). Although more aggressive treatment with thrombectomy or thrombolytic therapy has been used, it is usually reserved for cases with complicated massive unresolved ileofemoral thrombosis.

Promising New Strategies for Thromboprophylaxis Recently, new prophylactic medications have been developed. These drugs, known as pentasaccharides, are more effective than LMWH for patients who have undergone hip replacement. The disadvantages of the pentasaccharides are that they cost more than heparin, they are associated with increased bleeding among patients undergoing knee replacement, and their anticoagulant properties cannot be reversed. Several other orally administered prophylactic medications are being investigated.

PULMONARY EMBOLISM Overview One of the most devastating experiences for any surgeon is to perform a successful operation or to repair a complex injury, only to have the patient succumb to a massive unexpected PE. The methods of diagnosing PE are constantly evolving and the management of this complication remains controversial.

Incidence PE most commonly results from thrombi in the popliteal veins and the veins proximal to them in the lower extremity, although half of patients with PE have no detectable DVT. The risk factors for PE are similar to those for DVT because the two conditions occur at different stages of the same disease process. For patients undergoing surgery, those risk factors include increased age, prolonged immobility, malignancy, and trauma. Any thrombophilic condition will also increase the risk of PE and DVT. Rarely, patients undergoing surgery may experience PE in association with the presence of a central venous catheter (25).

Chapter 30: Venous Thromboembolism

383

Autopsy studies performed before prophylactic measures came into use detected PE in 4% to 16% of patients who had died of traumatic injury, and PE was considered to be the direct cause of death in approximately half of these patients (26,27). Currently, PE occurs in approximately 2% of patients undergoing general surgery without prophylaxis. The use of prophylaxis reduces this overall incidence by more than 50%: with prophylaxis, PE occurs in fewer than 1% of patients undergoing general surgery and in only 0.3% to 2% of patients who have suffered trauma (14). The risk of PE for patients undergoing general surgery is related to each patient’s risk factors and to the nature of the procedure. The risk of PE for patients who have suffered trauma is most closely related to the severity of the injury and to the presence of specific injuries involving fractures of the spine, head, pelvis, and long bones (28). For trauma patients, 6% of PEs will occur within the first day after injury and 25% will occur during the first week after injury (29), probably because prophylaxis cannot be administered to these patients early in their treatment course. Thus, PE is the third most common cause of death among trauma patients.

Arterial Blood Gas Analysis

Diagnosis Physical Examination

The results of plain chest radiography [chest X ray (CXR)] are abnormal but nonspecific for most patients with PE, except for the rare occasion when PE causes lung infarction with its suggestive wedge-shaped consolidation. Other radiographic signs specific for PE include the Westermark sign (decreased vascular marking). CXR is a valuable diagnostic tool because it can detect other pathological conditions that may explain the clinical picture, such as pneumothorax or pneumonia.

The fact that PE is often unsuspected underscores the low sensitivity of the signs and symptoms of this disease. In fact, PE is one of the most common unsuspected causes of death and is often detected only during autopsy. PE should be suspected whenever sudden, unexplained dyspnea occurs. All of the other classic signs, such as pleuritic pain and hemoptysis, are nonspecific for diagnosis. Other signs and symptoms of PE include low-grade fever, pleural rub, and accentuated second heart sound. Whether hemodynamic-cardiac signs will be evident depends on the extent of the embolus and the degree of acute right heart dysfunction that results. The hemodynamic-cardiac signs can be mild such as tachycardia and lightheadedness but massive obstruction can cause hypotension, cyanosis, neck vein distension, and sudden cardiac arrest (30).

Arterial blood gas analysis is probably the first test ordered by most physicians when PE is suspected because dyspnea and tachycardia are the most common symptoms presented. Hypoxemia is common but is not always present, especially in young patients with no underlying pulmonary pathology. Ten percent of patients with acute PE will have no demonstrable hypoxemia [Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) study)] (32). For approximately one-third of patients with PE who are younger than 40, the partial pressure of oxygen in the plasma phase of arterial blood (PaO2) will be higher than 80 mmHg (33). For 86% of patients with PE, the alveolar–arterial (a–A) pressure gradient is elevated; thus, the a–A value is considered to be a more sensitive diagnostic indicator than the PaO2 value, although it can also be normal in patients with PE who do not have underlying lung pathology. In general, hypoxemia or an elevated a–A gradient can indicate the possibility of PE, but normal values cannot rule out the diagnosis (34).

Plain Chest Radiography

D-Dimer Assays The advantages and disadvantages of D-dimer assays are the same for patients with suspected PE and for patients with suspected DVT. The negative predictive value of a plasma D-dimer value of less than 500 ng/mL is 95%. Unfortunately, D-dimer values are low in 25% of patients who do not have venous thromboembolism. Currently, no published results support the use of D-dimer assays to rule out PE (1).

Electrocardiography The importance of electrocardiography (ECG) in diagnosing PE has declined with the development of hightechnology imaging. ECG changes indicative of PE include axis deviation and nonspecific ST-wave and T-wave segment abnormalities in the lateral leads. The diagnostic pattern of the S1Q3T3 sign, right bundle branch block, P pulmonale, and right axis deviation is present in only one-third of patients with massive PE who exhibit clinical signs of acute cor pulmonale. Other nonspecific ECG changes will occur in approximately 87% of patients with proven PE and no underlying cardiac disease (31).

Ventilation/Perfusion Scans For many years, the ventilation/perfusion (V/Q) scan has been the cornerstone of the diagnostic work-up for suspected PE. The effort expended in finding other diagnostic methods proves the poor performance of the V/Q scan. The study uses two scans: one after intravenous injection of a radioisotope to evaluate lung perfusion and the other after inhalation of the radioisotope for evaluation of ventilation. The results of the scan are read as normal or as demonstrating low, intermediate, or high probability of PE. The main shortcoming of the test is that most types of coexisting

384

Klein and Ginzburg

lung disease affect the interpretation of the ventilation scan and, to some extent, that of the perfusion scan. When the results of CXR are normal, approximately 10% of V/Q scans will be interpreted as demonstrating intermediate probability of PE, but this number can be as high as 60% to 70% when CXR demonstrates conditions such as chronic lung disease or pneumonia. Thus, the diagnostic interpretation of V/Q scans is difficult. Most of the findings associated with the V/Q scan are from the PIOPED. This multicenter study compared the effectiveness of the V/Q scan for diagnosing PE with that of pulmonary angiography and autopsy studies (32). PIOPED demonstrated that better diagnostic accuracy is obtained when the level of clinical suspicion is combined with the results of the V/Q scan. The study showed that 96% of patients for whom the clinical suspicion of PE was high and whose V/Q results indicated a high probability of PE did in fact have a PE. Unfortunately, 40% of the patients for whom the clinical suspicion of PE was high but whose V/Q results indicated a low probability of PE also had a PE. When the clinical suspicion of PE was low, only 56% of patients whose V/Q results indicated a high probability of PE did indeed have a PE. Because CXR often yields abnormal results, especially when patients are being treated with mechanical ventilation, some physicians advocate using only the perfusion scan rather than both parts of the V/Q scan. One study showed that the diagnostic accuracy of the perfusion scan alone was similar to that of the full V/Q scan (35). The treatment of patients whose V/Q scanning results are inconclusive is controversial. One option is to perform serial noninvasive studies of the lower extremities in an attempt to detect DVT. When the results of V/Q scanning indicate a low probability of PE and the results of DVT studies are negative, the incidence of PE is 25% if the level of clinical suspicion is high (36). The treatment of these patients should be tailored to the clinical situation. All patients whose V/Q scanning results indicate an intermediate or high probability of PE should be treated with anticoagulation measures.

with the highest rate of complications. The risk of serious complications after this study is estimated at approximately 4% (38). Most complications are related to the contrast agent and the cateterization procedures used, but some critically ill patients with pulmonary hypertension and cor pulmonale may experience severe hemodynamic compromise and even cardiac arrest (39).

Helical Computed Tomography Technological advances in CT scanning during recent years have increased the use of this diagnostic method. During the past decade, the use of helical computed tomography (HCT) as part of the diagnostic work-up for PE has gained increasing popularity. HCT is appealing because it is less invasive than pulmonary angiography and because it can also demonstrate other pathological conditions of the thorax. The main disadvantage of HCT is that it only poorly shows peripheral areas and horizontally oriented vessels. HCT undoubtedly misses some peripheral PEs, but the clinical relevance of these PEs is uncertain. The sensitivity of HCT in detecting PEs in the main, lobar, and segmental (until the fourth order) pulmonary arteries is estimated at more than 90% (40). Not all studies support the reliability of HCT as a first-line method of diagnosing PE. At least two studies found that HCT was less sensitive than pulmonary angiography in diagnosing PE; disagreement among radiologists about the interpretation of HCT findings, including centrally located PEs, contributed to this lower sensitivity (41,42). In most centers, however, HCT has been adopted as a diagnostic method.

Magnetic Resonance Imaging The newest method for diagnosing PE is MR imaging. As is true of HCT, MR imaging has not been evaluated in large, controlled studies; thus, we have only limited definitive information about its performance. Only some small pilot studies have shown that the sensitivity and specificity of MR imaging in detecting PE are good (43). Currently, we do not have enough information to support the routine use of this costly method for diagnosing PE.

Pulmonary Angiography Pulmonary angiography is the gold standard for the diagnosis of PE. This study should be performed if other less-invasive tests fail to confirm or rule out a diagnosis of PE. The procedure begins with catheterization of a central vein; the catheter is advanced through the right heart to the pulmonary artery. The lobar or segmental arteries should be injected selectively. The presence of an intraluminal filling defect on two angiographic views is considered diagnostic for PE. Secondary diagnostic criteria are reduced flow, tortuous peripheral vessels, and delayed venous phase (37). Because pulmonary angiography is the most invasive method of diagnosing PE, it is associated

Echocardiography Although transthoracic or transesophageal echocardiography has occasionally demonstrated an embolus in the main pulmonary artery, the most common use of these studies is to evaluate right heart dysfunction. Right heart failure is the final cause of death for patients who suffer a massive PE. Although global or regional right ventricle dyskinesia is evident in more than 80% of patients with PE, it is nonspecific for diagnosis and can be associated with other clinically similar conditions such as chronic obstructive pulmonary disease. The role of echocardiography in diagnosing PE remains to be defined.

Chapter 30: Venous Thromboembolism

Summary Although we have no definitive information about the value of HCT and MR imaging in the diagnosis of PE, these methods are widely used in place of V/Q scans. Prospective studies are currently underway to investigate the accuracy of these modalities.

Prophylaxis and Treatment Prophylaxis Against Pulmonary Embolism Prophylaxis is the first line of defense against PE. Administering low-dose heparin reduces the risk of fatal PE from 0.8% to 0.2% (14). Unfortunately, prophylaxis is not perfect: approximately 80% of trauma patients who experience PE received adequate prophylaxis (28).

Treatment for Established Deep Venous Thrombosis or Pulmonary Embolism Full anticoagulation is the treatment of choice for DVT and PE. This therapy will reduce the risk of proximal thrombus propagation in the deep vein system and in the pulmonary arteries, and will also reduce the risk of recurrent PE.

Inferior Vena Cava Filter The concept of interrupting the flow through the IVC so as to prevent the passage of emboli from the venous system of the lower extremity was first introduced in 1960. The rationale behind using vena cava interruption is that some form of treatment is needed for the substantial number of patients for whom anticoagulation is not an option or who have experienced complications related to anticoagulation. The first procedure involved ligation of the IVC; however, the high morbidity and mortality rates associated with this procedure prevented it from gaining popularity. Use of the IVCF began in 1967, but this procedure was associated with a nearly 70% risk of IVC thrombosis and occlusion. A breakthrough in this techique occurred in 1973 with the development of the Geenfield–Cimray filter; use of this filter lowered the rate of long-term vena cava occlusion to less than 5%. In the early 1980s, the next technical advance came in the form of percutanous insertion of the IVCF (44). Currently, the two most commonly used filters are the titanium Greenfield filter and the bird’s nest filter that has a larger diameter. Insertion of the IVCF usually takes place in the invasive radiology suite. The first step in the procedure is venographic measurement of the vena cava; the appropriate filter is then placed under the right renal vein. Recently, it has been demonstrated that the filter can be inserted at the bedside with ultrasonographic guidance (45). Experiments indicate that retrievable or temporary filters may offer short-term

385

protection and reduce the incidence of long-term complications (46). The complications associated with the use of IVCFs can be early or late. The most common early complication is thrombosis of the superficial femoral vein; this complication is related to the size of the introducer. The recent development of introducers with smaller diameters may reduce the incidence of this complication. Other early complications include malpositioning of the filters; this complication occurs in 7% of placement procedures. Late complications include dislodgement of the filter, as a result of erroneous placement in the IVC, and tilting of the filter. IVC thrombosis occurs in 2% to 20% of cases, but recanalization will occur within four years in almost all cases. The most common late complication is DVT. About half of patients with permanent IVCFs will experience DVT, and most of them will exhibit clinical symptoms of venous insufficiency (47). Recurrent PE after insertion of an IVCF has been reported to occur in 3% of patients, mainly those with chronic hypercoagulable conditions such as cancer (48). There is no universal agreement about some indications for IVCF insertion. Most physicians agree that an IVCF should be placed in patients who have survived a massive PE but whose cardiopulmonary reserve is so limited that another embolic event would be devastating. The same holds true for patients who have undergone pulmonary embolectomy. The primary controversy is related to the use of an IVCF as a prophylactic measure for high-risk patients who have not experienced a documented thromboembolic event. Currently, no studies support the routine use of IVCFs for these patients as prophylaxis against PE.

Thrombolytic Therapy As described above, the routine treatment for PE is anticoagulation with intravenously administered heparin or subcutaneously administered LMWH. Such treatment will reduce the risk of recurrent emboli and the propogation of the existing thrombi. In most cases, recanalization of the obstructed pulmonary artery will take place. In extreme situations, when massive PE induces significant right heart failure and hemodynamic instability, the increased right heart afterload must be resolved quickly. In these rare situations, thrombolytic therapy with streptokinase, urokinase, or tissue plasminogen activator may be lifesaving. The most serious risk, naturally, is bleeding, but this risk should be weighed against the high mortality rates experienced by patients with massive pulmonary emboli who are in an unstable condition. Although there are reports of the administration of thrombolytic agents to such patients by systemic or local infusion, we do not have sufficient information to determine the success rate and the risks associated with this treatment. However, the use of thrombolytic therapy should be considered when patients are in this desperate situation (19).

386

Klein and Ginzburg

Pulmonary Embolectomy Another treatment option for patients with massive PE whose condition is rapidly deteriorating is surgical pulmonary embolectomy. This procedure is a valid option especially when thrombolysis has failed or is contraindicated. The mortality rate associated with this procedure has been reported to be as high as 75%. Pulmonary embolectomy has not been performed often enough to allow us to evaluate its effectiveness (19). Most authorities recommend the placement of an IVCF after embolectomy. A few recent reports have described attempts to mechanically fragment and extricate massive PEs by using a percutaneous rotating catheter. Promising results have been achieved with or without followup treatment with thrombolytic therapy, and the complication rate is very low (49).

2.

3.

4.

5.

6.

Summary If possible, every patient who suffers a PE should receive anticoagulation therapy with adjusted-dose intravenously administered heparin or subcutaneously administered LMWH. If anticoagulation therapy is contraindicated, fails, or produces complications, an IVCF should be placed. An IVCF should also be considered for patients with massive PE and low cardiopulmonary reserve or for those who have undergone pulmonary embolectomy. The use of prophylactic IVCFs is sometimes recommended for high-risk patients (e.g., those with spine or head injuries) for whom anticoagulation is contraindicated. When patients are in a hemodynamically unstable condition with acute cor pulmonale resulting from massive PE, decisions about whether to use thrombolytic therapy or surgical (or percutaneous, in the future) pulmonary embolectomy must be made for each case individually.

Outcome PE is associated with a high mortality rate: 10% to 40% of high-risk patients will die (50). Of those patients who survive the initial event, 1.5% will die of recurrent PE within a year (51). Occasionally, chronic pulmonary hypertension occurs as a result of recurrent nonfatal PE. This condition is readily diagnosed with V/Q scans. Surgical thrombectomy may be necessary so as to avoid the risk of right heart failure and deteriorating lung function (14).

The Future Currently, researchers are investigating new pentasaccharides and are studying the use of temporary vena cava occlusion devices. The findings are promising thus far, but further investigation is necessary.

7.

8.

9.

10. 11.

12.

13. 14. 15.

16.

17.

REFERENCES 18. 1. Tapson VF, Carroll BA, Davidson BL, et al. The diagnostic approach to acute venous thromboembolism.

Clinical practice guideline. American Thoracic Society. Am J Respir Crit Care Med 1999; 160(3):1043–1066. Greenfield LJ. Complications of venous thrombosis and pulmonary embolism. In: Greenfield LJ, ed. Complications in Surgery and Trauma. 2nd ed. Philadelphia: Lippincott, 1984:406–421. Tapson VF, Carroll BA, Davidson BL, et al. The diagnostic approach to acute venous thromboembolism. Clinical practice guideline. American Thoracic Society. Am J Respir Crit Care Med 1999; 160(3):43, 68, 90. Tapson VF, Carroll BA, Davidson BL, et al. The diagnostic approach to acute venous thromboembolism. Clinical practice guideline. American Thoracic Society. Am J Respir Crit Care Med 1999; 160(3):54. Tapson VF, Carroll BA, Davidson BL, et al. The diagnostic approach to acute venous thromboembolism. Clinical practice guideline. American Thoracic Society. Am J Respir Crit Care Med 1999; 160(3):103. Tapson VF, Carroll BA, Davidson BL, et al. The diagnostic approach to acute venous thromboembolism. Clinical practice guideline. American Thoracic Society. Am J Respir Crit Care Med 1999; 160(3):114. Tapson VF, Carroll BA, Davidson BL, et al. The diagnostic approach to acute venous thromboembolism. Clinical practice guideline. American Thoracic Society. Am J Respir Crit Care Med 1999; 160(3):8. Lensing AW, Hirsh J. 125I-fibrinogen leg scanning: reassessment of its role for the diagnosis of venous thrombosis in postoperative patients. Thromb Haemost 1993; 69:2–7. Cruickshank MK, Levine MN, Hirsh J, et al. An evaluation of impedance plethysmography and 125Ifibrinogen leg scanning in patients following hip surgery. Thromb Haemost 1989; 62:830–834. Elias A, Aptel I, Huc B, et al. D-dimer test and diagnosis of deep vein thrombosis: a comparative study of 7 assays. Thromb Haemost 1996; 76:518–522. Roussi J, Bentolila S, Boudaoud L, et al. Contribution of D-dimer determination in the exclusion of deep venous thrombosis in spinal cord injury patients. Spinal Cord 1999; 37:548–552. Tapson VF, Carroll BA, Davidson BL, et al. The diagnostic approach to acute venous thromboembolism. Clinical practice guideline. American Thoracic Society. Am J Respir Crit Care Med 1999; 160(3):138. Perone N, Bounameaux H, Perrier A. Comparison of four strategies for diagnosing deep vein thrombosis: a cost-effectiveness analysis. Am J Med 2001; 110:33–40. Geerts WH, Heit JA, Clagett GP, et al. Prevention of venous thromboembolism. Chest 2001; 119:132S–175S. Nurmohamed MT, Rosendaal FR, Buller HR, et al. Low-molecular-weight heparin versus standard heparin in general and orthopaedic surgery: a metaanalysis. Lancet 1992; 340:152–156. Hirsh J, Warkentin TE, Shaughnessy SG, et al. Heparin and low-molecular-weight heparin mechanisms of action, pharmacokinetics, dosing, monitoring, efficacy, and safety. Chest 2001; 119:64S–94S. Ginzburg E, Cohn SM, Lopez J, Jackowski J, Brown M, Hameed SM. Randomized clinical trial of intermittent pneumatic compression and low molecular weight heparin in trauma. Br J Surg 2003; 90:1338–1344. Gould MK, Dembitzer AD, Doyle RL, Hastie TJ, Garber AM. Low-molecular-weight heparins compared with unfractionated heparin for the treatment of acute deep

Chapter 30: Venous Thromboembolism

19. 20. 21. 22. 23. 24.

25. 26. 27. 28.

29. 30. 31. 32.

33. 34.

35.

36.

venous thrombosis: a meta-analysis of randomized, controlled trials. Ann Intern Med 1999; 130:800–809. Hyers TM, Agnelli G, Hull RD, et al. Antithrombotic therapy for venous thromboembolic disease. Chest 2001; 119:176S–193S. Hyers TM. Venous thromboembolism. Am J Respir Crit Care Med 1999; 159:1–14. Bernardi E, Prandoni P. The post-thrombotic syndrome. Curr Opin Pulm Med 2000; 6(4):335–342. Milne AA, Ruckley CV. The clinical course of patients following extensive deep venous thrombosis. Eur J Vasc Surg 1994; 8(1):56–59. Patel NH, Plorde JJ, Meissner M. Catheter-directed thrombolysis in the treatment of phlegmasia cerulea dolens. Ann Vasc Surg 1998; 12:471–475. Prandoni P, Villalta S, Bagatella P, et al. The clinical course of deep-vein thrombosis. Prospective long-term follow-up of 528 symptomatic patients. Haematologica 1997; 82(4):423–428. Rogers JZ, Thomas P, Apovian C, Jensen GL. Pulmonary embolus as a complication of a central venous catheter. Nutrition 1996; 12:271–273. McCartney JS. Pulmonary embolism following trauma. Surg Gynecol Obstet 1935; 61:369–379. Sevitt S, Gallagher N. Venous thrombosis and pulmonary embolism: a clinico-pathological study in injured and burned patients. Br J Surg 1961; 48:475–489. Winchell RJ, Hoyt DB, Walsh JC, Simons RK, Eastman AB. Risk factors associated with pulmonary embolism despite routine prophylaxis: implications for improved protection. J Trauma 1994; 37:600–606. Owings JT, Kraut E, Battistella F, Cornelius JT, O’Malley R. Timing of the occurrence of pulmonary embolism in trauma patients. Arch Surg 1997; 132:862–867. Tai NR, Atwal AS, Hamilton G. Modern management of pulmonary embolism. Br J Surg 1999; 86:853–868. The urokinase pulmonary embolism trial: a national cooperative study. Circulation 1973; 47:II1–II108. The PIOPED Investigators. Value of the ventilation/ perfusion scan in acute pulmonary embolism. Results of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). JAMA 1990; 263:2753–2759. Green RM, Meyer TJ, Dunn M, Glassroth J. Pulmonary embolism in younger adults. Chest 1992; 101:1507–1511. Stein PD, Terrin ML, Hales CA, et al. Clinical, laboratory, roentgenographic, and electrocardiographic findings in patients with acute pulmonary embolism and no pre-existing cardiac or pulmonary disease. Chest 1991; 100:598–603. Stein PD, Terrin ML, Gottschalk A, Alavi A, Henry JW. Value of ventilation/perfusion scans versus perfusion scans alone in acute pulmonary embolism. Am J Cardiol 1992; 69:1239–1241. Stein PD, Hull RD, Saltzman HA, Pineo G. Strategy for diagnosis of patients with suspected pulmonary embolism. Chest 1993; 103:1553–1559.

387

37. Newman GE. Pulmonary angiography in pulmonary embolic disease. J Thorac Imaging 1989; 4:28–39. 38. Stein PD, Athanasoulis C, Alavi A, et al. Complications and validity of pulmonary angiography in acute pulmonary embolism. Circulation 1992; 85:462–468. 39. Mills SR, Jackson DC, Older RA, Heaston DK, Moore AV. The incidence, etiologies, and avoidance of complications of pulmonary angiography in a large series. Radiology 1980; 136:295–299. 40. Remy-Jardin M, Remy J, Artaud D, Deschildre F, Fribourg M, Beregi JP. Spiral CT of pulmonary embolism: technical considerations and interpretive pitfalls. J Thorac Imaging 1997; 12:103–117. 41. Drucker EA, Rivitz SM, Shepard JA, et al. Acute pulmonary embolism: assessment of helical CT for diagnosis. Radiology 1998; 209:235–241. 42. Velmahos GC, Vassiliu P, Wilcox A, et al. Spiral computed tomography for the diagnosis of pulmonary embolism in critically ill surgical patients: a comparison with pulmonary angiography. Arch Surg 2001; 136:505–511. 43. Meaney JF, Weg JG, Chenevert TL, Stafford-Johnson D, Hamilton BH, Prince MR. Diagnosis of pulmonary embolism with magnetic resonance angiography. N Engl J Med 1997; 336:1422–1427. 44. Ferris EJ, McCowan TC, Carver DK, McFarland DR. Percutaneous inferior vena caval filters: follow-up of seven designs in 320 patients. Radiology 1993; 188: 851–856. 45. Benjamin ME, Sandager GP, Cohn EJ Jr., et al. Duplex ultrasound insertion of inferior vena cava filters in multitrauma patients. Am J Surg 1999; 178:92–97. 46. Hughes GC, Smith TP, Eachempati SR, Vaslef SN, Reed RL II. The use of temporary vena caval interruption device in high-risk trauma patients unable to receive standard venous thromboembolism prophylaxis. J Trauma 1999; 46:246–249. 47. Patton JH Jr., Fabian TC, Croce MA, Minard G, Pritchard FE, Kudsk KA. Prophylactic Greenfield filters: acute complications and long term follow-up. J Trauma 1996; 41:231–236. 48. David W, Gross WS, Colaiuta E, Gonda R, Osher D, Launti S. Pulmonary embolus after vena cava filter placement. Am Surg 1999; 65:341–346. 49. Schmitz-Rode T, Janssens U, Schild HH, Basche S, Hanrath P, Gunther RW. Fragmentation of massive pulmonary embolism using a pigtail rotation catheter. Chest 1998; 114:1427–1436. 50. O’Malley KF, Ross SE. Pulmonary embolism in major trauma patients. J Trauma 1990; 30:748–750. 51. Douketis JD, Kearon C, Bates S, Duku EK, Ginsberg JS. Risk of fatal pulmonary embolism in patients with treated venous thromboembolism. JAMA 1998; 279: 458–462.

PART VI Complications of Trauma

31 Epidemiological, Organizational, and Educational Aspects of Trauma Care Michael E. Ivy Hartford Hospital, Hartford, and University of Connecticut School of Medicine, Farmington, Connecticut, U.S.A.

Tissues are injured when the energy transferred from the environment to the body exceeds the tolerance of the tissues (1). Any form of energy can result in injury, but the energy transferred usually takes the form of kinetic energy. Examples include the energy transferred during a sudden deceleration to the chest that results in a tear of the thoracic aorta, or the kinetic energy of a knife being transferred to a small area of the body. However, other forms of energy, such as thermal injury, can also cause tissue injury. Much of the credit for our appreciation of this basic concept can be given to Dr. William Haddon, Jr., an engineer and physician who was a pioneer in injury prevention. Injuries were responsible for 146,941 deaths in the United States in 1998 (2). Motor vehicle crashes (MVCs) were responsible for 28.8% of injury deaths, and firearms accounted for 20.9% (2). It should be noted that the Centers for Disease Control and Prevention (CDC) include poisonings and medical misadventures in this total. In the United States, injuries are the leading cause of death in persons in the age group of 1 to 44 years (2). Consequently, injury is responsible for the loss of more years of productive life (years lost before age 65) than is cancer or heart disease. The CDC categorizes injuries as intentional or unintentional (Table 1). In 1998, unintentional injuries were responsible for 94,331 deaths (2). The unintentional injuries that are the main causes of mortality are MVCs, falls, drownings, most thermal injuries, and most occupational injuries. Intentional injuries that lead to death are suicides, homicides, and the results of legal interventions. Intentional injuries can result from a variety of mechanisms, including gunshot, piercing, assault, and thermal injuries. The state of Massachusetts, one of the safest states in the United States, reviewed its experience with injuries in 1989 and estimated that one in four residents is injured every year (3). For every person who died as a result of injury, an estimated 17 were hospitalized and another 535 were injured but not hospitalized.

BACKGROUND Motor Vehicle Crashes MVCs are one of the main causes of death and disability in the United States and around the world. In the year 2000 in the United States, 37,338 crashes caused 41,800 fatalities (4). Overall during that year, an estimated 6,266,000 nonfatal crashes occurred and 3,219,000 people were injured. The rates of MVC-associated death vary widely between states. Much of the difference can be explained by the difference in the amount of time that passes before the victims reach a hospital. Although the overall number of deaths due to MVCs has decreased marginally over the past 35 years, the rates per population and per 100 million vehicle miles traveled (vmt) have dropped substantially (Table 2). Efforts aimed at reducing the rates of deaths and disability due to MVCs have resulted in an important public health triumph. In 1966, 5.5 deaths occurred per 100 million vmt in the United States, for a total of 50,894 deaths that year (5). In response, the U.S. Government, the insurance industry through its proxy the Insurance Institute of Highway Safety, and the auto industry worked together to identify and correct obvious hazards. In 2000, the number of deaths per 100 million vmt had decreased to 1.6, for a total of 41,800 deaths (4). If the fatality rate per 100 million vmt had remained unchanged at 5.5, the year 2000 would have seen approximately 140,000 deaths. This reduction translates into the prevention of more than 100,000 fatalities every year. The impact of motor vehicle fatality prevention programs largely goes unrecognized by the public, but it is important that physicians caring for trauma patients recognize and understand the impact of these interventions on the population.

Falls Falls are an important cause of loss of life and of hospitalization. In the United States, falls currently

390

Ivy

Table 1 Number of Traumatic Deaths for All Age Groups, 1992

All injuries Unintentional Suicide Homicide Undetermined Other

All ages

0–14 yrs

15–44 yrs

45–64 yrs

Over 65 yrs

Age not known

146,941 94,331 30,575 17,893 3,746 386

7,537 5,848 324 1,178 184 3

72,538 40,210 16,337 13,327 2,362 311

29,099 17,568 8,094 2,452 922 63

37,560 30,605 5,803 881 253 18

207 109 17 55 25 1

result in an estimated 13,301 deaths and nearly 783,357 hospitalizations each year (2). The mortality rate associated with falls is highest among the elderly: for Americans older than 85, the mortality rate is 108.7 per 100,000 persons. Consequently, falls may become a more important cause of fatal injuries as the population continues to age. Efforts are beginning to be made to decrease the number of hip fractures that result when elderly persons fall. Potential areas for improvement include better floor surfaces that more effectively absorb energy, as well as padding in clothes for the elderly who are deemed to be at substantial risk of hip fracture with a fall. Kannus et al. (6) recently reported a trial of an effective hip protector for elderly patients; the rate of hip fractures was 46.0 per 1000 person-years for unprotected patients and 21.3 per 1000 person-years for protected patients, a rate that makes this protector remarkably effective.

(18–24 years), a healthy economy, better police interventions, the imprisonment of larger numbers of criminal offenders, and the legalization of abortion in 1973.

Burns For a variety of reasons, the number of fatalities resulting from burns has decreased over the past several decades and this continues to the present. In 1999, an estimated 3910 Americans died of burn injuries, while in 2002, 3,645 died from similar injuries 6b. Some of the decrease is no doubt due to prevention efforts. Specific areas that have been addressed include flame-retardant clothing, elimination of freestanding propane or gas heaters from many homes, improved training of professional firefighters in most major cities across the United States, and the creation of dedicated burn centers to better care for patients with severe thermal injuries.

Gunshot Wounds The incidence of fatal gunshot wounds has decreased substantially over the past decade. In the United States in 1991, 38,317 people died of gunshot wounds while only 30,242 died in 2002. (6a,6b). Of the deaths in 2002, the majority (17,108) were suicides. Of the remainder, 11,829 were homicides and a much smaller number were either undetermined, unintentional, or the result of legal interventions. Persons at greatest risk of homicide are those aged 15–24 years, and most victims are male (2). The greatest progress in reducing mortality rates seems to have been made for slightly older Americans. For example, the mortality rate for persons aged 25 to 34 declined from 18.2 per 100,000 in 1979 to 11.8 per 100,000 in 1998, a decrease of 35%. The causes of this decline are subject to debate, but may include a decrease in illegal drug use, a decline in the number of persons in the highest-risk age group Table 2 Numbers and Rates of Fatalities Related to Motor Vehicle Crashes Over Time

Years

Fatalities

Fatalities per 100,000 population

1966 1976 1986 2000

50,894 45,523 46,087 41,800

26.02 20.92 19.19 15.20

Abbreviation: vmt, vehicle miles traveled.

Fatalities per 100 million vmt 5.5 3.2 2.5 1.6

TRAUMA SYSTEMS As expected, the development of an organized, systematic approach to the care of injured patients has improved outcomes in this population. A landmark study by West et al. in 1979 compared the preventable injury mortality rates for patients in Orange County with the rates for patients in San Francisco (7). This study documented a marked decrease in preventable trauma deaths in San Francisco and concluded that the trauma system had improved the quality of care. Subsequent studies have confirmed these findings. For example, Shackford et al. reviewed the San Diego County experience and noted that an organized trauma system had improved patient care (8). A recent review of the trauma experience in England and Wales from 1989 to 1997 found that the odds ratio of mortality from injury was 0.72 in 1997; compared to that in 1989, this indicates a decrease in the mortality rate of nearly 30% (9). The timing of deaths due to injury has been another area of investigation. Baker et al. reviewed all traumatic deaths that occurred in San Francisco in 1977 (10). They noted that the distribution of deaths was trimodal: 53% occurred before arrival at the hospital, 21.5% occurred within 48 hours after arrival at the hospital, and 12.6% occurred more than seven days after injury. Of the late deaths, 78% were due to sepsis and multiple organ failure. Interestingly,

Chapter 31:

Epidemiological, Organizational, and Educational Aspects of Trauma Care

only 5% of deaths from penetrating trauma and 8% of deaths from blunt trauma were due to sepsis. On the other hand, burn patients were at substantial risk of death due to sepsis. (At that time, early excision of burn wounds was not routinely practiced.) Overall, 50% of deaths were due to brain injury, 31% were due to exsanguination, and 9.8% were due to sepsis and organ failure. Sauaia et al. performed a similar study in Denver County in 1992 (11). This study showed that 34% of deaths occurred during the prehospital phase, 53% occurred within 48 hours of arrival at the hospital, and only 9% were classified as late deaths. Most late deaths (61%) resulted from organ failure. Again, late deaths due to organ failure were unusual after penetrating trauma (3%), but were slightly more common after blunt trauma (13%). The authors reported that, overall, 42% of the deaths were due to brain injuries, 39% were due to exsanguination, and 7% were due to organ failure. These findings show a trend toward improvement in the care of hospitalized trauma patients, with fewer deaths due to late sepsis and organ failure, although improved burn care is at least partly responsible for the decrease in late deaths due to sepsis. The findings also show that because of faster and possibly improved prehospital care, the number of prehospital deaths has decreased, but the number of early hospital deaths has increased. The percentage of deaths due to sepsis and organ failure decreased from 9.8% (10) to 7% (11), a decrease of nearly 30% in the rate of death due to sepsis. Consequently, the distribution of deaths, which was trimodal in earlier studies, appears to be bimodal at present, although the difference in the percentage of late deaths is not great (12% in 1977; 9% in 1992). In 1991, Davis et al. reviewed the San Diego County Trauma System data from 1985 to 1988 and investigated the importance of errors in critical care delivered to trauma patients (12). Overall, 813 deaths occurred at trauma centers; 62 of these deaths were judged to have been preventable. Of the preventable deaths, nearly half (30 deaths) were due to errors in critical care. Most (67%) of these errors that were related to preventable deaths were errors in management. The authors concluded that errors in critical care are a substantial cause of mortality among trauma patients, and that surgeons caring for trauma patients must be skilled in caring for critically ill patients. Trauma systems have other important goals besides minimizing the number of preventable deaths; these goals include minimizing morbidity rates and maximizing rehabilitation efforts so that injured persons can return to work and function successfully in society. Rhodes et al. (13) found that 83% of 302 patients who had been admitted to their trauma center had returned to work within six months (13). After three years, 81% of patients who had sustained severe injuries, as defined by an Injury Severity Score (ISS) greater than 15, had returned to work. Documenting the effectiveness of trauma centers and

391

trauma systems is important if we are to justify the continuing investment by society and the medical community in maintaining and improving trauma care. Additionally, these findings and those of related studies are needed if we are to convince governments to expand the funding of trauma systems. A trauma system encompasses the care of the patient from the scene of the injury, through transport to the initial hospital, through transport to a definitive acute care hospital if necessary, and later through the rehabilitation phase. This level of organization requires the participation of hospital personnel, prehospital personnel, and officials of the local, county, and state governments. At the trauma center, a coordinated response involves the emergency department (ED), the operating room, the intensive care units, the inpatient floors, the radiology department, the blood bank and laboratory, and eventually the rehabilitation center. Such a coordinated response mandates the availability of skilled physicians who are trained in trauma care. Additionally because of the breadth in the variety of injuries that can occur, a wide range of surgical and nonsurgical specialists must be available. Surgical specialists included in the system are general surgeons, neurosurgeons, orthopedic surgeons, and plastic surgeons. Essential nonsurgical specialists include general internists, cardiologists, gastroenterologists, psychiatrists, radiologists, and pediatricians. A commitment to obtain the resources, talents, and abilities required by a trauma center can raise the overall level of care that a hospital offers, but it also requires an investment of time, money, and effort on the part of the administrators.

PREHOSPITAL CARE Rapid Prehospital Response The Emergency Medical Services (EMS) system is in many ways the descendent of the military evacuation systems that have been used for centuries, beginning with Baron Larrey during the Napoleonic Wars. According to McSwain (14), Larrey established the importance of three concepts: (i) rapid arrival of a well-equipped ambulance at the site of the injury; (ii) provision of prehospital care by educated, skilled personnel; and (iii) rapid transport from the scene to a hospital that can care for injured patients. The military continued to develop this concept during World Wars I and II, using mechanized vehicles for transport and mobile hospitals to provide definitive care soon after injury. In Korea, the military began to use helicopters to provide rapid transport to more secure hospitals. The civilian use of these concepts remained limited in the United States until the publication in the 1960s of the National Academy of Sciences report ‘‘Accidental Death and Disability: The Neglected Disease of Modern Society.’’ This report called for substantial improvements in the provision of prehospital trauma care.

392

Ivy

Education of Prehospital Personnel It is vital that injured patients receive organized care in the field, where the situation is often chaotic. Bringing order to such a scene and providing a high level of care can be quite difficult. Personnel must be specifically trained to provide this care and, over time, this requirement has mandated the creation of courses for prehospital care providers, from emergency medical technicians (EMTs) to paramedics. In 1969, the Department of Transportation published a manual for prehospital personnel and created a specific curriculum for the Emergency Medical TechnicianAmbulance (EMT-A). Further development of prehospital education occurred rapidly as its effectiveness was recognized. The National Association of EMTs created the Prehospital Trauma Life Support Course along guidelines similar to those of the Advanced Trauma Life Support Course that is offered to physicians (14). The American College of Emergency Physicians developed the Basic Trauma Life Support Course. Each of these courses is now regularly offered in locations across the United States to persons interested in providing prehospital care. Education of prehospital personnel should be and is supported by trauma centers. Such education requires that prehospital personnel participate in standard courses and that opportunities for continuing medical education be created for prehospital personnel.

Communication Communication between prehospital personnel and hospital physicians is of paramount importance, not only for trauma patients but also for a wide variety of patients with critical illnesses such as myocardial infarctions, arrhythmias, and cerebrovascular accidents. Communication is beneficial in several ways. It allows the physician to direct the care of critically injured patients during transport to the hospital, it allows hospital personnel to prepare for specific types of injuries, and it offers the opportunity to direct the injured patient to a hospital that offers an appropriate level of care. It is also important to maintain communication between prehospital personnel and physicians after the patient has been transported. Patients can benefit from the development of protocols and standing orders, but such protocols can be appropriately developed only with the collaborative efforts of all parties involved.

Triage Triage of injured patients occurs at many levels and is a crucial part of the trauma system. Triage began, as did much of the prehospital system of care, as a result of the efforts of Baron Larrey. Two important facets of the epidemiology of trauma combine to make rapid and accurate triage a vital component of any trauma system. The first factor is the recognition that nearly

50% of trauma patients die within the first four hours after injury (11). Most of these deaths are not preventable; however, some of them may be prevented with a fast and accurate assessment of the injuries involved. The second factor is that only 5% to 10% of trauma patients have injuries severe enough to require the services of a trauma center. Consequently, the EMT must triage the remaining 90% of injuries to find the patients who need rapid transport to a trauma center. The alternative is to treat all trauma patients with the same level of urgency, but in such a scenario, most systems would be overwhelmed by the minimally injured, and the care of the seriously injured would be compromised. Triage in the field is based on mechanism of injury and physiologic parameters. Frequently, a second triage occurs when patients arrive at the ED. Mechanism of injury plays an important role in determining a patient’s need for definitive care at a trauma center; triage criteria include MVC with ejection and penetrating injuries of the head, neck, or torso. Examples of physiologic triage criteria include systolic blood pressure below 90 mmHg or a Glasgow Coma Scale (GCS) score of less than 12. Patients can be reassigned to a different triage level on the basis of their initial hospital course while they are still in the ED. For example, a patient whose condition originally appears stable after a fall from a standing position may initially be assigned a low triage level in the field and in the ED, but this triage level may be subsequently upgraded if acute changes in mental status occur. The evaluation and management of mass casualties is another important component of a welldesigned trauma system. Incidents that result in a large number of casualties are relatively infrequent, but unless they are planned and practiced for, can result in a disproportionate number of fatalities. It is important that experienced personnel be present in the field to direct the expeditious triage of victims. Several different triage systems have been used over the years, but all have several features in common. Minimally injured patients are considered walking wounded, and in civilian settings, they are triaged to less urgent care. Patients with very severe injuries are triaged to the expectant category and are given palliative care. Patients with serious injuries who are expected to benefit from relatively minimal interventions are assigned the highest priority. The complexity of the decision-making process, and the need to make the decisions quickly and efficiently make the job very complex. These incidents typically require a coordinated response from area hospitals so that the resources available at any one center will not be completely overwhelmed.

Interventions in the Field The effectiveness of specific interventions in the prehospital setting has been evaluated. The insertion of an endotracheal tube is clearly beneficial for patients

Chapter 31:

Epidemiological, Organizational, and Educational Aspects of Trauma Care

with severe brain injury (26). There is good evidence that hypoxia soon after a severe brain injury greatly increases the risk of death and the risk of a poor outcome (27). Brain-injured patients may retain CO2, and this retention can result in inappropriate cerebral vasodilation and increased intracranial pressure (ICP). High levels of CO2 will also cause serious acidosis, which may lead to cerebral vasodilation and even higher ICP. Additionally, unconscious patients may actively aspirate oral and gastric secretions until an endotracheal tube is in place and secured. Stewart et al. (28) studied the ability of prehospital personnel to perform oral intubation of comatose patients in the field. They reported an overall 90% success rate for intubation in the field, although the success rate was only 79% for trauma patients. The complication rate was nearly 10%, although many of the complications, such as vomiting during intubation (0.9%), may well be unavoidable even for more experienced personnel. Attempts by trained hospital personnel to perform intubation on comatose patients in the field are justified. The effectiveness of administering intravenous fluids to patients before arrival at the hospital is less certain. A retrospective study by Kaweski et al. (29) found no difference in survival rates related to the prehospital administration of fluids. A highly publicized clinical trial in Houston evaluated the intravenous administration of fluids to patients with penetrating trauma to the torso (30). In that study, patients received either standard fluid resuscitation in the field or no fluid resuscitation before arrival in the operating room. The survival rate for patients who received no fluids was significantly higher than that for patients who received standard fluids (p ¼ 0.04). Thus, it appears that the prehospital administration of fluids may be harmful to this specific subpopulation of trauma patients. However, random assignment of patients to the groups was not properly carried out, the treatment options were offered on alternate days, the fluid volume regimens were not strictly followed, and the preoperative times were longer than would be acceptable at most centers. These problems lead to concerns about the generalizability of the study’s findings. Unfortunately, no properly performed, prospective, randomized controlled trial has been performed to clarify the issue. Other studies have indicated that the type of fluid given in the prehospital setting may affect the outcome of patients with head injury. A prospective randomized trial was conducted to evaluate the effectiveness of administering hypertonic saline and dextran (HSD) to injured patients in the prehospital setting (31). The rationale behind the study was that HSD is a better plasma volume expander than is isotonic crystalloid solution. Consequently, the small volumes given before arrival at the hospital could rapidly increase the blood pressure of hypotensive trauma patients and improve perfusion to vital organs while the patient is undergoing resuscitation and definitive treatment. The study found that, overall,

393

there were no significant differences in outcome between patients given HSD and those given other types of resuscitation fluids. However, the outcomes were better (p ¼ 0.068) for the subgroup of patients with a head Abbreviated Injury Score of 4, 5, or 6 who were treated with HSD. This finding is consistent with those of other studies that have documented a significantly worse outcome for patients with brain injury, who experience even transient hypotension. In fact, Chesnut et al. (27) reported that a 50% increase in mortality rates was associated with one or more episodes of hypotension experienced in the ED by patients with a GCS score of less than 8. Given the demonstrated importance of hypotension in the ED in determining the outcome of patients with brain injury, it may well be that HSD can better minimize hypotension during the early postinjury period and therefore minimize the risk of secondary brain injury. Additionally, on the basis of numerous studies that began with that by Weed and McKibben in 1919, it has been well established that the sodium concentration of the infused fluid plays a key role in the volume of fluid in the brain (32). It may simply be the case that administering hypertonic saline soon after injury increases the sodium concentration of the plasma and shrinks the extracellular volume of the brain, thereby lowering ICP and increasing cerebral perfusion pressure. The other important aspect of this study is the finding that aggressive fluid resuscitation administered before bleeding has been controlled may increase bleeding and subsequently put the patient at risk of developing complications and a higher likelihood of mortality. The results of good animal studies demonstrate the adverse consequences of excess fluid resuscitation in the early stages of uncontrolled hemorrhage (33). This adverse consequence of blood pressure restoration in some patients may compensate for the benefit achieved in others and may explain the lack of difference in overall outcomes.

HOSPITAL CARE To be a successful trauma center, a hospital must be committed as an organization to providing excellent trauma care. A hospital must meet certain standards if it is to be classified as a trauma center. Its ED must be open 24 hours a day and it must be adequately staffed with personnel who have been trained to care for injured patients. These personnel must complete continuing education courses on trauma care. An operating room must be immediately available and appropriately staffed at all times. This requirement is of paramount importance if life-saving operations are to be performed in a timely manner. Because the most severely injured patients will require intensive care after arrival, an intensive care unit must be available and its surgeons must be involved in the care of the patients. As patients recover from their injuries, they

394

Ivy

must receive physical therapy, occupational therapy, and speech therapy as appropriate for their injuries. The complexity of the trauma center concept prompted the American College of Surgeons (ACS) to develop a process for verification and consultation in 1987. The Trauma Center Verification program has resulted in an organized approach to the care of injured patients in designated trauma centers. In implicit recognition of the fact that hospitals in some regions of the United States cannot muster the resources to become Level I or Level II trauma centers, the ACS developed guidelines that encompass several levels of care. This approach is based on the belief that care can be improved if a hospital develops an organized trauma system and has a coordinated response to the arrival of a trauma patient. The ACS mandated the creation of quality improvement processes in trauma centers to ensure that cases are reviewed, errors are identified, and, when possible, actions are taken to prevent similar errors in the future.

number of patients admitted with ISS higher than 15. Several of the hospitals included in their study are very high volume trauma centers, and all of the centers in the study would probably meet the criteria for highvolume centers in the other two studies discussed above. These authors found that an increasing volume was associated with a statistically significant increase in mortality rates. This association may result from the fact that very high volumes of trauma patients can occasionally overwhelm the available resources at trauma centers. On the basis of the findings of these studies, it appears that the survival rates of severely injured patients are higher at high-volume trauma centers. Between 600 and 650 trauma admissions are needed annually if trauma centers are to maintain an optimal level of care. However, it is possible that very high volumes may compromise care if hospital resources are overwhelmed.

Physician Involvement Trauma Center Volume Substantial debate persists about the optimal volume of trauma patients that a hospital should care for in a year. It seems intuitively obvious that centers with very low volume will provide less than ideal care, but the exact volume that will allow the best care has not been determined. Three recent studies have examined this issue in depth. Nathens et al. (15) performed a multivariate analysis of the data from the University Healthsystem Consortium Trauma Benchmarking Study and found that hospital volume was a significant predictor of survival for high-risk patients. Specifically, for patients who arrived at the hospital with hypotension resulting from penetrating abdominal trauma, the odds ratio for mortality was 0.02 at high-volume centers as compared with that at low-volume centers. For comatose patients with multisystem blunt trauma, the odds ratio for mortality was 0.49 at high-volume centers as compared with that at low-volume centers. There was no statistically significant difference in mortality rates for patients with less severe injuries, such as penetrating abdominal trauma without hypotension. In this study, the volume used to distinguish low-volume and high-volume trauma centers was 650 trauma admissions per year. Pasquale et al. (16) analyzed the Pennsylvania trauma center database and found that, of all factors reviewed, only the trauma center’s case volume affected survival rates. The 12 busiest centers were considered high-volume trauma centers, and the 12 slowest centers were considered low-volume trauma centers. The author identified that the transition from low-volume to high-volume center occurred between 607 and 627 trauma admissions per year. Margulies et al. (17) reviewed the experience at five Level I trauma centers in Los Angeles County. Their analysis of hospital volume was based on the

Trauma care is an integral part of general surgery. It is essential that general surgeons be intimately involved in the care of patients at a trauma center. This does not mean that all general surgeons should be caring for trauma patients. There are a number of reasons why surgeons do not want to care for injured patients. Trauma care requires caring for patients who are often uninsured and the care is frequently needed at very inconvenient times. Directing trauma resuscitation can be a very intense experience and many surgeons are simply not interested in providing this kind of care. A separate concern is that having too many surgeons on the call schedule will dilute the trauma experience and could decrease the quality of care being provided. Consequently, in most large hospitals, many of the general surgeons do not take trauma call. However, in smaller communities, it is necessary for all or most of the general surgeons to participate in the care of injured patients. This necessity can and should be viewed as an opportunity to serve the community in a unique and special fashion. Unfortunately, most general surgeons simply do not want to care for trauma patients. When Orange County attempted to organize a trauma system in 1980, only 23 of 225 general surgeons offered to take trauma call (18). A survey by Richardson and Miller found that only 20% of surgery residents wanted to care for trauma patients as part of their practice (19). Those who are willing to take trauma call frequently demand reimbursement for being on the call schedule. A growing concern in the trauma community is the decreasing number of operative trauma cases and the need for surgeons who are providing trauma care to maintain their operative skills. It is likely that trauma surgeons will need to expand their general surgery practice if they are to perform enough surgical procedures to maintain their operating skills at a high level.

Chapter 31:

Epidemiological, Organizational, and Educational Aspects of Trauma Care

Volume of Trauma Cases for Individual Surgeons There is debate about the volume of trauma patients that individual surgeons need to treat if they are to provide optimal care. It seems intuitively obvious that, as is the case for trauma centers, surgeons need some volume of patients if they are to maintain their skills. The ACS has essentially declared that surgeons must care for at least 35 trauma patients with an ISS higher than 15 each year if they are to maintain an optimal level of expertise (20). This declaration has been based on the findings of the Pennsylvania study reported by Konvolinka et al. in 1995 (21). This study found that the outcome improved when surgeons treated an average of at least 35 trauma patients with an ISS higher than 12 each year. The study did not examine the case volume of individual surgeons but examined instead the average volume over a system; additionally, it used an ISS higher than 12 instead of an ISS higher than 15 as the definition of serious injury. Other studies have not found a strong correlation between a trauma surgeon’s case volume and patient outcome. A more recent review of the Los Angeles County experience found no link between trauma surgeon volume and survival rates (17). Surgeons in this study treated an annual average of 10 to 20 patients with an ISS higher than 15. It is not clear if this study really addresses the ACS guidelines, but it is remarkable that none of the surgeons in the five Level I trauma centers in Los Angeles County meet the proposed case volume requirements. The debate about case volume will continue, but at present there is no convincing evidence that surgeons need an annual volume of more than 35 injured patients with an ISS higher than 15.

Coverage by In-House Attending Physicians Another raging debate concerns the need for in-house trauma coverage by attending physicians at trauma centers. The results of several studies are contradictory. The study by Thompson et al. (22) determined that the presence of an in-house trauma-attending physician was irrelevant to patient outcome. However, a study by Rogers et al. (23) at two university Level I trauma centers in the same metropolitan area found that coverage by an in-house attending physician improved outcome. Their study has been criticized because of concerns that other differences between the centers could have confounded its results. Luchette et al. (24) reported that the presence of an in-house surgeon with added qualifications in surgical critical care resulted in more rapid resuscitation but did not decrease mortality rates. The study by Pasquale et al. (16) found no increase in survival in association with the presence of an in-house trauma surgeon.

Training of Surgical Residents There is a growing concern about the adequacy of training for general surgery residents in trauma care;

395

this concern is specifically focused on operative volume, which for several reasons is decreasing across the country. Automobiles and roads are safer than in the past, and, although airbags may increase the incidence of orthopedic trauma, they are likely to decrease the incidence of other types of injuries. The increase in the use of nonoperative management for blunt trauma to the liver, spleen, and kidneys has greatly decreased residents’ operative experience. Hawkins et al. (25) reviewed the operative trauma experience of their residents and noted a 25% decrease from 1991–1993 to 1994–1996. The rising popularity of angiographic embolization for managing injuries to solid organs will also decrease the operative experience at some centers. A final reason for the decrease in operative volume is the decrease in the incidence of penetrating trauma in almost all cities across the United States. How many operative trauma cases are enough to develop a surgeon’s competency in caring for injured patients? How do we ensure that future trauma surgeons perform at least this minimum number of operations? These questions remain unanswered.

Education of Hospital Personnel The outreach efforts of the ACS have been instrumental in improving the care of injured patients. The creation of the Advanced Trauma Life Support Course has standardized the approach to injured patients and allowed a large number of surgeons and other medical personnel to improve their skills and knowledge. The ACS also requires that surgeons who care for trauma patients at a trauma center obtain continuing medical education credits on topics specifically related to trauma issues. To assure that surgeons can obtain such credits, the ACS annually sponsors several outstanding educational conferences for surgeons interested in trauma care. The ACS manual on the care of the surgical patient includes several sections that deal specifically with the treatment of injured patients. The creation of the field of emergency medicine has increased the number of EDs that are staffed with physicians who have been formally trained in providing trauma care.

RURAL AND URBAN TRAUMA CENTERS The trauma system approach has been very successful in metropolitan and suburban settings. Unfortunately, much work remains to be done in rural settings. The fatality rates due to all kinds of injuries are much higher in rural settings than in urban or suburban settings. Although the rural population accounts for only one-third of the population of the United States, it accounts for more than 50% of the MVC fatalities (34). Problems include a longer time between an accident and the recognition that an accident has occurred, a longer time before the arrival of prehospital personnel, long transport times to Level II or III trauma centers, and long transport times, often via

396

Ivy

helicopters or airplanes, to more definitive care centers. Such delays can clearly be life-threatening.

TRIAGE IN THE HOSPITAL Once the patient has reached the hospital, the value of a well-organized response to trauma has been clearly demonstrated and is widely accepted in the United States. Competing concerns about costs and resource utilization have mandated a graded response to trauma. The highest response should be prompted by several physiologic criteria, including hypotension, a respiratory rate of more than 40 breaths per minute, a GCS of less than 8, and penetrating injury to the head, neck, chest, or abdomen. The ACS recognizes that triage is not an exact science, so it mandates that rates of overtriage and undertriage deficiencies.

MANAGEMENT OF PEDIATRIC TRAUMA There is ongoing debate about the role of pediatric surgeons in the treatment of injured children. D’Amelio et al. (35) published their experience with pediatric trauma patients treated by surgeons not trained as pediatric surgeons. Their results compared favorably with those obtained by a review of national standards. Consequently, these authors stated that, in areas with a shortage of pediatric surgeons, general surgeons can safely and effectively treat injured children. Tepas et al. (36) analyzed a multi-institutional pediatric trauma registry and reported that mortality rates were higher in very busy pediatric trauma centers. These authors suggested that this fact may reflect the inadequacy of available resources at these particularly busy centers.

PREVENTION OF TRAUMA The science of injury analysis and prevention began in World War II, largely because of the efforts of Colonel Hugh DeHaven. In 1942, DeHaven published a review of the U.S. military’s experience with soldiers who had survived falls from heights of 50 to 150 ft (37). He demonstrated that the forces on the body could be decreased to tolerable levels if the deceleration was extended over a brief but significant time and was spread out over a substantial area of the body. The implication of this work, that extending the deceleration time spreads out the forces over a greater area of the body, was further investigated. A leading force in the early analysis of injury was Colonel John Stapp. The U.S. Air Force was interested in decreasing the likelihood of injury during airplane crashes. The construction of a rocket sled with controls for speed or deceleration allowed the testing of a variety of innovations in safety equipment. Stapp rode the sled at a speed of 632 mph and stopped it within 1.4 seconds; he sustained only minor injuries, in large part because he was wearing a four-point restraint (38). This type of testing was expanded to include automobile crash

testing, and innovations aimed at improving safety soon followed. Nevertheless, before 1965, the highways in America were exceedingly hazardous and little progress was being made in efforts aimed at reducing the numbers of deaths occurring on the highways. In 1966, the government created the National Highway Transportation Safety Administration (NHTSA) and appointed William Haddon Jr., as its director. Dr. Haddon’s approach to injury prevention focused on passive interventions, those that prevent injuries but do not require the active participation of the person at risk of injury (39). For example, the use of restraints had been known since Colonel Stapp’s work in injury prevention, but restraints require the active and willing participation of the person at risk and therefore had not been particularly successful in preventing MVC fatalities in the United States. Helmets were known to be effective in preventing brain injuries, but such knowledge was not and still is not a serious consideration for passenger vehicles because of the need for active participation. In contrast, the installation of shatterproof glass in windshields is a very effective way of preventing certain types of injuries. In 1972, Haddon (40) introduced the Haddon Matrix, a tool used to analyze injury events and to design preventive measures. The matrix is essentially a 3
3 matrix with human, vehicular, and environmental factors on one side and event phases (preevent, event, and post-event) on the other (Table 3). The effectiveness of the Haddon matrix in decreasing the occurrence of injurious events and in minimizing the consequences of those events when they do occur can be clearly demonstrated by the decline in MVC fatalities over the past three decades. Lifesaving interventions have resulted from the use of the Haddon matrix and the combined efforts of the NHTSA, the automobile industry, the construction industry, and the hospital industry, as well as consumer advocacy groups. These interventions include crumple zones in cars, padding of dashboards and steering wheels, better seat belts, three-point restraints, child car seats, air bags in numerous locations, wider highways, breakdown lanes, breakaway light poles, and Jersey barriers, as well as the creation of trauma systems and prehospital emergency care systems.

Table 3 Sample Haddon Matrix for Motor Vehicle Crashes Phases Pre-event Event

Post-event

Factors Driver’s education Cell phone use

Bystander first aid

Center brake Speed bumps, light street lights Air bag; padded Guard rails; Jersey dashboard barriers and steering wheel Measures to decrease EMS, trauma incidence of center car fires

Abbreviation: EMS, emergency medical services.

Chapter 31:

Epidemiological, Organizational, and Educational Aspects of Trauma Care

It is also important to study innovations to ensure that they are effective. There are several classic examples of ideas that seemed to be effective, but either were not effective in practice or needed modification. Robertson et al. (41) conducted a study evaluating the impact of an intensive advertising campaign on the use of seat belts. Two cities were randomly assigned to receive advertising or no advertising through their cable television system. The intervention had no impact on the rate of seat belt use. Driver’s education programs are commonly believed to be a logical way to save the lives of adolescents. When the State of Connecticut eliminated mandatory driver’s education classes, some towns and cities continued to offer driver’s education classes, although others did not. Robertson (42) reported that significantly fewer MVC fatalities occurred among teenagers in the towns and cities that did not offer the driver’s education classes. The towns that did offer the courses experienced lower rates of MVCs per teenaged driver, but because there were so few teenaged drivers in the areas that did not offer the classes, the rate of MVCs per teenaged driver was much lower in these areas. It is reasonable to conclude that, although driver’s education programs are better than no education for individual teenaged drivers, these programs are not as effective in decreasing the overall incidence of MVCs as is delaying licensure for two more years. Installation of air bags in cars and light trucks was a much anticipated improvement in the safety of automobiles. Air bags are credited with a 16% decrease in driver fatalities and a 23% decrease in deaths due to frontal crashes (43). Even for passengers, air bags are responsible for a 14% decrease in the rate of fatalities due to frontal crashes among passengers wearing seat belts and a 23% decrease in such fatalities among those not wearing belts (44). Unfortunately, air bags are also associated with a 34% increase in the risk of death among children under the age of 10 who ride in the front passenger seat. In an attempt to minimize this risk, important educational efforts have been undertaken and air bags have been modified. Innovations that reduce the number of deaths due to injury will continue to be introduced. As surgeons caring for the people in our communities, we need to ensure that the innovations are effective. If they are, we need to actively promote their use.

REFERENCES 1. Haddon W Jr. A note concerning accident theory and research with special reference to motor vehicle accidents. Ann N Y Acad Sci 1963; 107:635–646. 2. Murphy SL. National Vital Statistics Reports. Vol. 48, Number 11. 3. Schuster M, Cohen BB, Rodgers CG, Walker DK, Friedman DJ, Ozonoff VV. Overview of causes and costs of injuries in Massachusetts: a methodology for analysis of state data. Public Health Rep 1995; 110:246–250.

397

4. National Highway Transportation Safety Administration. 1999 Annual Report and 2000 early assessment files. Washington D.C.: U.S. Department of Transportation, 2001. 5. National Highway Traffic Safety Administration. Traffic Safety Facts 1995. Washington D.C.: U.S. Department of Transportation, 1996. 6. Kannus P, Parkkari J, Niemi S, et al. Prevention of hip fracture in elderly people with use of a hip protector. N Engl J Med 2000; 343:1506–1513. 6a. National Center for Health Statistics, Center for Disease Control and Prevention. National Vital Statistics Report 199. Vol 48. 2000. 6b. National Center for Health Statistics, Center for Disease Control and Prevention. National Vital Statistics Report 2002. Vol 54, Number 10. 2006. 7. West JG, Trunkey DD, Lim RC. Systems of trauma care. A study of two counties. Arch Surg 1979; 114:455–460. 8. Shackford SR, Hollingworth-Fridlund P, Cooper GF, Eastman AB. The effect of regionalization upon the quality of trauma care as assessed by concurrent audit before and after institution of a trauma system: a preliminary report. J Trauma 1986; 26:812–820. 9. Lecky F, Woodford, Yates DW. Trends in trauma care in England and Wales 1989–97. UK Trauma Audit and Research Network. Lancet 2000; 355:1771–1775. 10. Baker CC, Oppenheimer L, Stephens B, Lewis FR, Trunkey DD. Epidemiology of trauma deaths. Am J Surg 1980; 140:144–150. 11. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma deaths: a reassessment. J Trauma 1995; 38: 185–193. 12. Davis JW, Hoyt DB, McArdle MS, Mackersie RC, Shackford SR, Eastman AB. The significance of critical care errors in causing preventable deaths in trauma patients in a trauma system. J Trauma 1991; 31:813–819. 13. Rhodes M, Aronson J, Moerkirk G, Petrash E. Quality of life after the trauma center. J Trauma 1988; 28: 931–938. 14. McSwain NE Jr. Pre-hospital care. In: Feliciano DV, Moore EE, Mattox KL, eds. Trauma. Stamford, CT: Appleton & Lange, 1996:107–122. 15. Nathens AB, Jurkovich GJ, Maier RV, et al. Relationship between trauma center volume and outcomes. JAMA 2001; 285:1164–1171. 16. Pasquale MD, Peitzman AB, Bednarski J, Wasser TE. Outcome analysis of Pennsylvania trauma centers: factors predictive of nonsurvival in seriously injured patients. J Trauma 2001; 50:465–474. 17. Margulies DR, Cryer HG, McArthur DL, Lee SS, Bongard FS, Fleming AW. Patient volume per surgeon does not predict survival in adult level 1 trauma centers. J Trauma 2001; 50:597–603. 18. Trunkey DD. What’s wrong with trauma care? Bull Am Coll Surg 1990; 75:10–15. 19. Richardson JD, Miller FB. Will future surgeons be interested in trauma care? Results of a resident survey. J Trauma 1992; 32:229–235. 20. American College of Surgeons Committee on Trauma. Optimal Hospital Resources for Care of the Injured Patient. Chicago: American College of Surgeons, 1999. 21. Konvolinka CW, Copes WS, Sacco WJ. Institution and per-surgeon volume versus survival outcome in Pennsylvania’s trauma centers. Am J Surg 1995; 170: 333–340.

398

Ivy

22. Thompson CT, Bickell WH, Siemens RA, Sacra JC. Community hospital level II trauma center outcome. J Trauma 1992; 32:336–343. 23. Rogers FB, Simons R, Hoyt DB, Shackford SR, Holbrook T, Fortlage D. In-house board-certified surgeons improve outcome for severely injured patients: a comparison of two university centers. J Trauma 1993; 34:871–877. 24. Luchette F, Kelly B, Davis K, Johanningman J, Heink N, James L, Ottaway M, Hurst J. Impact of the in-house trauma surgeon on initial patient care, outcome, and cost. J Trauma 1997; 42:490–497. 25. Hawkins ML, Wynn JJ, Schmacht DC, Medeiros RS, Gadacz TR. Nonoperative management of liver and/ or splenic injuries: effect on resident surgical experience. Am Surg 1998; 64:552–557. 26. Winchell RJ, Hoyt DB. Endotracheal intubation in the field improves survival in patients with severe head injury. Trauma Research and Education Foundation of San Diego. Arch Surg 1997; 132:592–597. 27. Chesnut RM, Marshall LF, Klauber MR, et al. The role of secondary brain injury in determining outcome from severe head injury. J Trauma 1993; 34:216–222. 28. Stewart RD, Paris PM, Winter PM, Pelton GH, Cannon GM. Field endotracheal intubation by paramedical personnel. Success rates and complications. Chest 1984; 85:341–345. 29. Kaweski SM, Sise MJ, Virgilio RW. The effect of prehospital fluids on survival in trauma patients. J Trauma 1990; 30:1215–1219. 30. Bickell WH, Wall MJ Jr, Pepe PE, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med 1994; 331:1105–1109. 31. Vassar MJ, Perry CA, Gannaway WL, Holcroft JW. 7.5% sodium chloride/dextran for resuscitation of trauma patients undergoing helicopter transport. Arch Surg 1991; 126:1065–1072. 32. Weed LH, McKibben PS. Experimental alteration of brain bulk. Am J Physiol 1919; 48:531–558. 33. Capone AC, Safar P, Stezoski W, Tisherman S, Peitzman AB. Improved outcome with fluid restriction

34.

35.

36.

37. 38. 39.

40. 41.

42. 43. 44.

in treatment of uncontrolled hemorrhagic shock. J Am Coll Surg 1995; 180:49–56. Rogers FB, Osler TM, Shackford SR, Martin F, Healey M, Pilcher D. Population-based study of hospital trauma care in a rural state without a formal trauma system. J Trauma 2001; 50:409–414. D’Amelio LF, Hammond JS, Thomasseau J, Sutyak JP. ‘‘Adult’’ trauma surgeons with pediatric commitment: a logical solution to the pediatric manpower problem. Am Surg 1995; 61:968–974. Tepas JJ III, Patel JC, DiScala C, Wears RL, Veldenz HC. Relationship of trauma patient volume to outcome experience: can a relationship be defined? J Trauma 1998; 44:827–831. De Haven H. Mechanical analysis of survival in falls from heights of fifty to one hundred and fifty feet. War Med 1942; 2:586–596. Stapp JP. Effects of mechanical force on living tissues. I. Abrupt deceleration and windblast. J Aviat Med 1955; 26:268–288. Haddon W Jr, Goddard JL. An analysis of highway safety strategies. In: Passenger car design and highway safety. Association for the Aid of Crippled Children and Consumers Union of New York, 1962: 6–11. Haddon W Jr. A logical framework for categorizing highway safety phenomena and activity. J Trauma 1972; 12:193–207. Robertson LS, Kelley AB, O’Neill B, Wixon CW, Eiswirth RS, Haddon W Jr. A controlled study of the effect of television messages on safety belt use. Am J Public Health 1974; 64:1071–1080. Robertson LS. Crash involvement of teenaged drivers when driver education is eliminated from high school. Am J Public Health 1980; 70:599–603. Lund AK, Ferguson SA. Driver fatalities in 1985– 1993 cars with airbags. J Trauma 1995; 38:469– 475. Braver ER, Ferguson SA, Greene MA, Lund AK. Reductions in deaths in frontal crashes among right front passengers in vehicles equipped with passenger air bags. JAMA 1997; 278:1437–1439.

32 Competing Priorities in the Trauma Patient Michael E. Ivy Hartford Hospital, Hartford, and University of Connecticut School of Medicine, Farmington, Connecticut, U.S.A.

Many general surgery residents would prefer that all patients who come to an emergency department (ED) require treatment for gunshot wounds to the anterior abdomen. Because the priorities are clear, the algorithm for such treatment is straightforward, the workup required is minimal, the surgical approach necessitates a laparotomy, and interaction with other services is minimal. Alas, the management of trauma is usually not this simple. At most trauma centers, blunt trauma predominates, injuries regularly require an extensive workup, the management algorithm is complex, and clear supportive evidence for treatment choices is not always found in the medical literature. Optimal care in these situations is achieved only when the general surgeon can interact smoothly with a variety of other specialists, particularly neurosurgeons and orthopedic surgeons. We frequently depend on the services of anesthesiologists, cardiac surgeons, and interventional radiologists. Coordinating the actions of our colleagues, prioritizing treatment for various injuries, and selecting appropriate diagnostic tests in the optimal sequence require an understanding of the complex interactions between different types of injuries and the perspectives of our subspecialty colleagues. Prominent areas of controversy include the evaluation and treatment of patients with combined head and abdominal injuries, combined blunt aortic injury and abdominal injury, and the timing of fracture fixation in patients with severe pulmonary or brain injury. For many of these issues, there is no single correct approach. Every trauma service needs to develop a system that delivers optimal patient care in their hospital. One example of such a system involves the timing of patient evaluation by subspecialty consultants. This timing is a matter of institutional preference. In some hospitals, the orthopedic surgeons and neurosurgeons are part of the initial trauma response team, whereas in many other hospitals, these specialists respond only when they are called for. In a large university teaching hospital, the presence of neurosurgery and orthopedic surgery residents makes their inclusion in the initial trauma response team more likely. On the other hand, a community hospital trauma center that does not have surgical subspecialty

residents would be unlikely to demand that an attending neurosurgeon or orthopedic surgeon participate in the evaluation of every trauma patient. However, the fact that institutional practices are different does not necessarily imply that better care is delivered with one arrangement or the other. Because all patients are unique and the evaluation and treatment of the multisystem trauma patient will always be challenging, we must do our best to ensure that the care we provide is individualized. It is important to recognize that some combinations of injuries are likely to prompt disagreement about the best way to manage them. Collaborative discussion about these issues and agreement on a treatment protocol can improve the quality of the care provided.

BRAIN AND ABDOMINAL TRAUMA Each year, roughly 50,000 deaths occur in the United States as a result of traumatic brain injury (1). In some published series, brain injuries have been responsible for 42% of traumatic deaths (2). The combination of brain injury and exsanguination is responsible for another 6% of traumatic deaths (2). The combination of intra-abdominal bleeding and serious brain injury is uncommon but not rare and can present a difficult management problem. This combination of injuries is particularly lethal because bleeding that results in hypotension will render cerebral perfusion inadequate, thereby worsening the brain injury and, ultimately, the chance for recovery. When Chesnut et al. reviewed the Traumatic Coma Databank (TCDB), they defined hypotension as systolic blood pressure (SBP) below 90 mmHg, hypoxia as a partial pressure of arterial oxygen of less than 60 mmHg, and severe brain injury as a Glasgow Coma Scale (GCS) score of less than 8 upon arrival in the ED (3). Patients with hypotension and hypoxia upon arrival had only a 5.8% chance of a good or moderately disabled neurological outcome and 75% of those patients died. Hypotension alone increased the mortality rate from

400

Ivy

27% to 65%. The authors reported that 35% of their patients were hypotensive and 46% were hypoxic during the early period after injury; they also emphasized the impact of hypotension and hypoxia on brain-injured patients. These results are particularly discouraging because they imply that the outcome of many of these patients is fixed soon after their arrival in the ED. General surgeons have interpreted the TCDB study to mean that hypotension in patients with brain injury is to be avoided at all costs. This concept has a sound physiologic basis. An SBP of 90 mmHg roughly corresponds to a mean arterial pressure (MAP) of 60 mmHg. If the patient’s intracranial pressure (ICP) is 20 mmHg or more as a result of the injury, the cerebral perfusion pressure (CPP, defined as MAP minus ICP) is below 40 mmHg; this perfusion pressure is inadequate to meet the needs of an injured brain. All efforts should be made to quickly correct fluid deficits, rapidly diagnose, and stop ongoing bleeding. Rarely, neurosurgeons will argue that the need for emergent computed tomography (CT) of the brain takes priority over the need for laparotomy. This argument does not apply to hypotensive patients. Fortunately, the combination of laparotomy and craniotomy is required for only 0.4% of trauma patients who require CT of the head or a neurosurgical procedure (4). The frequency of this combination on injuries may well be decreasing due to the improvements in automobile safety (e.g., air bags) over the past decade. The clinical presentation of patients with injury to both the brain and the abdomen may vary widely. Therefore, the diagnostic evaluation and treatment must be tailored to each specific patient. Accurate clinical assessment when the patient arrives at the ED is fundamental. Rapid and correct calculation of the GCS score and prompt detection of lateralizing signs are crucial. Lateralizing signs include unilateral dilation of a pupil, hemiplegia, hemiparesis, and asymmetric posturing. The presence of a serious abdominal injury can be suggested by the mechanism of injury. In such cases, clinical clearance of the abdomen is justified only if patients are conscious with an intact sensorium and exhibit no cardiovascular symptoms. Patients with hypotension should immediately undergo diagnostic procedures such as diagnostic peritoneal lavage (DPL) or abdominal ultrasonography (US). Patients with a GCS score of less than 13 should undergo CT of the head; however, the indications for such scanning when the GCS score is 13 or higher are less clear. Harad and Kerstein (5) reviewed the cases of 302 patients with a GCS score of at least 13 and found that the 18% of patients had abnormal results from CT of the head and that 4% required neurosurgical intervention. Stiell et al. (6) found that 8% of 3121 patients with brain injury and a GCS score of at least 13 had a clinically important brain injury and that 1% required neurosurgical intervention. These authors identified five risk factors that predicted the

need for neurosurgical intervention with 100% sensitivity (Table 1). Two other risk factors, along with the five high-risk factors, helped to predict the presence of clinically significant brain injury with a sensitivity of 98.4%. The presence of any one of the risk factors should prompt CT of the head. Hypotensive patients who have suffered blunt trauma and have a minor head injury should undergo abdominal US during their secondary survey. If the results of US are positive for free intraperitoneal blood and the patients remain hypotensive, laparotomy should be performed without CT of the head. If US is unavailable, DPL is indicated. If the results of DPL are positive, laparotomy is indicated. If the results of DPL are negative, the cause of hypotension should be diagnosed and treated. A more complex challenge is presented by the hypotensive patient with a GCS score of less than 13. A review of the medical literature demonstrates that no prospective randomized trials have compared the effectiveness of treatment algorithms and that most studies have not considered the role of US in the examination of patients with brain injury whose condition is unstable. Thomason et al. (7) prospectively studied 14,255 patients from eight trauma centers in North Carolina. Five percent of these patients were alive but hypotensive when they arrived in the ED. Of the hypotensive patients, nearly 10% died in the ED and another 21% required laparotomy. Although 40% of the hypotensive patients were found to have a serious head injury, only 2.5% required emergency craniotomy. Only six hypotensive patients (0.8%) needed both craniotomy and laparotomy. The authors concluded that laparotomy and control of abdominal bleeding are the highest priorities in such patients. Wisner et al. (4) retrospectively reviewed the cases of 800 consecutive trauma patients who underwent CT of the head or a neurosurgical procedure. In this series, 52 patients (6.5%) required craniotomy, 40 (0.5%) required therapeutic laparotomy, and 63 (7.9%) underwent nontherapeutic laparotomy. Only three patients (0.4%) required both craniotomy

Table 1 Indications for Computed Tomography of the Head for Patients with a GCS Score of 13 or Higher Factors indicating high risk of head injury GCS score less than 15 two hours after injury Suspected depressed- or open-skull fracture Any sign of basal skull fracture Two or more episodes of vomiting Age 65 years or more Factors indicating medium risk of head injury Amnesia involving events that took place more than 30 min before the injury Mechanism of injury (pedestrian struck by car, ejection from vehicle, fall from a height of more than 3 ft or 5 stairs) Abbreviation: GCS, Glasgow Coma Scale.

Chapter 32: Competing Priorities in the Trauma Patient

and therapeutic laparotomy. These authors found that only two predictors, intubation in the field (odds ratio, 5.0) and lateralizing findings (odds ratio, 4.0), were significantly associated with the need for craniotomy. Thirty patients in this series exhibited lateralizing signs upon arrival in the emergency room; 10 of these patients required craniotomy. Wisner et al. (4) recommended a straightforward algorithm for the treatment of patients with trauma to the head and the abdomen (Fig. 1). All hypotensive trauma patients whose condition remains unstable but who have no other obvious cause of cardiovascular instability undergo laparotomy. Hypotensive patients whose conditions stabilize with fluid resuscitation but with peritoneal aspirate grossly positive for blood undergo laparotomy if they do not exhibit lateralizing signs; if they do exhibit lateralizing signs, they undergo CT of the head before celiotomy. Patients whose conditions improve with fluid resuscitation and with peritoneal aspirate negative for blood undergo simultaneous CT of the head and completion of DPL. Winchell et al. (8) retrospectively reviewed the cases of 212 hypotensive patients with blunt trauma and a suspected head injury. Although most studies define unstable condition as SBP below 90 mmHg, this study defined instability as SBP below 100 mmHg. Patients whose conditions did not respond to fluid resuscitation underwent immediate surgical procedures; those whose conditions did respond to fluid resuscitation underwent CT of the head before operative treatment was performed. For patients whose conditions were unstable, DPL was the primary method of diagnosing abdominal injury. For patients whose conditions continued to be unstable after

Figure 1 Proposed algorithm using the results of diagnostic peritoneal lavage to guide the management of combined brain and abdominal injuries. Abbreviation: CT, computed tomography. Source: From Ref. 4.

401

fluid resuscitation, the intraoperative mortality rate was 70% and the overall hospital mortality rate was 80%. The authors reported that a low GCS score predicts the need for craniotomy: 19% of patients with a GCS score of less than 8 underwent craniotomy, whereas only 9% of those with a GCS score of 8 to 13 required craniotomy. Although CT scans have revolutionized the management of brain injury, performing them takes time and can delay treatment. Winchell et al. (8) reported that the average time from arrival in the ED to the start of a surgical procedure was 47 minutes for patients in unstable condition and 115 minutes for patients who underwent CT of the head before laparotomy. Because scanners are now faster, the delay should be less than 68 minutes, but much of the delay is unavoidable because of the time required to transport and position patients for the scan. Huang et al. (9) evaluated the usefulness of US in determining whether these patients should undergo immediate laparotomy or should first undergo CT of the head. These authors used a simple scoring system to calculate the amount of fluid demonstrated by initial US. A US score or 3 or higher was correlated with the presence of at least one liter of free intraperitoneal fluid. All 14 patients in this study with a US score of 3 or higher required therapeutic laparotomy, and all of these patients underwent emergent CT of the head after laparotomy. Two patients required intraoperative placement of an ICP monitor because of evidence of serious brain injury. A more recent study by McKenney and coworkers (10) evaluated a new scoring system for abdominal US after trauma and found that the sensitivity of a US score of 3 or higher in predicting the need for therapeutic laparotomy was 83%. The algorithm developed by Huang et al. (9) is representative of current actual practice at many trauma centers (Fig. 2). Rapid US is performed as part of the resuscitation procedure in the ED, and laparotomy is performed for patients with continued hypotension and positive US results. Patients with positive US results, persistent unstable conditions persists, and the need neurosurgical monitoring because of low GCS scores or lateralizing signs, undergo placement of an ICP monitor during abdominal exploration. Those patients are taken directly from the operating room to the CT scanner if their physiologic status permits. At trauma centers where US is not readily available, DPL is used to evaluate abdominal injuries among hypotensive patients. Patients with SBP below 90 mmHg and positive results from paracentesis undergo immediate laparotomy. CT of the head can be performed immediately after the operation if the patient’s condition permits. Patients whose conditions respond rapidly and appropriately to fluid resuscitation should undergo CT of the head and abdomen instead of immediate laparotomy.

402

Ivy

Figure 2 Proposed algorithm using the results of ultrasonography to guide the management of combined brain and abdominal injuries. Abbreviations: US, ultrasonography; CT, computed tomography; ICP, intracranial pressure. Source: From Ref. 9.

INJURIES TO THE THORACIC AORTA AND THE ABDOMEN Controversy persists about the evaluation and treatment of patients with abdominal injuries and potential injuries to the thoracic aorta. Tears of the thoracic aorta are the second leading cause of death after blunt injury. Most of these deaths occur in the field very soon after the event. Patients with partial aortic tears who survive to reach the hospital must receive appropriate diagnostic tests and undergo rapid and safe repair of the aorta. General surgeons must understand the controversies surrounding the diagnosis and management of injuries to the thoracic aorta among patients who have suffered blunt trauma. If the care of these patients is to be properly directed, surgeons must be aware of several issues, including the role of CT in diagnosis and screening, the role of transesophageal echocardiography (TEE) in diagnosis, the optimal timing of abdominal and thoracic surgery, and the roles of delayed repair and nonoperative management. The multicenter American Association for the Surgery of Trauma (AAST) trial (11) reported the status of diagnosis and management of aortic injury in 1997. These researchers enrolled 274 patients from 50 trauma centers over a 30-month period and recorded 86 deaths, for a mortality rate of 31%. Of the injuries, 81% were due to motor vehicle crashes and another 14% were the result of motorcycle crashes or of motor

vehicles striking pedestrians. The mortality rate for hemodynamically stable patients who underwent repair of aortic injury was only 8% (22 patients). Eight of those deaths occurred in the operating room when control of the proximal aorta was lost. Most of the deaths in this study (54 of 86) were due to rupture of the aorta. Other causes of death were head injury (11 deaths), multiple organ failure, or adult respiratory distress syndrome (ARDS; 17 deaths). Because a substantial amount of energy is required to tear the aorta, it is not surprising that other injuries are frequently associated with such tears. Overall, 51% of the patients in the AAST study had a brain injury, 21% had an abdominal injury, and 34% had a pelvic or long-bone fracture (11). The study reported that 93 laparotomies, 124 orthopedic operations, and eight craniotomies were performed. In the AAST study (11), the mean time from injury to diagnosis of aortic injury was 4.7 hours, and the time to repair averaged nearly 10 hours after diagnosis and 16 hours after injury. One potential explanation for this delay in treatment is the inclusion in the study of patients who were managed nonoperatively for an extended period of time because of the presence of other injuries or significant comorbid disease. Reporting the median time rather than the mean time to diagnosis and repair might have been more meaningful. Because most (91%) of the aortic repairs were performed by cardiothoracic surgeons (11), another potential explanation for the prolonged times to diagnosis and repair is a lack of coordination between the trauma service and the cardiac surgery service. Each institution should develop a multidisciplinary protocol for optimizing the diagnosis and management of aortic injury; such protocols should be based on published findings and on the unique resources available at each institution. For example, it can be wasteful and potentially dangerous to spend time performing TEE if the standard practice at the hospital is to perform aortic repair only after angiography. Any patient who has sustained high-energy impact to the chest is at risk of a partial tear of the aorta. Obtaining a routine chest radiograph (CXR) as part of the initial evaluation when the patient arrives in the ED is important in diagnosing aortic injury. The AAST study (11) found that 85% of patients with blunt chest trauma had a widened mediastinum, 8% had another finding suggestive of aortic injury, and only 7% had normal results from CXR. Radiography of the chest may be a useful screening test; however, for some patients with normal CXR results, an aortic injury will be missed without more extensive evaluation. Unfortunately, the exact indications for obtaining other tests after CXR has yielded normal results have not been well defined. The usefulness of CT of the chest in the diagnosis of aortic injury has been evaluated by large prospective studies. Mirvis et al. (12) reported using contrast-enhanced CT of the chest to screen 1104 of 7826 patients admitted for blunt trauma; an aortic injury was found in 25 of these

Chapter 32: Competing Priorities in the Trauma Patient

patients. CT provided direct evidence of aortic injury for all 25 patients; the results of CT were equivocal for three other patients, and angiography was required for these patients. In this series, no aortic injuries were missed by CT of the chest. TEE can accurately diagnose aortic injuries. Chirillo et al. (13) reported that TEE has a sensitivity of 93% and a specificity of 98% in diagnosing such injuries. However, TEE has several potential drawbacks. First, TEE neither allows good visualization of the aortic arch, nor provides visualization of the proximal great vessels after they branch from the aorta (13). Second, like many US techniques, TEE is an operator-dependent tool; some cardiologists and anesthesiologists who are accustomed to evaluating the heart may not evaluate the aorta with the same sensitivity as that reported in the literature. On the other hand, when TEE is performed by experienced physicians, it may reveal small intimal tears and thrombi that are not clinically significant and require no further intervention (14). The primary advantage of TEE is that it can save time compared to other diagnostic modalities such as angiogram. In the series by Chirillo et al. (13), the procedure saved an average of more than 40 minutes. Such time-savings can be particularly important for the cardiovascularly unstable patient who requires emergent laparotomy. In such cases, TEE can diagnose aortic injury in the operating room and can minimize treatment delays. However, realizing this potential benefit requires the availability of experienced personnel around the clock. One of the most important controversies related to the treatment of patients with combined abdominal and aortic injuries concerns the timing of surgical procedures. The AAST (11) study found that 66% of patients who needed both laparotomy and aortic repair underwent the abdominal procedure first. That study also demonstrated that patients who are hypotensive because of free rupture of the aorta die before repair can be effected. Another study (15) found that, although as many as 31% of patients with aortic injuries were hypotensive after arrival, as few as 24% of these patients were hypotensive because of the aortic injury. Intra-abdominal injury was a more common cause of cardiovascular instability than was aortic injury. Consequently, the alternative of delaying a therapeutic laparotomy for a hypotensive patient until a thoracotomy and full heparinization can be performed seems unwise. The general recommendation is that laparotomy should be performed first with the goal of preventing exsanguination from abdominal injuries. If there is a sense of urgency about the aortic laceration, it is reasonable to perform a ‘‘damage control’’ laparotomy first so that ongoing bleeding and enteric spillage can be stopped. Controversy exists about the nonoperative management and the delayed operative management of tears of the thoracic aorta. First because of the increased sensitivity of diagnostic tests, we are diagnosing isolated intimal injuries or small intraluminal

403

thrombi that may not require repair (14). Second, there is a growing recognition that surgery can be delayed for some patients with other life-threatening injuries until they are able to tolerate thoracotomy or until they no longer require operative treatment. Karmy-Jones et al. (15) reported the results of nonoperative treatment of patients with severe pulmonary injury, symptomatic coronary artery disease, severe brain injury, and comorbid chronic diseases. They noted that three of the patients who received this type of treatment died after free rupture of the aorta. Pate et al. (14) reported the outcomes of 11 patients who underwent nonoperative treatment and 15 patients who underwent delayed treatment. These authors used beta-blockade to maintain a heart rate of less than 90 beats per minute and added vasodilators as necessary to maintain an SBP below 100 mmHg. No patients died of free rupture of the aorta. The authors selected nonoperative therapy for patients with severe head, pulmonary, or abdominal injuries and for patients with preexisting medical conditions such as renal failure or symptomatic coronary artery disease. Follow-up studies showed that patients with small intimal flaps or thrombi had no aortographic evidence of aortic abnormality. It appears that in select circumstances, nonoperative management or delayed repair of the thoracic aorta is acceptable provided that the patient’s blood pressure and heart rate are strictly controlled.

TIMING OF FRACTURE FIXATION FOR PATIENTS WITH BRAIN INJURIES Controversy exists about the optimal timing of fracture fixation for patients with serious brain injury. This scenario involves multiple surgical specialists, including general surgeons, orthopedic surgeons, neurosurgeons, and anesthesiologists. There are essentially two perspectives. Some surgeons strongly believe that fixation within 48 hours of injury reduces the incidence of pulmonary complications, decreases the length of stay, and improves the eventual outcome of rehabilitation. Other surgeons believe that early fixation may worsen brain edema and increase mortality rates and that the benefits related to early mobilization may be lost for comatose patients with head injuries. Several articles pertinent to this controversy have been published, but none of them reports the results of a large, prospective, randomized controlled trial that can help us establish a definitive standard of care. An important problem with all of the studies that have addressed this complex issue is the lack of statistical power because of small sample size. Fakhry et al. (16) reported the outcome of 87 patients with a brain injury and an Abbreviated Injury Scale (AIS) score of 3 or higher who underwent femur fixation. Early fixation was defined as fixation within 48 hours of arrival in the ED. The mortality rate was 8.5% for patients in the early fixation group and

404

Ivy

3.6% for patients in the delayed fixation group; this difference was not statistically significant. Reynolds et al. (17) prospectively studied 105 patients with brain injury who underwent fixation of femoral fractures. The mean AIS score was 1.96 in the early fixation group and 2.36 in the delayed fixation group; the difference was not statistically significant. The mortality rate was 4.2% for the early fixation group; no deaths occurred in the late fixation group, but this difference was not statistically significant. The number of days spent in the intensive care unit (ICU) and the number of days spent on a ventilator were also nearly identical between groups. Jaicks et al. at Yale (18) retrospectively reviewed the cases of 33 patients with long-bone fractures and brain injuries. The mean AIS score was 3.3 for patients in the early fixation group and 3.1 for patients in the delayed fixation group. Although 53% of the patients in the early fixation group and only 36% of the patients in the late fixation group had femur fractures, this difference was not statistically significant. The authors reported a mortality rate of 11% in the early fixation group and no deaths occurred in the late fixation group, but this difference was also not statistically significant. The mean number of days spent on a ventilator was 6.4 for patients in the early fixation group and 6.5 for patients in the late fixation group. The most interesting finding of the study was that 16% of the patients in the early fixation group and 7% of the patients in the late fixation group exhibited intraoperative hypotension, defined as SBP below 90 mmHg, and that 11% of the patients in the early fixation group but only 7% of the patients in the late fixation group experienced hypoxia, defined as partial pressure of oxygen below 60 mmHg. Neither of these differences were statistically significant. Hofman and Goris (19) retrospectively reviewed the cases of 58 patients with severe brain injury and long-bone fractures. The mean GCS score was 4.6 for patients who underwent early fixation and 4.7 for patients who underwent late fixation. These authors reported a statistically significant difference in mortality rates: 13% for patients in the early fixation group and 47% for patients in the late fixation group (p < 0.02). On the other hand, Sanker et al. (20) reviewed the cases of patients with a GCS score between 4 and 8; they reported mortality rates of 59% for the early fixation group and 12% for the delayed fixation group (p < 0.01). Both of these studies are Level III studies and are not definitive. The Eastern Association for the Surgery of Trauma (EAST) Practice Management Guidelines Work Group has evaluated all the evidence and has posted its conclusions on their web site (www.east. org). There is no Level I evidence to guide our management practices (21). Because the Level II and Level III evidence are flawed, no strong recommendations can be based on those studies. Scalea et al. (22) developed a ‘‘damage control’’ approach to the management of long-bone fractures

for patients with severe multisystem trauma. These authors retrospectively compared 43 severely injured patients who underwent early external fixation of the femur with 284 patients who were treated with placement of primary intramedullary rods. The patients who underwent external fixation were more severely injured with a mean ISS of 26.8 versus 16.8 (p ¼ 0.001). In the external fixation group, 46% of the patients had sustained a head injury with an overall mean GCS of 11 versus 14.2 (p ¼ 0.001). Over onequarter of the external fixation group required ICP monitoring, they had a median opening ICP of 22 mmHg and a median increase up to 27 mmHg. The external fixation procedures were performed either at the bedside or in the operating room and required an average of 35 minutes with an average estimated blood loss of 90 mL. In the cohort managed with external fixation, 9% of the patients died, 65% were discharged to a rehabilitation facility, and 26% were discharged to home a median hospital length of stay of 17.5 days. Less than 1% of the group treated with a primary intramedullary nail died, 51% were discharged to home and 49% were discharged to a rehabilitation facility. Unfortunately, given the variation in severity of injury, it is difficult to determine the effectiveness of the ‘‘damage control’’ approach in this patient population from this study. This innovative procedure may benefit some severely injured patients, but we will need controlled studies before we can state this with confidence.

MANAGEMENT OF LONG-BONE FRACTURES AND PULMONARY INJURIES Another important controversy surrounds the treatment of patients with fractures and pulmonary injury. Advocates of early fracture fixation believe that outcomes are better and hospital stays are shorter when fracture fixation occurs during the initial 48 hours after injury. Other surgeons argue that early fracture fixation increases the risk of complications and exacerbates the pulmonary disease process. Fakhry et al. (16) studied 96 patients with a chest injury and an AIS score of at least 3 who underwent delayed or early fixation of femur fractures (16). The mortality rate for patients in the early fixation group was 5%; no deaths occurred in the late fixation group, but the difference was not statistically significant. Reynolds et al. reviewed the impact of chest injury on outcome (17). They reported that the mortality rate was higher for the early fixation group (4.2%) than for the late fixation group (no deaths), but the difference was not statistically significant. Charash et al. (23) performed a nonrandomized, prospective study of 82 patients who underwent fixation of femur fractures. There was no statistically significant difference between the two groups in mortality rates, but pulmonary complications were more likely to occur among patients in the delayed fixation group.

Chapter 32: Competing Priorities in the Trauma Patient

However, it is not clear whether this difference in pulmonary complications is due to differences in injury severity or in patient management. This question arises because 71% of patients in the delayed fixation group were intubated during their initial resuscitation whereas only 24% (p ¼ 0.01) of patients of the early fixation group were intubated initially. Thus it is very likely that the delayed fixation patients had sustained more severe pulmonary injuries overall. The Level II evidence does not demonstrate that a convincing difference in the incidence of ARDS or pneumonia, or the number of days a patient spends on a ventilator or in the ICU, can be clearly attributed to fracture management. The findings of Level III studies are contradictory. Boulanger et al. (24) demonstrated that a delay in fixation was associated with a statistically insignificant increase in the incidence of ARDS (from 4% to 20%; p > 0.05). Pape et al. (25) reported that the incidence of ARDS was 33% after early fixation and 8% after late fixation (p ¼ 0.03). In light of this contradictory evidence, strong recommendations about the optimal management of femur fractures cannot be made. The EAST Management Guidelines Work Group has done a superb job of summarizing the evidence obtained to date (21). The Group does not endorse either early or late fixation of fractures for patients with clinically significant pulmonary injuries. It remains to be seen whether the ‘‘damage control’’ approach proposed by Scalea et al. (22) will improve the outcomes of these patients. In the end, we surgeons need to provide each patient with individualized treatment based on the patient’s condition.

REFERENCES 1. Sosin DM, Sniezek JE, Waxweiler RJ. Trends in death associated with traumatic brain injury, 1979 through 1992. Success and failure. JAMA 1995; 273:1778–1780. 2. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma deaths: a reassessment. J Trauma 1995; 38: 185–193. 3. Chesnut RM, Marshall LF, Klauber MR, et al. The role of secondary brain injury in determining outcome from severe head injury. J Trauma 1993; 34:216–222. 4. Wisner DH, Victor NS, Holcroft JW. Priorities in the management of multiple trauma: intracranial versus intra-abdominal injury. J Trauma 1993; 35:271–278. 5. Harad FT, Kerstein MD. Inadequacy of bedside clinical indicators in identifying significant intracranial injury in trauma patients. J Trauma 1992; 32:359–363. 6. Stiell IG, Wells GA, Vandemheen K, et al. The Canadian CT Head Rule for patients with minor head injury. Lancet 2001; 357:1391–1396. 7. Thomason M, Messick J, Rutledge R, et al. Head CT scanning versus urgent exploration in the hypotensive blunt trauma patient. J Trauma 1993; 34:40–45. 8. Winchell RJ, Hoyt DB, Simons RK. Use of computed tomography of the head in the hypotensive blunttrauma patient. Ann Emer Med 1995; 25:737–742.

405

9. Huang MS, Shih HC, Wu JK, et al. Urgent laparotomy versus emergency craniotomy for multiple trauma with head injury patients. J Trauma 1995; 38:154–157. 10. McKenney KL, McKenney MG, Cohn SM, et al. Hemoperitoneum score helps determine need for therapeutic laparotomy. J Trauma 2001; 50:650–656. 11. Fabian TC, Richardson JD, Croce MA, et al. Prospective study of blunt aortic injury: Multicenter Trial of the American Association for the Surgery of Trauma. J Trauma 1997; 42:374–383. 12. Mirvis SE, Shanmuganathan K, Buell J, Rodriguez A. Use of spiral computed tomography for the assessment of blunt trauma patients with potential aortic injury. J Trauma 1998; 45:922–930. 13. Chirillo F, Totis O, Cavarzerani A, et al. Usefulness of transthoracic and transesophageal echocardiography in recognition and management of cardiovascular injuries after blunt chest trauma. Heart 1996; 75:301–306. 14. Pate JW, Govant ML, Weiman DS, Fabian TC. Traumatic rupture of the aortic isthmus: program of selective management. World J Surg 1999; 23:59–63. 15. Karmy-Jones R, Carter YM, Nathens A, et al. Impact of presenting physiology and associated injuries on outcome following traumatic rupture of the thoracic aorta. Am Surg 2001; 67:61–66. 16. Fakhry SM, Rutledge R, Dahners LE, Kessler D. Incidence, management, and outcome of femoral shaft fracture: a statewide population-based analysis of 2805 adult patients in a rural state. J Trauma 1994; 37:255–261. 17. Reynolds MA, Richardson JD, Spain DA, Seligson D, Wilson MA, Miller FB. Is the timing of fracture fixation important for the patient with multiple trauma? Ann Surg 1995; 222:470–481. 18. Jaicks RR, Cohn SM, Moller BA. Early fracture fixation may be deleterious after head injury. J Trauma 1997; 42:1–6. 19. Hofman PA, Goris RJ. Timing of osteosynthesis of major fractures in patients with severe brain injury. J Trauma 1991; 31:261–263. 20. Sanker P, Frowein RA, Richard KE. Multiple injuries: coma and fractures of the extremities. Neurosurg Rev 1989; 12(suppl 1):51–54. 21. Dunham CM, Bosse MJ, Clancy TV, et al and EAST Practice Management Guidelines Work Group. Practice management guidelines for the optimal timing of longbone fracture stabilization in polytrauma patients: the EAST Practice Management Guidelines Work Group. J Trauma 2001; 50:958–967. 22. Scalea TM, Boswell SA, Scott JD, Mitchell KA, Kramer ME, Pollak AN. External fixation as a bridge to intramedullary nailing for patients with multiple injuries and with femur fractures: damage control orthopedics. J Trauma 2000; 48:613–623. 23. Charash WE, Fabian TC, Croce MA. Delayed surgical fixation of femur fractures is a risk factor for pulmonary failure independent of thoracic trauma. J Trauma 1994; 37:667–672. 24. Boulanger BR, Stephen D, Brenneman FD. Thoracic trauma and early intramedullary nailing of femur fractures: are we doing harm? J Trauma 1997; 43:24–28. 25. Pape HC, Auf’m’Kolk M, Paffrath T, Regel G, Sturm JA, Tscherne H. Primary intramedullary femur fixation in multiple trauma patients with associated lung contusion—a cause of posttraumatic ARDS? J Trauma 1993; 34:540–548.

33 Complications of Fractures Bar Ziv Yaron, Kosashvili Yona, Gelfer Yael, and Halperin Nahum Department of Orthopedic Surgery, Tel Aviv University, Sackler Faculty of Medicine, Assaf Harofeh Medical Center, Zeriffin, Israel

A fracture is a discontinuity of the bone, which is caused by the application of an external force. The amount of energy absorbed by the bone to create the fracture depends on the magnitude of the force and the direction of the load (1). The force required to break a bone depends on the material’s properties and the bone’s geometry. Fractures may be classified in a number of ways: by the amount of energy absorbed by the bone (high energy or low energy), by the direction of the force (direct trauma or indirect trauma), or by the extent of the fracture (complete fracture or incomplete greenstick fracture in children). The amount of energy involved in creating a fracture is the most important factor because high-energy trauma causes an open-comminuted fracture with damage to soft tissues, including neurovascular structures. The severity of the soft-tissue injury affects the bone-healing process and the complications associated with the fracture. The mechanism of injury determines whether the trauma is direct or indirect. In the case of direct trauma, force is applied directly to the fracture site and this direct application causes tapping fractures, crush fractures, and penetrating fractures (2). Tapping fractures typically affect the transverse line in the bone, whereas crush and penetrating injuries results in open-comminuted fractures. A typical result of direct trauma is a fracture of the tibial plateau that results when the tibia is struck directly by an automobile bumper. In the case of an indirect trauma, a force is applied to an area at a distance from the fracture site. A typical result of indirect trauma is a fracture of the upper extremity. For example, falling on an outstretched hand may generate a fracture of the navicular bone, the distal radius, the head of the radius, the humerus, or the clavicle, depending on the point of loading. Several fracture patterns are produced by indirect force: tension fractures, angulation fractures, rotation fractures, and compression fractures. Fractures are usually produced by a combination of forces rather than by a single force acting alone (3). A typical tension fracture is created when the knee is forcibly flexed while the extensor mechanism is in contraction. Similarly, the deltoid ligament in eversion and external rotation may pull off the medial malleolus. Compression forces may cause

impacted fractures such as fractures of the spinal vertebrae. Angulation fractures, rotation fractures, or combinations of indirect forces typically cause spiral and oblique lines; the pattern affects the stability of the fracture. A pathological fracture occurs when a degree of stress that would leave normal bone intact is applied to a bone weakened by factors such as an underlying tumor, infection, or metabolic disease (4). The most common pathological fractures involve osteoporotic bone in the proximal femur (5), the spine, the proximal humerus, or the distal radius. A stress fracture results from fatigue brought about by repeated loading. These fractures are seen most frequently among military recruits undergoing strenuous training (6) and among ballet dancers and athletes. Muscle fatigue allows an abnormal concentration of stress, which results in failure of the bone (7). The incidence of complications may be reduced by careful planning of all treatment aspects, including primary assessment, emergency care, definitive treatment, and rehabilitation. The complications that may occur after fracture can be acute or chronic. Acute complications are related to injuries to soft tissues, such as vascular, pulmonary, gastrointestinal, and neurological structures. These complications may jeopardize life or limb, depending on the severity of the injury and the systemic response. Long-term complications involve abnormalities of bone healing (malunion, delayed union, and nonunion), posttraumatic osteoarthritis, and avascular necrosis (AVN). The amount of energy and the intensity of the force are responsible for the incidence of early complications after fracture. An unstable fracture of the pelvic ring is very likely to lead to severe systemic complications such as hemorrhagic shock, neurovascular injuries, thromboembolism, and pulmonary embolism (PE). Long-bone fractures resulting from high-energy trauma are often associated with a risk of fat embolism (FE) syndrome and compartment syndrome. Immobility of the limb after a fracture is a risk factor for thromboembolic disease and PE. The risk of complications is substantially increased after a high-energy open fracture. Stabilizing the fracture is the main goal of emergency care and will reduce the likelihood of early complications. Restoring the bones to a position as close as possible to the

408

Yaron et al.

normal anatomical position, by surgical or nonsurgical treatment, is necessary for realigning neurovascular structures, providing optimal circulation to the injured extremity, and minimizing the risk of peripheral nerve compromise. Reestablishing bony length decreases the size of hematomas and the extent of soft-tissue disruption. It also improves venous and lymphatic return and reduces soft-tissue swelling. When soft tissue and neurovascular structures around broken bones are undamaged, later complications such as delayed union, nonunion, malunion, AVN, and heterotopic ossification are less likely.

ACUTE COMPLICATIONS Shock Shock is a clinical condition manifested by poor tissue perfusion with resultant tissue hypoxia, which may damage vital organs. Hemorrhagic (hypovolemic) shock is by far the most common type of shock affecting patients with multiple skeletal injuries. The bone injuries that most commonly produce severe hypovolemic shock are fractures of the pelvis. Ostrum et al. reported that, of 100 patients with closed, isolated femoral shaft fractures, none experienced class III or IV shock (8). Among these patients, bleeding from the femur fracture was insufficient to produce hypotension. Thus, a meticulous search for a second source of blood loss must be performed when a patient with a closed femoral shaft fracture is hypotensive. More than 90% of pelvic ring injuries are associated with other injuries (9). Although pelvic ring injuries are notorious for hemorrhage, this bleeding is the primary cause of death for only 7% to 18% of fatalities (10). Burgess et al. have shown that hemorrhage from pelvic fractures strongly depends on the type of fracture; fractures associated with complete dissociation of the posterior pelvis (anterior– posterior type III) are associated with the greatest loss of blood, typically more than 20 units of blood in the first 24 hours, and consequently the highest mortality rates related to shock (11). The goal of treating hypovolemic shock is to safely restore adequate intravascular volume and oxygen-carrying capacity.

Hemorrhage and Pelvic Fractures Approximately 90% of bleeding associated with fractures of the pelvis arises from low venous plexus pressure and fractured cancellous bone surfaces (12–14). The remaining 10% of bleeding originates from named arteries that can be embolized (15). The initial treatment of the fracture is stabilization of the pelvis. Stabilizing the pelvis prevents constant movement during transport and resuscitation (16). It is also believed that decreasing the volume of the pelvis will restore the tamponade effect and lower the volume of bleeding into the retroperitoneum (13,16,17). This stabilization can be achieved by applying an external

fixator, which can be used temporarily as an emergency stabilizer or definitively for an open-book fracture. The fixator is applied by inserting pins into the ilium and is held in place with transverse bars that stabilize the pelvis. The patient’s condition is considered hemodynamically unstable if the systolic blood pressure upon arrival at the emergency department is below 90 mmHg and remains low after the administration of 2 L of intravenous (IV) crystalloids, or if the pulse rate is greater than 110 beats per minute. Patients whose condition is hemodynamically unstable should undergo suprapubic diagnostic peritoneal lavage (DPL). Negative results from DPL indicate that the bleeding is of retroperitoneal origin and suggest that an external fixator should be applied, if this action has not already been taken (12,13). If the external fixator does not stabilize the patient’s hemodynamic condition, the surgeon should be alerted to the possibility that the source of bleeding is arterial and should consider embolization under angiographic guidance (18,19). If the patient’s hemodynamic condition is deteriorating, no time should be wasted in transfer to the angiography unit; exploratory laparotomy and retroperitoneal packing (with or without angiographic guidance) should be performed in the operating room (18).

Pearls and Pitfalls &

&

High grade of suspicion in fracture patterns notorious for hypovolemic shock. Ongoing monitoring of hemodynamic status is of outmost importance.

Vascular Injury The following arterial injuries are commonly associated with fractures. Injury to the brachial artery is associated with proximal fractures of the humeral shaft or with supracondylar fractures of the humerus. Injuries to the deep femoral artery are associated with subtrochanteric fractures of the femur. Injuries to the popliteal artery below the level of Hunter’s canal are associated with supracondylar fractures of the femur, whereas injuries to the popliteal artery at the level of trifurcation are associated with metaphyseal fractures of the tibia. Injuries to the anterior tibial artery are associated with fractures to the middle-third of the tibia. Vascular injuries should also be considered in fracture dislocations, especially of the knee. The rate of arterial injuries ranges between 5% and 30% in anterior and posterior dislocations where the insult results from tethering of the popliteal artery. Subsequently, the likelihood of arterial injuries must be considered with every fracture or joint dislocation. Injury to a major artery of the extremity may result in histologic changes to nerves and muscles if blood supply is not restored within four to six hours (20). ‘‘Hard’’ signs of vascular injury include massive external bleeding, a rapidly expanding hematoma, and the

Chapter 33:

classic signs of arterial occlusion: lack of pulse, pallor, aesthesia, pain, paralysis, and poikilothermia. When these signs are present, immediate surgical exploration and vascular repair are indicated, particularly if there is substantial external bleeding or the threat of limb loss (21). ‘‘Soft’’ signs of vascular injury include a questionable history of arterial bleeding, proximity of a penetrating wound to an artery, a small nonpulsatile hematoma, or a neurologic deficit. Patients with these signs can be treated with arteriography or observation alone (21). Patients with pulse deficits distal to fractures or dislocation should undergo immediate reduction before any other treatment is undertaken. The presence of normal distal pulses does not rule out a serious vascular injury, and the limb must be repeatedly examined so that an occult lesion is not missed. Lynch and Johansen reported on 100 patients with blunt or penetrating limb trauma; all patients had ankle-brachial indexes (ABIs) measured and were studied by arteriography. ABI less than 0.90 predicted an injury that required intervention with 87% sensitivity and 97% specificity (22). The treatment of vascular injury requires a complementary team effort by vascular and orthopedic surgeons. The aim of surgery is to shorten the ischemic period as much as possible by promptly restoring the arterial flow to the limb. On the other hand, performing the vascular procedure first puts the repair at risk during the orthopedic manipulation. Priorities in such situations should be jointly established by the vascular and orthopedic surgeons after a complete discussion of the nature of the combined injuries and the duration of the planned repair of both the arterial and the orthopedic injuries (23).

Pearls and Pitfalls &

&

&

&

Fractures of long bones should be stabilized before vascular repair is undertaken. If the extremity may be jeopardized by a delay in arterial repair, the vascular injury should be repaired first. Fasciotomy is indicated if any of the following has occurred:  restoration of blood flow has been delayed  the patient has experienced an episode of substantial hypotension  considerable swelling is noted Specific fractures or dislocations go along with specific neurovascular injuries.

Nerve Injury Nerve injuries result from direct contusion or laceration by the fracture fragment or by penetrating missiles. Nerves may also be stretched by excessive forces that produce fractures or dislocation about the joint (see Chapters 33, 34). The following nerve injuries are commonly associated with fractures or dislocations: injuries to the axillary nerve are associated with fracture or dislocation of the shoulder; injuries to the radial nerve are commonly associated with fracture

Complications of Fractures

409

of the humeral shaft; injuries to the median nerve at the wrist are commonly associated with displaced fractures of the radius; injuries to the sciatic nerve are commonly associated with fracture or dislocation of the hip; and injuries to the peroneal nerve are commonly associated with fracture of the neck of the fibula.

Fat Embolism Syndrome FE syndrome is a complex clinical syndrome, consisting of fever, tachycardia, and confusion in association with arterial hypoxemia and other pertinent laboratory findings. Modern techniques for managing respiratory distress have decreased the mortality and morbidity rates associated with this syndrome. Its incidence correlates positively with the number of long bones fractured, with the presence of open fractures and with fractures caused by vehicular accidents (24,25). The pathogenesis of the FE syndrome is a subject of controversy. Most investigators agree that bone marrow is the source of embolic fat seen in the lungs (26–31). Considerably fewer investigators agree about the exact role of this fat in the production of the clinical FE syndrome (32–39). Peltier’s original hypothesis states that lipase, endogenous to the lung, converts neutral fat to toxic free-fatty acids (40–42). Barie et al. have demonstrated that free-fatty acids are bound rapidly by albumin and transported through the bloodstream and the lymphatic channels in this benign form (32). An abundance of tissue thromboplastin is released with the marrow elements after long-bone fracture. This release activates the complement system and the extrinsic coagulation cascade (35,39) and results in the production of by-products of intravascular coagulation, such as fibrin and fibrin-degradation products. These blood elements, along with leukocytes, platelets, and fat globules, combine to increase pulmonary vascular permeability, both by their direct actions on the endothelial lining and through the release of numerous vasoactive substances (43,44). Suppression of the fibrinolytic system in the injured patient may then aggravate an ongoing accumulation of cellular aggregates, fat macroglobules, and clotting factors that are concentrated in the lung by virtue of its filtering action on venous blood before that blood is recycled to the systemic circulation (43). It has become increasingly apparent that embolic marrow fat and other elements may be only the catalyst for a single early step in a long chain of events that lead to the final common pathway of increased pulmonary vascular permeability in response to many forms of systemic injury.

Clinical Findings The onset of the clinical symptoms of FE syndrome may be immediate or may not occur for two or three days after the trauma (45). Diagnosing this syndrome is largely a process of exclusion and depends on the clinician’s index of suspicion. Symptoms are

410

Yaron et al.

shortness of breath followed by restlessness and a changing neurologic picture and disorientation, which are followed by marked confusion, stupor, or coma. Arterial hypoxemia is the hallmark of the syndrome. PE should always be excluded by the results of a helical computed tomography scan of the chest before the diagnosis of FE syndrome is considered. Other clinical signs are temperature elevation to 39 C or 40 C; tachypnea, with rates of 30 breaths per minute or higher; and tachycardia, with rates of 140 beats per minute or higher. On the second or third day after injury, petechiae may be seen, characteristically located on the chest, the axilla, the root of the neck, and the conjunctivae. These clinical manifestations result from reduced blood flow to these vital organs. Other clinical signs are blindness and focal neurologic findings, including seizures. No pathognomonic laboratory test exists for FE syndrome, but arterial hypoxemia is the hallmark of this condition. The measurement of arterial hypoxemia is a sensitive index of the degree of pulmonary FE and monitors the response to treatment. In the early stages, thrombocytopenia may occur with platelet values of less than 150  103/mL. The hematocrit often decreases (39). Chest radiographs demonstrate progressive snowstorm-like pulmonary infiltrations. The changes apparent on chest radiographs are characteristic but not specific (38). A cryostat-frozen specimen of clotted blood, which reveals the presence of fat, has been used in identifying FE syndrome; a biopsy of a skin petechial lesion can also demonstrate the presence of embolic intravascular fat (44).

Treatment The initial (and perhaps the only specific) treatment for FE is aimed at decreasing hypoxemia. Oxygen should be administered immediately upon admission to the emergency department. Accurate monitoring of blood gases is crucial. If the degree of hypoxemia is relatively mild, oxygen can be given by mask; if the degree of hypoxemia is severe and respiratory failure is impending, prompt mechanical ventilatory assistance is mandatory. A few specific drugs have been suggested as treatment for fat emboli— including ethanol, heparin, and hypertonic glucose— but these drugs are not effective in reducing the incidence of pulmonary failure. Recent clinical research suggests that early administration of corticosteroids may be helpful in treating FE syndrome (44). Another intervention that has been successful is angiographically guided occlusion of a patent foramen ovale in a patient with persistent neurologic sequelae resulting from FE syndrome. Finally, early fracture fixation and patient mobilization can decrease the incidence of FE syndrome and improve pulmonary function (34,46–59). The prognosis for recovery from FE syndrome is poor for patients who have experienced marked pulmonary failure and coma. The mortality rate associated with these complications is high. Mild cases of this syndrome often go undetected, and mortality

rates are low for patients who do not experience severe pulmonary insufficiency or cerebral manifestations.

Pearls and Pitfalls &

When encountered with respiratory or neurological distress, exclude PE and do not forget FE.

Compartment Syndrome Acute compartment syndrome is a well-recognized complication of various conditions but most commonly of fracture. Compartment syndrome is defined as increased pressure within a fascial space, which compromises oxygen supply to the muscles and nerves within that compartment. The numerous causes of compartment syndrome include complications of open and closed fractures, arterial injury, temporary vascular occlusion, snakebite, drug abuse, burns, physical exertion, and gunshot wounds. Other possible causes are pulsatile lavage and contusion in patients with hemophilia (60).

Pathogenesis The most common cause of compartment syndrome is muscle injury that leads to muscle edema, which is usually related to the amount of tissue damage. Pressure is increased within a closed space, first by intracellular swelling and then by the formation of a hematoma from the fractured bone. Because the extremities are composed of relatively unyielding fascial compartments, circulatory compromise ultimately occurs as tissue pressure rises, and this condition causes ischemia and tissue damage that result in leakage of intracellular fluid and a further increase in intracompartmental pressure. In the case of arterial injury, the muscle is deprived of its blood supply, and intracellular injury results. When circulation is reestablished, reperfusion injury occurs as the muscle swells, with secondary elevation of tissue pressure and further ischemic damage (60). When injuries produce complete ischemia, skeletal muscle that is deprived of oxygen may survive for as long as four hours without irreversible damage (60,61). Total ischemia of eight hours duration produces a complete irreversible change. Peripheral nerves conduct impulses for one hour after the onset of total ischemia and can survive for four hours with only neurapraxic damage. After eight hours, axonotmesis and irreversible damage occur (62). Ischemia caused by reduction or cessation of blood flow to the muscle results when the perfusion gradient in the compartment tissue falls below a critical level. Thus, perfusion is directly related to the patient’s blood pressure. Experimentally measured terminal arterial pressure is equal to diastolic pressure, which is therefore used as the crucial measurement (60,61). When the intracompartmental pressure is 20 mmHg below the diastolic pressure, tissue perfusion in injured tissues is substantially decreased. Fasciotomy should be performed when the intracompartmental pressure approaches 20 mmHg

Chapter 33:

below the diastolic pressure or if an extremity has been completely ischemic for six hours or if the patient’s clinical condition is worsening or if substantial tissue injury is present or if the tissue pressure is increasing (63). Prophylactic treatment is important. Fasciotomy will not reverse the changes caused by the initial trauma but can prevent changes that result from secondary ischemia.

Clinical Evaluation Pain, pallor, paralysis, paresthesia, and pulselessness are the clinical hallmarks of compartment syndrome. It is important to note that these are the signs and symptoms of an established compartment syndrome with ischemic injury; fasciotomy at this stage yields poor results (60). Pain and aggravation of pain by passive stretching of the muscles in the compartment in doubt are the most sensitive and generally the only clinical findings (60). Because pain is a subjective symptom, it is diagnostically useful only when patients are conscious and can respond cognitively to examination. In unconscious patients at risk of compartment syndrome, tissue-pressure measurements may be the only objective criteria for diagnosis. Lately, several studies have investigated the use of near infrared spectroscopy in the diagnosis of compartment syndrome. Although their results seem promising, clinical data is still required (64–66).

Complications of Fractures

411

Tissue-Pressure Measurements The location of the highest tissue pressure can be determined by taking measurements in all of the extremity’s compartments, at the level of the fracture and proximal and distal to it (63). The highest pressure noted should serve as a basis for determining the need for fasciotomy. When tissue pressure is increasing, careful follow-up is required with repeated physical examination and pressure measurements every one to two hours until fasciotomy is indicated or pressure has decreased to a safe level and clinical signs and symptoms have improved. The newer techniques of compartment monitoring allow the introduction of catheters coupled to transducers, which allow continuous monitoring of one or more compartments. However, obtaining repeated measurements at multiple areas in the same compartment is difficult with an indwelling device. Therefore, needle methods are more appropriate for multiple sites and repeated measurements. A number of methods of tissue-pressure measurement have been described. If properly used, all of them are accurate and equally measure the same phenomenon (60). The infusion technique is simple and inexpensive; it requires only mercury, two plastic IV extension tubes, two 18-gauge needles, one 20-mL syringe, one three-way stopcock, and normal saline (Fig. 1).

Figure 1 Infusion techniques.

412

Yaron et al.

Pearls and Pitfalls &

Look for the 5 P’s—pain, pallor, paralysis, paresthesia, and pulselessness

Open Fractures With an open fracture, a break in the skin and the underlying soft tissues leads directly into or communicates with the fracture. The serious consequences of such an injury are contamination and crushing, stripping, and devascularization of soft tissues and the bone they cover (67). A minor injury, managed properly, arouses no great concern; a major injury may necessitate immediate or early amputation (68). The prognosis of an open fracture is determined primarily by the amount of devitalized soft tissue caused by the injury and by the level and type of bacterial contamination. These two factors working in combination, rather than the configuration of the fracture itself, are the primary determinants of outcome (67). The extent of soft-tissue devitalization is defined by the energy absorbed by the limb at the time of injury. The most important and ultimate goal in the treatment of open fractures is to restore the function of the limb and of the patient as early and as fully as possible. To achieve this goal, the surgeon must prevent infection, restore soft tissues, achieve bone union, avoid malunion, and institute early joint motion and muscle rehabilitation. Of these goals, the most important is avoiding infection because infection is the event that most commonly leads to malunion, nonunion, and loss of function. Prehospital care is crucial to the outcome of open fractures. The fractures should be aligned and splinted, and sterile dressings should be applied to the wound. In the hospital, all open fractures must be treated as an emergency. Surgery should be initiated as soon as the patient’s general condition will allow it. A thorough initial evaluation should be carried out to diagnose other life-threatening injuries, and appropriate antibiotic therapy should be started in the emergency room or, at the latest, in the operating room before the procedure (69,70). The wound should be immediately debrided and the fracture should be stabilized. The wound is then left open for five to seven days.

Classification of open fractures is important because it provides guidelines for prognosis and suggests some methods of treatment. The wound classification system of Gustilo and Anderson (71,72) with subsequent modifications by Gustilo et al. (72–74) is the most widely accepted and quoted system. The crucial factors in this classification system are the degree of soft-tissue injury and the degree of contamination. Table 1 given below is a clarification of that classification (75). An open fracture should generally be stabilized by using the method that provides adequate stability with a minimum of further damage to the vascularity of the zone of injury and its associated soft tissues (76). For type I wounds, any technique that is suitable for closed fracture management is satisfactory. Treatment of type II and type III wounds is more controversial; investigators have suggested traction, external fixation, nonreamed intramedullary nailing, and, occasionally, plate and screw fixation. Generally, external fixation is preferred for metaphyseal– diaphyseal fractures, with occasional limited internal fixation with screws. For the upper extremity, casting, external fixation, and plate-and-screw fixation are the more popular methods of stabilization. For the lower extremity, open diaphyseal femoral and tibial fractures have been successfully treated with intramedullary nailing, and type I, type II, and type IIIA fractures have been treated with nonreamed intramedullary nails, with encouraging results. External fixation is still the primary method of treatment for salvageable type IIIB and type IIIC fractures (67,68). After surgery, the extremity must be efficiently immobilized. Gross motion delays wound healing and increases the risk of infection. Edema of the extremity is reduced by elevation and pressure dressings. The patient should be carefully observed so that any impairment in circulation or signs of infection can be promptly discovered.

Pearls and Pitfalls &

&

Irrigation and debridement are the mainstay of treatment. Add IV antibiotics.

Table 1 Classification of Open Fractures Type

Wound

Level of contamination

Soft-tissue injury

Bone

I II IIIa IIIA

1 cm long

Clean Moderate

Minimal Moderate, some muscle damage

Simple, minimal comminution Moderate comminution

Usually >10 cm long

High

Severe with crushing

IIIB

Usually >10 cm long

High

Very severe loss of coverage

IIIC

Usually >10 cm long

High

Very severe loss of coverage plus vascular injury requiring repair

Usually comminuted; soft-tissue coverage of bone possible Bone coverage poor; usually requires soft-tissue reconstructive surgery Bone coverage poor; usually requires soft-tissue reconstructive surgery

a

Segmental fractures, farmyard injuries, fractures occurring in a highly contaminated environment, shotgun wounds, or high-velocity gunshot wounds automatically result in classification as a type III open fracture. Source: From Ref. 72.

Chapter 33:

Thromboembolism Venous thromboembolic disease is the most common complication after trauma to the lower extremities in adults, and PE is the most common fatal complication (77,78). Thrombophlebitis occurs in 40% to 60% of patients older than 40 years after fractures of the femur or tibia (10). Of these patients, 10% are at risk of pulmonary emboli, and, unless adequate protection is provided, 2% will die as a result of this complication (79). Risk factors associated with an increased incidence of venous thrombosis include age, obesity, myocardial infarction, heart failure, malignant disease, and the use of oral contraceptives (70,80–82). A congenital deficiency of antithrombin III and the presence of a lupus anticoagulant are also associated with an increased incidence of venous thrombosis (83). When diagnosis is based solely on clinical findings such as complaints of pain, swelling, tenderness, Homan’s sign, fever, and leukocytosis, thromboembolic disease is markedly underdiagnosed (84). Only 5% to 30% of instances of thrombophlebitis will be detected by physical examination alone (83). Objective testing is mandatory for an accurate diagnosis. Measurement of venous hemodynamics by Doppler ultrasonography can detect obstructions to venous blood flow that are produced by proximal thrombi. These methods are most accurate for thrombi in major veins (thigh) and are less accurate for thrombi in small veins (calf) because of the insignificant difference in outflow (85–87). The diagnostic gold standard is still radiocontrast venography (84,88,89). This is the most precise evaluation technique, with high accuracy in diagnosis of clots in the femoral and calf veins and approximately 70% reliability for clots in the iliac veins. Its disadvantages are that it causes some pain, is difficult to repeat frequently, and may induce thrombophlebitis in 5% of patients and an allergic reaction in 0.02% (83).

Treatment Prevention is the best treatment for thromboembolic disease. Patients who are at highest risk of this complication are those who are obese, those who are elderly and have undergone multiple injuries or operations, those with a history of associated cardiovascular or pulmonary disease, and those with a prior episode or family history of deep venous thrombosis (DVT) who are about to undergo major musculoskeletal surgery (70,80–82). The prevention methods for venous thrombosis include sodium warfarin [Coumadin1 (BristolMyers Squibb Company, New York, U.S.A.)], heparin, low-molecular-weight heparin (LMWH), aspirin, dextran, and intermittent pneumatic compression (IPC) devices. The largest clinical comparative study for trauma patients to date found both LMWH and IPC to be equally effective in the prevention of DVT (90). For established DVT, the treatment is heparin administered by continuous IV infusion to maintain

Complications of Fractures

413

the activated partial thromboplastin time at 1.5 to 2.5 times normal. This treatment is continued for approximately one week. Oral warfarin therapy is usually started approximately three days after the initiation of heparin therapy. This overlap allows the prothrombin time to reach the effective range (1.5 to 2 times normal). Orally administered anticoagulant therapy is continued for four to six months, so that recurrent episodes of DVT can be prevented (83,91).

Pulmonary Embolism PE occurs after lower extremity trauma in 5% to 19% of patients with DVT, who are unprotected by anticoagulants. Such emboli are fatal in 5% to 10% of patients after hip fracture. Prophylaxis decreases the incidence and mortality rates associated with PE (79). Diagnosis is based on findings of dyspnea, chest pain, tachypnea, decreased arterial oxygen tension, and changes in the electrocardiogram (S1Q3 pattern of cor pulmonale, axis shift, bundle-branch blocks, and nonspecific ST- and T-wave changes). Each of these changes is nonspecific, but a new appearance in the appropriate clinical setting is highly suggestive of PE. Although 10% of patients with PE never exhibit hypoxemia (92), almost half of trauma patients with sustained hypoxemia experience PE, as demonstrated by pulmonary angiogram (93). Radionuclide ventilation-perfusion scanning is the test most commonly used to obtain an objective diagnosis of PE (94). Scans are considered to indicate a low, intermediate, or high probability of PE on the basis of the degree of conformity to a vascular pattern and the degree of ventilation-perfusion mismatch. Ninety percent of patients with high-probability lesions but only 5% of patients with low-probability lesions will have a PE as demonstrated by angiography (94). Pulmonary angiography is the most accurate method of detecting PE (95). Angiography and lung scanning are complementary studies; both should be performed when necessary and their results should be correlated with those indicated by plain chest radiographs. However, although pulmonary angiography is an invasive procedure, the risks associated with its use for the diagnosis of PE are less than those associated with a severe bleeding episode related to the therapeutic administration of heparin to patients with multiple injuries or those who have undergone surgery (96). Heparinization is the treatment of choice for PE, followed by oral warfarin therapy for four to six months (91). Full anticoagulation has been shown to be necessary for reducing the risk of recurrent pulmonary emboli. For life-threatening massive emboli, open thoracotomy and embolectomy may be necessary (97,98). In cases of repeated PE, partial obstruction of the inferior vena cava with an inferior vena cava filter may be necessary for avoiding a fatal embolus (91). &

Consider DVT prophylaxis in every immobilizing fracture.

414

Yaron et al.

LATE COMPLICATIONS Malunion Malunion occurs when a fracture heals in an unsatisfactory position. Surgical correction of the resulting deformity is indicated if it causes unacceptable cosmetic or functional disabilities. Deformities in the weight-bearing bones may cause abnormal stress through the joints and lead to early osteoarthritis. Four main types of malunion occur: angulation, rotation, shortening, and translation. When malunions are treated, the following facts must be considered. Of the four characteristics that define the acceptability of fracture reduction, the most important is alignment, the second is rotation, the third is restoration of normal length, and the fourth and least important is the actual position of the fragments. A slight deformity can be seriously disabling when a malunion involves a joint or is near a joint. Sometimes, when malunion causes only slight disability, function cannot be improved enough to justify surgery; however, a rotational deformity can be so disabling that surgery is required. Most deformities may correct themselves spontaneously (especially for young patients with immature skeletons) to a functionally satisfying degree during the remodeling phase of fracture healing. Unfortunately, such selfcorrection will not occur with rotational deformities. Before surgical correction is attempted, the patient should participate in intensive physical therapy aimed at regaining the maximal range of motion of the joints. Surgical correction creates osteotomies to correct the deformity; either internal or external fixation may be used. Leg length discrepancy can be treated, if indicated, by lengthening the affected limb. For growing children, an epiphysiodesis of the unaffected bone may be performed at an appropriately planned time (99,100).

Nonunion Nonunion occurs when a fracture has not united within a prescribed period of time, and all healing processes have ceased even though bone continuity has not been restored. The diagnosis is based on clinical and radiographic findings. The amount of time required for a fracture to unite is directly proportional to the amount of energy imparted to the extremity by the injury. The time required for union differs with different bones. Basically, all fractures will unite within four months; a fracture that has not healed within six months may be declared a delayed union and a fracture that has not healed within eight to nine months is defined as a nonunion. Pseudoarthrosis will develop if continued motion occurs at the fracture site and leads to the formation of a pseudocapsule and true synovial lining. Pseudoarthrosis is the final status of nonunion. There are two types of nonunion. In the first type, the ends of the fragments are hypervascular or hypertrophic and are capable of biological reaction. In the second, the ends of the fragments are avascular

or atrophic and are inert and incapable of biological reaction (99,100). The likelihood of nonunion is increased when fractures are open, infected, segmental with an impaired blood supply (usually to the middle fragment), comminuted by severe trauma, uncertainly fixed, immobilized for an insufficient time, treated by ill-advised open reduction, or distracted either by traction or by a plate and screws. Traditional nonsteroidal anti-inflammatory medications have been shown to delay fracture union (101). This effect may be smaller with cyclo-oxygenase-2-specific inhibitors (102). The treatment of nonunion is individual for each case. In general, hypertrophic (hypervascular) nonunions can be treated by stable fixation of the fragments alone, whereas atrophic (avascular) nonunions require decortication and bone grafting for healing. Improvements in electrical and electromagnetic bone growth stimulators are currently being developed. Bone growth stimulators are usually used in conjunction with cast immobilization and weight bearing. External electrical stimulation is especially advantageous for managing infected nonunion or for treating patients for whom surgical intervention is contraindicated (100).

Avascular Necrosis AVN or osteonecrosis is the cellular death of bone tissue after a fracture or dislocation; it occurs when the bone is deprived of a sufficient supply of arterial blood during the fracture, the reduction, or the operative fixation (103–107). This condition may be caused by factors other than trauma, but the pathomechanism is the same: intraluminar vascular compression or disruption of a blood vessel. After trauma, AVN occurs most commonly in the femoral head, the carpal scaphoid, and the body of the talus because of the specific retrograde blood supply to these bones (108). Biopsy is the definitive method of diagnosis. Noninvasive studies include radiography and bone scanning. Advances in magnetic resonance imaging have made earlier diagnosis of AVN of the femoral head possible and have allowed determination of the exact stage and extent of the pathological process, without the use of invasive methods (109,110). Histologically, AVN begins with disappearance of hematopoietic elements and fat cell necrosis, followed by total necrosis of the marrow and osteocytes. The condition is repaired as the body lays new woven bone onto dead trabeculae. As the dead trabeculae are reabsorbed, the remaining unstressed woven bone may fracture, and this fracturing leads to collapse and fragmentation and eventually to an altered joint configuration (108). The main clinical presentation is pain in the involved joint, pain that is aggravated by activity. The natural history of AVN is unpredictable. The best treatment for osteonecrosis depends on the cause, stage of involvement, symptoms, and location. No method of treatment apart from reducing weight

Chapter 33:

bearing for a period of time has proved effective, but the rate and course of progression are variable, and the radiographic picture may not correlate with the clinical symptoms. The most common site of osteonecrosis requiring surgical treatment is the femoral head. There are several surgical treatment options for a femoral head that has not collapsed, such as core biopsy osteotomies and vascular-free fibular grafts, but these techniques have not consistently produced satisfactory results (111–113). Once osteochondral fracture and collapse occur in the femoral head, most patients require major reconstructive surgery. Osteonecrosis of the talus may also require major surgical reconstruction. When osteonecrosis affects the humeral head, healing may occur without surgical intervention, and many patients will continue to function well, even with moderate deformity. Those who experience severe symptoms can be treated successfully with a prosthetic replacement (108).

Heterotopic Ossification Bone commonly forms around joints after trauma and surgery. We distinguish between ectopic calcification, heterotopic bone formation (or ossification), and myositis ossificans. Ectopic calcification is the mineralization of soft-tissue structures, areas in which calcification should not occur or is out of place. For example, calcification of the collateral ligaments is very common after elbow dislocation; it presents no functional limitation and requires no treatment (114). Heterotopic bone formation or ossification refers to the formation of trabecular bone at a location where it does not belong (114,115). Unlike ectopic calcification, this bone forms not necessarily by mineralizing a definable structure but rather by forming new bone in areas of previous hematoma and fibroblast activity. The term ‘‘heterotopic ossification’’ refers to bone that has formed around a joint and is blocking motion. This condition has also been associated with head trauma (116) and burns (117). Myositis ossificans is a subset of heterotopic bone formation. Myositis ossificans is a specific histopathologic condition that occurs in striated muscle, not simply around the capsule or around the joint. Patients with this condition usually experience swelling, pain, and a decreased range of motion. The most common locations include the thigh (the quadriceps muscle), the arm (the brachialis muscle), the shoulder (the deltoid and scapular muscles), and the hand. Initially, radiographs may show only faint, irregular radiodensities, but as the lesion matures, the radiographic appearance changes to that of more solid bone formation. In most patients, the lesion is not attached to the underlying bone, but it can be attached if it lies near the bone and if the original injury induced an adjacent periosteal reaction. Serial radiographs over a period of years will show that the volume of heterotopic bone gradually decreases (118).

Complications of Fractures

415

Treatment is aimed at restoring function. Initially, patients are treated with rest and anti-inflammatory agents. Gentle physical therapy is instituted when local heat, edema, and pain subside. Isometric muscle strengthening and gentle, active-assisted range-of-motion exercises (within the limits of pain) are progressively increased until full function has been recovered. Manipulation should be avoided. Excision of the ectopic bone is rarely indicated.

SUMMARY A fracture is a discontinuity of the bone that is caused by the application of an external force. In most cases, healing of the bone is relatively uneventful, especially in low-energy, closed fractures. However, one has to bear in mind the possible complications that may consequently arise. These complications could be classified into acute and late, each having distinct characteristics and impact. Acute complications may be life threatening, including hypovolemic shock and fatal PE or FE syndrome. Acute limb threatening complications should be promptly recognized, namely, vascular compromise, neurological deterioration and impending infection. Therefore, an appropriate level of suspicion and alertness may dramatically improve the clinical outcome of these patients. Late complications such as nonunion, malunion, and AVN mainly influence the long-term life quality of the patient, causing considerable morbidity leading to recurrent surgical interventions. Some of these complications are preventable, whereas others are related to the trauma mechanism. Good clinical practice and effective trauma team cooperation might lower the incidence and severity of undesired complications.

REFERENCES 1. Evans FG. Relation of the physical properties of bone to fractures. Instr Course Lect 1961; 18:110–121. 2. Hipp JA, Hayes WC. Biomechanics of fractures. In: Browner BD, Jupiter JB, Levine AM, Trafton PG, eds. Skeletal Trauma. Vol. 1. 3d ed. Philadelphia: WB Saunders, 2003:90–119. 3. Allum RL, Mowbray MA. A retrospective review of the healing of fractures of the shaft of the tibia with special reference to the mechanism of injury. Injury 1980; 11:304–308. 4. Pentecost RL, Murray RA, Brindley HH. Fatigue, insufficiency, and pathologic fractures. JAMA 1964; 187:1001–1004. 5. Kyle RF, Gustilo RB, Premer RF. Analysis of six hundred and twenty-two intertrochanteric hip fractures. J Bone Joint Surg Am 1979; 61:216–221. 6. Beck TJ, Ruff CB, Shaffer RA, Betsinger K, Trone DW, Brodine SK. Stress fracture in military recruits: gender differences in muscle and bone susceptibility factors. Bone 2000; 27:437–444.

416

Yaron et al.

7. Baker J, Frankel VH, Burstein AH. Fatigue fractures: biomechanical considerations [abstr]. J Bone Joint Surg Am 1972; 54:1345–1346. 8. Ostrum RF, Verghese GB, Santner TJ. The lack of association between femoral shaft fractures and hypotensive shock. J Orthop Trauma 1993; 7:338–342. 9. Poole GV, Ward EF. Causes of mortality in patients with pelvic fractures. Orthopedics 1994; 17:691–696. 10. Fox MA, Mangiante EC, Fabian TC, Voeller GR, Kudsk KA. Pelvic fractures: an analysis of factors affecting prehospital triage and patient outcome. South Med J 1990; 83:785–788. 11. Burgess AR, Eastridge BG, Young JW, et al. Pelvic ring disruptions: effective classification system and treatment. J Trauma 1990; 30:848–856. 12. Buckle R, Browner BD, Morandi M. Emergency reduction for pelvic ring disruptions and control of associated hemorrhage using the pelvic stabilizer. Tech Orthop 1995; 9:258–266. 13. Ghanayem AJ, Wilber JH, Lieberman JM, Motta AO. The effect of laparotomy and external fixator stabilization on pelvic volume in an unstable pelvic injury. J Trauma 1995; 38:396–401. 14. Jerrard DA. Pelvic fractures. Emerg Med Clin North Am 1993; 11:147–163. 15. Agnew SG. Hemodynamically unstable pelvic fractures. Orthop Clin North Am 1994; 25:715–721. 16. Sanders R, DiPasquale T. External fixation of the pelvis: application of the resuscitation frame. Tech Orthop 1990; 4:60–64. 17. Ganz R, Krushell RJ, Jakob RP, Kuffer J. The antishock pelvic clamp. Clin Orthop 1991; 267:71–78. 18. Ben-Menachem Y. Pelvic fractures: diagnostic and therapeutic angiography. Instr Course Lect 1988; 37:139–141. 19. Siliski JM. Fractures of the pelvic ring. In: Browner BD, Jupiter JB, Levine AM, Trafton PG, eds. Skeletal Trauma. Vol. 2. 3d ed. Philadelphia: WB Saunders, 1992:2052–2056. 20. Malan E, Tattoni G. Physio-and anatomo-pathology of acute ischemia of the extremities. J Cardiovasc Surg 1963; 4:212–225. 21. Frykberg ER, Dennis JW, Bishop K, Laneve L, Alexander RH. The reliability of physical examination in the evaluation of penetrating extremity trauma for vascular injury: results at one year. J Trauma 1991; 31:502–511. 22. Lynch K, Johansen K. Can Doppler pressure mea surement replace ‘‘exclusion’’ arteriography in the diagnosis of occult extremity arterial trauma? Ann Surg 1991; 214:737. 23. Feliciano DV, Moore EE, Mattox KL, eds. Trauma. 3d ed. Publisher, 1996. 24. Collins JA, Gordon WC Jr., Hudson TL, Irvin RW Jr., Kelly T, Hardaway RM III. Inapparent hypoxemia in casualties with wounded limbs: pulmonary fat embolism? Ann Surg 1968; 167:511–520. 25. Collins JA, Hudson TL, Hamacher WR, Rokous J, Williams G, Hardaway RM III. Systemic fat embolism in four combat casualties. Ann Surg 1968; 167:493–499. 26. Armin J, Grant RT. Observations on gross pulmonary fat embolism in man and the rabbit. Clin Sci 1951; 10:441–469. 27. Jacobs RR, Wheeler EJ, Jelenko C III, McDonald TF, Bliven FE. Fat embolism: a microscopic and ultrastruc-

28. 29.

30. 31.

32. 33. 34. 35.

36. 37.

38. 39.

40. 41. 42. 43. 44.

45. 46. 47.

ture evaluation of two animal models. J Trauma 1973; 13:980–993. Kerstell J. Pathogenesis of post-traumatic fat embolism. Am J Surg 1971; 121:712–715. Kerstell J, Hallgren B, Rudenstam CM, Svanborg A. The chemical composition of the fat emboli in the post-absorptive dog. Acta Med Scand 1969; suppl 499:3–18. Meek RN, Woodruff B, Allardyce DB. Source of fat macroglobules in fractures of the lower extremity. J Trauma 1972; 12:432–434. Schnaid E, Lamprey JM, Viljoen MJ, Joffe BI, Seftel HC. The early biochemical and hormonal profile of patients with long bone fractures at risk of fat embolism syndrome. J Trauma 1987; 27:309–311. Barie PS, Minnear FL, Malik AB. Increased pulmonary vascular permeability after bone marrow injection in sheep. Am Rev Respir Dis 1981; 123:648–653. Bergentz SE. Studies on the genesis of posttraumatic fat embolism. Acta Chir Scand 1961; suppl 282:1–72. Bone L, Bucholz R. The management of fractures in the patient with multiple trauma. J Bone Joint Surg Am 1986; 68:945–949. Hammerschmidt DE, Weaver LJ, Hudson LD, Craddock PR, Jacob HS. Association of complement activation and elevated plasma-C5a with adult respiratory distress syndrome. Pathophysiological relevance and possible prognostic value. Lancet 1980; 1:947–949. Kaplan JE, Saba TM. Humoral deficiency and reticuloendothelial depression after traumatic shock. Am J Physiol 1976; 230:7–14. LeQuire VS, Shapiro JL, LeQuire CB, Cobb CA Jr., Fleet WF Jr. A study of the pathogenesis of fat embolism based on human necropsy material and animal experiments. Am J Pathol 1959; 35:999–1015. Maruyama Y, Little JB. Roentgen manifestations of traumatic pulmonary fat embolism. Radiology 1962; 79:945–952. Tennenberg SD, Jacobs MF, Solomkin JS. Complement-mediated neutrophil activation in sepsis- and trauma-related adult respiratory distress syndrome. Clarification with radioaerosol lung scans. Arch Surg 1987; 122:26–32. Peltier LF. An appraisal of the problem of fat embolism. Surg Gynecol Obstet 1957; 104:313–324. Peltier LF. The diagnosis and treatment of fat embolism. J Trauma 1971; 11:661–667. Peltier LF, Adler F, Lai SP. Fat embolism: the significance of an elevated serum lipase after trauma to bone. Am J Surg 1960; 99:821–826. Crocker SH, Eddy DO, Obenauf RN, Wismar BL, Lowery BD. Bacteremia: host-specific lung clearance and pulmonary failure. J Trauma 1981; 21:215–220. Gossling HR, Pellegrini VD Jr. Fat embolism syndrome: a review of the pathophysiology and physiological basis of treatment. Clin Orthop 1982; 165: 68–82. Evarts CM. The fat embolism syndrome: a review. Surg Clin North Am 1970; 50:493–507. Whitesides TE, Heckman MM. Acute compartment syndrome: update on diagnosis and treatment. J Am Acad Orthop Surg 1996; 4:209–218. Bone LB, Johnson KD, Weigelt J, Scheinberg R. Early versus delayed stabilization of femoral fractures.

Chapter 33:

48.

49. 50.

51.

52.

53. 54. 55.

56.

57.

58. 59.

60. 61.

62.

63. 64.

A prospective randomized study. J Bone Joint Surg Am 1989; 71:336–340. Fabian TC, Hoots AV, Stanford DS, Patterson CR, Mangiante EC. Fat embolism syndrome: a prospective evaluation in 92 fracture patients. Crit Care Med 1990; 18:42–45. Goris RJ. The injury severity score. World J Surg 1983; 7:12–18. Goris RJ, Gimbrere JS, Van Niekerk JL, Schoots FJ, Body LH. Early osteosynthesis and prophylactic mechanical ventilation in the multitrauma patient. J Trauma 1982; 22:895–903. Johnson KD, Cadambi A, Seibert GB. Incidence of adult respiratory distress syndrome in patients with multiple musculoskeletal injuries: effect of early operative stabilization of fractures. J Trauma 1985; 25:375–384. Lozman J, Deno DC, Feustel PJ, et al. Pulmonary and cardiovascular consequences of immediate fixation or conservative management of long bone fractures. Arch Surg 1986; 121:992–999. Pipkin G. The early diagnosis and treatment of fat embolism. Clin Orthop 1958; 15:171–182. Riska E, Myllynen P. Fat embolism in patients with multiple injuries. J Trauma 1982; 22:891–894. Riska EB, von Bonsdorff H, Hakkinen S, Jaroma H, Kiviluoto O, Paavilainen T. Primary operative fixation of long bone fractures in patients with multiple injuries. J Trauma 1977; 17:111–121. Rokkanen P, Lahdensuu M, Kataja J, Julkunen H. The syndrome of fat embolism: analysis of thirty consecutive cases compared to trauma patients with similar injuries. J Trauma 1970; 10:299–306. Ruedi T, Wolff G. Vermeidung postraumatischer komplikationen durch fruhe definitive versorgung von polytraumatisierten mit frakturen des bewegungsapparats. [Prevention of post-traumatic complications through immediate therapy of patients with multiple injuries and fractures.]. Helv Chir Acta 1975; 42: 507–512. Seibel R, LaDuca J, Hassett JM, et al. Blunt multiple trauma (ISS 36), femur traction, and the pulmonary failure-septic state. Ann Surg 1985; 202:283–295. Talucci RC, Manning J, Lampard S, Bach A, Carrico CJ. Early intramedullary nailing of femoral shaft fractures: a cause of fat embolism syndrome. Am J Surg 1983; 146:107–111. ten Duis HJ, Nijsten MW, Klasen HJ, Binnendijk B. Fat embolism in patients with an isolated fracture of the femoral shaft. J Trauma 1988; 28:383–390. Whitesides TE, Haney TC, Morimoto K, Harada H. Tissue pressure measurements as a determinant for the need of fasciotomy. Clin Orthop Relat Res 1975; 113:43–51. Worlock P, Slack R, Harvey L, Mawhinney R. The prevention of infection in open fractures. An experimental study of the effect of antibiotic therapy. J Bone Joint Surg Am 1988; 70:1341–1347. Tornetta P III, Templeman D. Compartment syndrome associated with tibial fracture. Instr Course Lect 1997; 46:303–308. Garr JL, Gentilello LM, Cole PA, Mock CN, Matsen FA III. Monitoring for compartmental syndrome using near-infrared spectroscopy: a

65.

66.

67.

68. 69. 70.

71.

72. 73. 74. 75. 76.

77. 78.

79.

80. 81. 82.

Complications of Fractures

417

noninvasive, continuous, transcutaneous monitoring technique. J Trauma 1999; 46(4):613–616. Giannotti G, Cohn SM, Brown M, Varela JE, McKenney MG, Wiseberg JA. Utility of near-infrared spectroscopy in the diagnosis of lower extremity compartment syndrome. J Trauma 2000; 48(3):396–399. Gentilello LM, Sanzone A, Wang L, Liu PY, Robinson L. Near-infrared spectroscopy versus compartment pressure for the diagnosis of lower extremity compartmental syndrome using electromyography-determined measurements of neuromuscular function. J Trauma 2001; 51(1):1–8. Beherens FF, Sirkin MS. Fractures with soft tissue injuries. In: Browner BD, Jupiter JB, LevineAM, Trafton PG, eds. Skeletal Trauma. Vol. 13. 3d ed. Philadelphia: WB Sauders, 2003:293–319. Gustilo RB, Merkow RL, Templeman D. The management of open fractures. J Bone Joint Surg Am 1990; 72:299–304. Waterman NG, Howell RS, Babich M. The effect of a prophylactic antibiotic (cephalothin) on the incidence of wound infection. Arch Surg 1968; 97:365–370. Crandon AJ, Peel KR, Anderson JA, Thompson V, McNicol GP. Postoperative deep vein thrombosis: identifying high-risk patients. Br Med J 1980; 281:343–344. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am 1976; 58:453–458. Gustilo RB, Gruninger RP, Davis T. Classification of type III (severe) open fractures relative to treatment and results. Orthopedics 1987; 10:1781–1788. Gustilo RB. Current concepts in the management of open fractures. Instr Course Lect 1987; 36:359–366. Gustilo RB, Simpson L, Nixon R, Ruiz A, Indeck W. Analysis of 511 open fractures. Clin Orthop 1969; 66:148–154. Chapman MW. The role of intramedullary fixation in open fractures. Clin Orthop 1986; 212:26–34. Worlock P, Slack R, Harvey L, Mawhinney R. The prevention of infection in open fractures: an experimental study of the effect of fracture stability. Injury 1994; 25: 31–38. Tubiana R, Duparc J. Frequency and prevention of thromboembolic complications in orthopaedic and accident surgery. Acta Orthop Belg 1962; 28:605–611. Hamilton HW, Crawford JS, Gardiner JH, Wiley AM. Venous thrombosis in patients with fracture of the upper end of the femur. A phlebographic study of the effect of prophylactic anticoagulation. J Bone Joint Surg Br 1970; 52(2):268–289. Sikorski JM, Hampson WG, Staddon GE. The natural history and aetiology of deep vein thrombosis after total hip replacement. J Bone Joint Surg Br 1981; 63:171–177. Coon WW, Coller FA. Some epidemiologic considerations of thromboembolism. Surg Gynecol Obstet 1959; 109:487–501. Coon WW, Willis PW. Deep venous thrombosis and pulmonary embolism: prediction, prevention and treatment. Am J Cardiol 1959; 4:611–621. Gallus AS, Hirsh J. Prevention of venous thromboembolism. Semin Thromb Hemost 1976; 2:232–290.

418

Yaron et al.

83. Aaron RK, Ciombor D. Venous thromboembolism in the orthopedic patient. Surg Clin North Am 1983; 63:529–537. 84. Couch NP. AMA Archives symposium on diagnostic techniques in phlebothrombosis. Arch Surg 1972; 104:132–133. 85. Flinn WR, Sandager GP, Cerullo LJ, Havey RJ, Yao JS. Duplex venous scanning for the prospective surveillance of perioperative venous thrombosis. Arch Surg 1989; 124:901–905. 86. Froehlich JA, Dorfman GS, Cronan JJ, Urbanek PJ, Herndon JH, Aaron RK. Compression ultrasonography for the detection of deep venous thrombosis in patients who have a fracture of the hip. A prospective study. J Bone Joint Surg Am 1989; 71:249–256. 87. Lensing A, Prandoni P, Brandjes D, et al. Detection of deep-vein thrombosis by real-time B-mode ultrasonography. N Engl J Med 1989; 320:342–345. 88. Borow M, Goldson H. Postoperative venous thrombosis. Evaluation of five methods of treatment. Am J Surg 1981; 141:245–251. 89. Culver D, Crawford JS, Gardiner JH, Wiley AM. Venous thrombosis after fractures of the upper end of the femur. A study of incidence and site. J Bone Joint Surg Br 1970; 52:61–69. 90. Ginzburg E, Cohn SM, Lopez J, Jackowski J, Brown M, Hameed SM. Miami Deep Vein Thrombosis Study Group Randomized clinical trial of intermittent pneumatic compression and low molecular weight heparin in trauma. Br J Surg 2003; 90(11):1338–1344. 91. Herndon GH. Chapter 10: Circulatory, respiratory, hematopoietic, gastrointestinal, genitourinary and integumentary systems. In: Fitzgerald RH Jr., ed. Orthopaedic Knowledge Update 2. Park Ridge, Illinois: American Academy of Orthopaedic Surgeons, 1987:107–122. 92. Henry JW, Stein PD, Gottschalk A, Relyea B, Leeper KV Jr. Scintigraphic lung scans and clinical assessment in critically ill patients with suspected acute pulmonary embolism. Chest 1996; 109(2): 462–466. 93. Brathwaite C, O’Malley K, Ross S, Pappas PJ, Alexander JB, Spence RK. Continuous pulse oximetry and the diagnosis of pulmonary embolism in critically ill trauma patients. J Trauma 1992; 33(4):528–531. 94. McBride K, LaMorte WW, Menzoian JO. Can ventilation-perfusion scans accurately diagnose acute pulmonary embolism? Arch Surg 1986; 121:754–757. 95. Bookstein JJ. Segmental arteriography in pulmonary embolism. Radiology 1969; 93:1007–1012. 96. Patterson B, Marchand R, Ranawat C. Complications of heparin therapy after total joint arthroplasty. J Bone Joint Surg Am 1989; 71:1130–1134. 97. Mavor GE, Galloway JM. Iliofemoral venous thrombosis. Pathological considerations and surgical management. Br J Surg 1969; 56:45–49. 98. Miller GA. The diagnosis and management of massive pulmonary embolism. Br J Surg 1972; 59:837–839. 99. Palei D. Principles of deformity correction. In: Browner BD, Jupiter JB, Levine AM, Trafton PG, eds. Skeletal Trauma. Vol. 62. 3d ed. Philadelphia: WB Sauders, 2003:2519–2576. 100. Beaty J. Congenital anomalies of lower extremity. In: Canale ST, ed. Campbell’s Operative Orthopedics. Vol. 26. 10th ed. Philadelphia: Mosby, 2003:1048–1054.

101. Giannoudis PV, MacDonald DA, Matthews SJ, Smith RM, Furlong AJ, De Boer P. Nonunion of the femoral diaphysis. The influence of reaming and non-steroidal anti-inflammatory drugs. J Bone Joint Surg Br 2000; 82(5):655–658. 102. Brown KM, Saunders MM, Kirsch T, Donahue HJ, Reid JS. Effect of COX-2-specific inhibition on fracture-healing in the rat femur. J Bone Joint Surg Am 2004; 86-A(1):116–123. 103. Brodetti A. The blood supply of the femoral neck and head in relation to the damaging effects of nails and screws. J Bone Joint Surg Br 1960; 42:794–801. 104. Brodetti A. An experimental study on the use of nails and bolt screws in the fixation of fractures of the femoral neck. Acta Orthop Scand 1961; 31:247–271. 105. Catto M. A histological study of avascular necrosis of the femoral head after transcervical fracture. J Bone Joint Surg Br 1965; 47:749–776. 106. Catto M. The histological appearances of late segmental collapse of the femoral head after transcervical fracture. J Bone Joint Surg Br 1965; 47:777–791. 107. Rizzo PF, Gould ES, Lyden JP, Asnis SE. Diagnosis of occult fractures about the hip. Magnetic resonance imaging compared with bone-scanning. J Bone Joint Surg Am 1993; 73:395–401. 108. Levine M. In: Beaty JH, ed. Orthopaedic Knowledge Update 6: Home Study Syllabus. Rosemont, Illinois: American Academy of Orthopaedic Surgeons, 1999; 38:455–492. 109. Greenspan A. Radiologic evaluation of trauma. In: Greenspan A, ed. Orthopedic Radiology. 3d ed. Philadelphia: Lippincot Williams and Wilkins, 1999:71–76. 110. Lang P, Jergesen HE, Genant HK, Moseley ME, Schulte-Monting J. Magnetic resonance imaging of the ischemic femoral head in pigs. Dependency of signal intensities and relaxation times on elapsed time. Clin Orthop 1989; 244:272–280. 111. Brunelli G, Brunelli G. Free microvascular fibular transfer for idiopathic femoral head necrosis: long-term follow-up. J Reconstr Microsurg 1991; 7:285–295. 112. Fujimaki A, Yamauchi Y. Vascularized fibular grafting for treatment of aseptic necrosis of the femoral head— preliminary results in four cases. Microsurgery 1983; 4:17–22. 113. Urbaniak JR. Aseptic necrosis of the femoral head treated by vascularized fibular graft. In: Urbaniak JR, ed. Microsurgery for Major Limb Reconstruction. St. Louis: Mosby, 1987:178–184. 114. Ackerman LV. Extra-osseous localized non-neoplastic bone and cartilage formation (so-called myositis ossificans): clinical and pathological confusion with malignant neoplasms. J Bone Joint Surg Am 1958; 40:279–298. 115. Coventry MB. Ectopic ossification about the elbow. In: Morrey BF, ed. The Elbow and Its Disorders. Philadelphia: WB Saunders, 1985:464–471. 116. Garland DE, Hanscom DA, Keenan MA, Smith C, Moore T. Resection of heterotopic ossification in the adult with head trauma. J Bone Joint Surg Am 1985; 67:1261–1269. 117. Seth MK, Khurana JK. Bony ankylosis of the elbow after burns. J Bone Joint Surg Br 1985; 67:747–749. 118. Greenspan A. Radiologic evaluation of trauma. In: Greenspan A, ed. Orthopedic Radiology. 3d. Philadelphia: Lippincot Williams and Wilkins, 1999:63–80.

34 Complications of Dislocations Howard Richter and Gregory A. Zych Department of Orthopedics and Rehabilitation, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

A dislocation is a complete disruption of a joint so that the articular surfaces are no longer in contact. Most often, the cause of the dislocation is a traumatic event and the result is a loss of structural stability of the joint. Traumatic dislocation usually causes pain, deformity of the involved extremity, and marked limitation of joint motion. Appropriate treatment of a dislocation involves careful neurovascular evaluation, radiographic studies, and prompt reduction of the involved joint. Complications and long-term sequelae of traumatic dislocation include neurovascular injuries, avascular necrosis, heterotopic bone formation, posttraumatic arthritis, musculotendinous injuries, joint instability, and joint stiffness.

Vascular injuries have also been reported after elbow dislocations. Approximately 30 cases of brachial artery disruption have been reported after elbow dislocations (3). Brachial artery injury occurs most often with open dislocations and in the presence of associated fractures. The most common dislocation resulting in vascular injury is dislocation of the knee (Fig. 1). Green and Allen, in a review of 245 knee dislocations, found a 32% incidence of popliteal artery injury in association with traumatic dislocation of the tibiofemoral

VASCULAR INJURIES ASSOCIATED WITH DISLOCATION Vascular injury can occur in the involved extremity because the joint is forcibly displaced from its anatomic location. The medical literature contains reports of 200 cases of vascular injury after shoulder dislocations (1). Most of these cases involved elderly patients, whose vessels are stiffer and more fragile. The axillary artery consists of three sections that lie medial to, behind, and lateral to the pectoralis minor muscle. The second section of the artery is most commonly injured when the thoracoacromial trunk is avulsed, and the third section is most commonly injured when the subscapular and circumflex branches are avulsed (2). The mechanism of injury is believed to be the forced abduction and external rotation of the shoulder. Because the humeral head dislocates anteriorly, the artery becomes taut and is displaced forward. Because the artery is relatively fixed at the lateral margin of the pectoralis minor muscle, this forward displacement causes the pectoralis minor muscle to act as a fulcrum over which the artery is deformed and ruptured. Patients with axillary artery injuries resulting from shoulder dislocation experience pain, an expanding hematoma, pulse deficit, peripheral cyanosis, pallor, and neurologic dysfunction. Treatment requires emergent vascular repair.

Figure 1 Posterior dislocation of a knee. The direction of the dislocation is determined by the position of the tibia relative to that of the femur.

420

Richter and Zych

Figure 2 Angiogram of a knee after dislocation. Note the beading of the vessel just proximal to the knee joint. This finding indicates intimal damage to the popliteal artery.

joint (4). Anatomically, the popliteal artery is tethered proximally to the femur in the adductor hiatus and distally to the fibula by the fibrous bands of the soleus fascia. This tethering of the artery explains the high incidence of vascular injury in association with disruption and subsequent displacement of the knee joint (Fig. 2). Posterior dislocations (posterior displacement of the tibia) can result in complete transection of the artery because the vessel impacts the posterior rim of the tibial plateau. Anterior dislocations (anterior displacement of the tibia) typically cause a contusion of the vessel with intimal injury. Popliteal artery injuries are a surgical emergency requiring immediate repair. Consensus exists among traumatologists that circulation to the extremity needs to be restored within six to eight hours if the risk of amputation is to be minimized because an amputation rate of 85% has been reported for cases in which the popliteal artery injury was left untreated or was not repaired within eight hours (5).

NEURAL INJURIES ASSOCIATED WITH DISLOCATION In the event of the joint being dislocated, the surrounding neural structures can be contused, stretched, or even lacerated. The sciatic nerve, specifically the peroneal component, is injured in 8% to 19% of patients who sustain a posterior dislocation of the hip (6). Epstein et al. (7) postulated that one of the most important factors in the production of sciatic nerve injury is the marked internal rotation of the hip that occurs at the time of dislocation. The internal rotation causes a winding and tightening of the sciatic nerve. As the hip dislocates posteriorly, it directly contuses or entraps the nerve, and this contusion or entrapment often results in peroneal nerve palsy. These nerve injuries must be recognized early because nerve tissue does not tolerate pressure well

and permanent ischemic changes occur quickly. Treatment again involves a thorough physical and radiographic examination followed by a prompt reduction. Most reports of series of patients show that 60% to 70% of patients with sciatic nerve palsy eventually experience functional recovery (8). The axillary nerve is the neural structure most commonly injured in association with shoulder dislocations. Axillary nerve injury has been reported to occur with 5% to 33% of all shoulder dislocations; it is most common among elderly patients, after highenergy trauma, and in association with long-standing dislocations (9). The axillary nerve lies directly across the anterior surface of the subscapularis muscle. As the humeral head displaces the subscapularis muscle and tendon forward and anterior in glenohumeral dislocations, traction and direct pressure are produced on the axillary nerve and result in injury to the neural structures. The diagnosis of nerve injury is made on the basis of neurologic signs such as weakness or numbness after dislocation. Blom and Dahlback demonstrated that the usual sensory testing of the axillary nerve on the skin of the lateral arm just above the deltoid insertion yields unreliable diagnostic findings (10). Most axillary nerve injuries are traction neuropraxias and will recover completely. Common peroneal nerve palsies can result after knee dislocations. Typically, 20% to 40% of knee dislocations, primarily those involving lateral and posterolateral dislocations of the tibia, result in peroneal nerve palsies (11). Approximately half of these palsies are permanent. Treatment involves symptomatic care with the use of assistive ambulation devices, possible surgical exploration, tendon transfers, or nerve grafting.

AVASCULAR NECROSIS AFTER DISLOCATION Avascular necrosis of the femoral head is a well-recognized complication after posterior dislocation of the hip. The incidence of avascular necrosis varies from 6% to 40% after posterior dislocation of the hip (12). The cause of the avascular change is believed to be ischemia caused by damage to the vessels of the ligamentum teres and the retinaculum of Weitbrecht. Both the degree of initial trauma and the time during which the hip remains dislocated have been found to directly correlate with the likelihood of avascular necrosis. Hougaard and Thomsen reported that reduction within six hours of injury substantially decreased the incidence of avascular necrosis (13). Therefore, prompt reduction is mandated in all cases of hip dislocation. Avascular necrosis has been reported to occur as long as two to five years after posterior dislocation of the hip. Thus, careful monitoring and follow-up must be maintained after reduction if possible avascular changes are to be detected. If avascular necrosis is not diagnosed early, femoral head collapse and traumatic arthritis will result.

Chapter 34: Complications of Dislocations

HETEROTOPIC BONE FORMATION AFTER DISLOCATION Heterotopic ossification of the surrounding soft tissue can occur as a result of traumatic joint dislocation. This condition is most frequently noted after elbow dislocation and occurs in as many as 75% of cases (14). The most common sites of periarticular calcification are the anterior elbow region and the collateral ligaments. Ectopic bone formation is associated with delayed surgical intervention, closed head injury, and aggressive passive elbow joint manipulation after dislocation. If heterotopic bone formation is limiting joint motion, resection of the involved bone should be delayed until the ossification appears mature on plain radiographs, typically six months after the initial trauma. Radiographic maturation is characterized by well-defined cortical margins with linear trabeculations. Ectopic bone formation has also been reported after hip dislocation (Fig. 3). Epstein reported a 2% incidence of myositis ossificans after hip dislocation (15). The ectopic bone formation is believed to result from the initial muscle damage, the formation of hematoma, and the influx of inflammatory mediators. The severity of the traumatic dislocation seems to be the best predicator of the occurrence of bone formation in the surrounding tissue. Restriction of motion is not common; therefore, treatment is based on symptoms and the recommended excision time of the extraosseous tissue correlates with the maturity of the ectopic bone.

POSTTRAUMATIC ARTHRITIS AFTER DISLOCATION The traumatic process of a joint dislocation can permanently and irreversibly injure the articular cartilage

Figure 3 Heterotopic ossification of the soft tissues surrounding the hip joint after posterior dislocation of the hip.

421

lining the involved joint. This injury to the cartilage can lead to the development of osteoarthritis. The severity of the initial trauma and the structural damage to the articular surface are the primary factors in determining the later development of posttraumatic arthritis. Hip and ankle dislocations are most commonly associated with the later development of arthritic changes.

MUSCULOTENDINOUS INJURY AFTER DISLOCATION Dislocation may disrupt surrounding muscles and tendons and cause a functional disability. Anterior and inferior glenohumeral dislocations can injure the rotator cuff of the shoulder. Tijmes et al. (16) reported rotator cuff tears in association with 28% of anterior dislocations. The frequency of this complication increases with the age of the patient. Thirty percent of patients older than 40 years and 80% of patients older than 60 years typically sustain rotator cuff tears with dislocations (16). Injury to the rotator cuff causes pain and weakness with external rotation and abduction of the shoulder. Treatment involves appropriate radiographic studies to assess the extent of the rotator cuff tear, conservative therapy, and possibly surgical repair.

INSTABILITY AFTER DISLOCATION Traumatic dislocations can seriously injure the supportive structures of the joint, thus rendering it unstable during physiologic motion. This instability is most common after glenohumeral dislocations in young patients. McLaughlin and MacLellan (17) observed recurrence of 95% of 181 primary traumatic dislocations of the shoulder in teenagers. Most of the secondary dislocations occurred within two years of the initial traumatic dislocation. This increased rate of instability is linked to the high incidence of disruption of the labral attachment of the anterior inferior glenohumeral ligament and of fracture of the anterior inferior glenoid rim (Bankart lesion) after dislocations. Disruption of the anterior inferior glenohumeral ligament, which is the primary static restraint to anterior shoulder dislocation, increases the patient’s susceptibility to repeated dislocation during abduction and external rotation. Treatment involves sling immobilization and possibly surgical intervention. Hip dislocation can also result in persistent joint instability. Lutter reported the occurrence of repeated dislocations after approximately 1% of hip dislocations (18). Instability is believed to be linked to repeated injury, a shallow acetabulum or deficient posterior rim, and massive soft-tissue injury. Treatment typically involves some form of capsular plication. After traumatic elbow dislocation, insufficiency of the lateral elbow ligaments can lead to elbow instability. Injury to the ulnar lateral collateral ligament is primarily responsible for this lack of stability. With injury to the ulnar collateral ligament, the elbow

422

Richter and Zych

subluxates with elbow supination and flexion. Typically, nonoperative treatment with elbow immobilization is adequate to allow regaining the stability of the joint. Most dislocations of the knee involve tears of the central pivot, including both the posterior cruciate ligament and the anterior cruciate ligament. The collateral ligaments are also frequently disrupted in dislocations of the tibiofemoral joint. Incompetence of these ligaments— which are the primary stabilizers to anterior, posterior, varus, and valgus stresses— makes the knee quite unstable and severely limits the patient’s ability to ambulate. Current treatment guidelines recommend early (within one to two weeks) surgical reconstruction or repair of all involved ligamentous structures so that stability and mobility of the knee can be maximized.

JOINT STIFFNESS AFTER DISLOCATION Stiffness of the elbow joint is very common after traumatic dislocation. Most patients will lose the terminal 10 to 15 of elbow extension after elbow dislocation. The stiffness is frequently caused by thickening and fibrosis of the anterior joint capsule. Early active mobilization, usually within two weeks of injury, is necessary if this complication is to be minimized. Elbow capsular release can be considered if an elbow contracture of more than 30 persists after six months of therapy.

REFERENCES 1. Gugenheim S, Sanders RJ. Axillary artery rupture caused by shoulder dislocation. Surgery 1984; 95:55–58. 2. Beeson M. Complications of shoulder dislocation. Am J Emerg Med 1999; 17:288–295. 3. Cohen MS, Hastings H II. Acute elbow dislocation: evaluation and management. J Am Acad Orthop Surg 1998; 6:15–23.

4. Green NE, Allen BL. Vascular injuries associated with dislocation of the knee. J Bone Joint Surg Am 1977; 59: 236–239. 5. Good L, Johnson RJ. The dislocated knee. J Am Acad Orthop Surg 1995; 3:284–292. 6. Stewart MJ, McCarroll HR Jr., Mulhollan JS. Fracturedislocation of the hip. Acta Orthop Scand 1975; 46: 507–525. 7. Epstein HC, Wiss DA, Cozen L. Posterior fracture dislocation of the hip with fractures of the femoral head. Clin Orthop 1985; 201:9–17. 8. Epstein HC. Traumatic Dislocation of the Hip. Baltimore: Williams & Wilkins, 1980. 9. Rockwood C, Wirth M. Subluxations and dislocations about the glenohumeral joint. In: Bucholz, Heckman, eds. Rockwood and Green’s Fractures in Adults. Vol. 2. 5th ed. Philadelphia: Lippincott-Raven Publishers, 2001:1109–1201. 10. Blom S, Dahlback LO. Nerve injuries in dislocations of the shoulder joint and fractures of the neck of the humerus. A clinical and electromyographical study. Acta Chir Scand 1970; 136:461–466. 11. Siliski JM, Plancher K. Dislocation of the knee. Presented at the Annual Meeting of the American Academy of Orthopedic Surgeons, 1989. Proceedings of the American Academy of Orthopedic Surgeons, 1989. 12. Upadhyay SS, Moulton A. The long-term results of traumatic posterior dislocation of the hip. J Bone Joint Surg Br 1981; 63:548–551. 13. Hougaard K, Thomsen PB. Coxarthrosis following traumatic posterior dislocation of the hip. J Bone Joint Surg Am 1987; 69:679–683. 14. Josefsson PO, Johnell O, Gentz CF. Long-term sequelae of simple dislocation of the elbow. J Bone Joint Surg Am 1984; 66:927–930. 15. Epstein HC. Traumatic dislocations of the hip. Clin Orthop 1973; 92:116–142. 16. Tijmes J, Loyd HM, Tullos HS. Arthrography in acute shoulder dislocations. South Med J 1979; 72:564–567. 17. McLaughlin HL, MacLellan DI. Recurrent anterior dislocation of the shoulder. II. A comparative study. J Trauma 1967; 7:191–201. 18. Lutter LD. Post-traumatic hip redislocation. J Bone Joint Surg Am 1973; 55:391–394.

35 Complications of Amputations Yoram Klein Division of Trauma and Emergency Surgery, Kaplan Medical Center, Rehovot and Department of Surgery, Hadassah EIN, Kerem Medical Center, Jerusalem, Israel Mauricio Lynn Division of Trauma and Surgical Critical Care, DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

GENERAL CONSIDERATIONS Amputation is one of the earliest recorded operative procedures in the history of mankind. There is evidence that amputations were preformed as early as the Neolithic era. Hammurabi, King of Babylon (1792–1750 B.C.), dictated the use of amputation as a punishment for offenders, including surgeons who killed or blinded their patients during treatment (1). A few cultures still use amputation as a punishment today, mainly for thieves. The most important year in the history of amputation was probably 1338, the year in which gunpowder was introduced. Subsequently, surgery evolved along with the history of war, and military surgeons led the way in improving surgical techniques and perioperative care. Surgical complications changed over the years, as did indications for amputation, surgical techniques, and perioperative care. Hemorrhage, which was the most deadly complication in the early era, was treated with hot oil coagulation and by amputating through devitalized tissue, as advised by Hippocrates. In the Middle Ages, pressure bandages made of ox bladder were used to control postoperative bleeding. In the 16th century, the tourniquet was introduced and hemostatic ligature was accepted as the preferred means of controlling bleeding. Intraoperative pain was another problem. The most common technique for minimizing pain was to perform the procedure as quickly as possible; most surgeons performed amputations within only a few minutes. Other methods of pain control were cooling with ice and administering opium and alcohol, until general anesthesia was introduced in the 19th century. Lister’s 1867 discovery of the value of antisepsis had a tremendous impact on the postoperative infection rate and the indications for amputation. Before antisepsis techniques were introduced, almost every limb with a compound fracture was amputated because of gas gangrene. Other milestones in the evolution of war amputation were the development of antibiotics and transfusion therapy, which decreased mortality rates related to amputation from 8% during World War I to 2.5% during the Korean conflict.

Amputation is considered one of the most common surgical procedures (1); more than 100,000 amputations are performed each year in the United States (3). During the first half of the 20th century, military trauma was by far the leading reason for amputation. Since then, peripheral vascular disease and diabetes mellitus have become the most common indications for amputation. Peripheral vascular occlusive disease is responsible for approximately 60% to 70% of extremity amputations, diabetes mellitus for 10% to 20%, and trauma for 10% to 20% (4). Other indications are acute or chronic infections, tumors, nondiabetic neuropathic disorders, congenital anomalies, and iatrogenic complications (i.e., intra-arterial catheterization, extravasation of intravenous drugs, etc.). Most of the information in this chapter addresses lower-extremity amputations because they are much more common than upper-extremity amputations, which account for only 15% of all amputations. The accepted levels of amputation are illustrated in Figure 1.

PREOPERATIVE CONSIDERATIONS Before surgery, the clinician should optimize the patient’s general condition in an effort to increase the chances for uncomplicated healing. The need for emergent amputation may limit preoperative diagnostic and therapeutic efforts. Several conditions may increase the likelihood of complications. As is true with any other surgical procedure, amputation is more likely to be associated with complications when the patient has cardiovascular (e.g., myocardial ischemia, congestive heart failure, or malignant dysrhythmias) or respiratory disease. Conditions that interfere with wound healing, such as diabetes mellitus, malnutrition, or steroid treatment, are associated with a higher rate of wound

424

Klein and Lynn

KEY POINTS &

&

&

&

&

&

&

The most common reasons for amputation are peripheral vascular disease and diabetes mellitus. A delay in amputation after its indication has been determined to be correlated with an increased risk of perioperative complications. The level of amputation substantially affects the likelihood of wound complications and the potential for successful rehabilitation. This decision should be based on clinical judgment and the results of preoperative studies. Meticulous surgical technique should be maintained so that the fitting of a prosthesis can be optimized. Primary closure may shorten healing time, but revision amputation is required in 10% of cases of wound healing failure. Appropriate application of rigid dressings and early physical therapy will minimize the risk of contractures and chronic pain syndrome. Thirty percent of patients will die within two years of the amputation.

infections, dehiscence, and repeated amputation. Patients with chronic renal failure have a higher mortality rate (24% vs. 7%) and a higher complication rate (61% vs. 9%) than patients without renal failure (5). Smoking also has a deleterious effect on this group of patients. The higher rate of infection (42% vs. 22%) and wound dehiscence among active smokers is attributed to the nicotine-induced release of catecholamine, which reduces cutaneous blood flow, and to the procoagulation effect of nicotine, which increases the occurrence of microthrombi (6). Patients should stop smoking at least one week before amputation so that platelet aggregation and fibrinogen levels can return to normal (7). The influence of prior vascular reconstruction on the complication rate associated with amputation is still controversial. Several laboratory abnormalities have been shown to be correlated with increased rates of postoperative complications, especially infection. These abnormalities include hypoalbuminemia, hypocalcemia, hyperglycemia, and azotemia (8). One of the most important preoperative determinants of postoperative complications is the delay between admission and amputation (8). Amputation early in the hospital course will limit suffering (both preoperative and chronic postoperative pain), disorientation, uncontrolled diabetes, and ascending venous thrombosis (9). Minimizing the delay in definitive management will also reduce the risk of wound infection and increase the likelihood of success with the use of more distal amputation sites (10). The only reason to delay a necessary amputation is the assumption that the patient will not survive the surgical stress. In this case, a ‘‘physiological amputation’’ by placing dry ice on the diseased limb should be considered (11). Sepsis caused by an ischemic, infected limb is also a tremendous physiological stressor for the critically ill patient in unstable condition. The decision about the level of amputation is a crucial preoperative step that has a tremendous impact on the rate of complications. Several authors have

identified the choice of the level of amputation as the most important cause of repeated amputation and wound infection (12,13). It is known that a lower level of amputation is better for the outcome of rehabilitation efforts. On the other hand, amputation through compromised tissue without adequate blood supply will lead to an increased rate of wound dehiscence and infection, a longer hospitalization, and an increased incidence of repeated amputation at a higher level. Several invasive and noninvasive methods have been recommended for determining the optimal level of amputation. Doppler ultrasonography, angiography, scintigraphy, and photoplethysmography have all been used to predict healing with varying degrees of success. Clinical judgment based on the findings of physical examination, especially the location of the most distal pulse, is still very useful (14,15), although some authors state that this method is not satisfactory in and of itself (16). Another determinant of the level of amputation is the anticipated success of postoperative rehabilitation efforts. A patient with no chance of being able to use prosthesis because of physiological status (severe heart or lung insufficiency) or neurological condition (stroke with complications and spinal injury with paralysis) should undergo an amputation at the level associated with the lowest risk of postoperative wound complications. Preexisting local limb pathology, such as proximal joint deformities and contractures, will usually prevent the surgeon from performing amputation at a lower level because these conditions will preclude the use of prosthesis.

INTRAOPERATIVE CONSIDERATIONS Technical details of the surgical procedure are outside the scope of this chapter. Nevertheless, a few common operative considerations for avoiding complications must be emphasized. The process of evaluation for the best level of amputation must continue in the

Chapter 35: Complications of Amputations

425

the posterior flap in a below-knee amputation), further shortening of bone may be required, with the risk of compromising future function. General rules of good surgical technique— such as minimizing the use of electrocautery, avoiding extensive debridement of devitalized tissue, and avoiding unnecessary tension during suturing—must be followed religiously because the tissue is already compromised and susceptible to infection. The next decision to be made is whether to close the wound primarily, to leave it to secondary healing, or to use a delayed primary closure. The most important factor in this decision is the clinical circumstances. The failure rate of primary healing in an extremity with vascular compromise is approximately 20% (16). Nevertheless, primary closure, when successful, is associated with the shortest healing time and hospitalization time. The rehabilitation process may also be facilitated by primary closure (17). The advantages of primary closure must be weighed against the 10% rate of stump revision in the case of failure (18). Even in the case of amputation due to mine explosion, which was traditionally treated with an open technique for secondary healing, the success rates are as high as 87% for primary closure (19) and 84% for delayed primary closure (20). Placing a drain in a stump that has been closed primarily is still a controversial practice, although it is commonly performed in an effort to avoid retained hematoma, which may become secondarily infected. Flexion contracture of the knee after a below-knee amputation is best avoided by using a rigid plaster dressing or a posterior plaster splint.

POSTOPERATIVE COMPLICATIONS

Figure 1 Accepted levels of amputation.

operating room. If the skin and muscle do not seem to bleed or to contract to electrical stimuli, as would be expected of viable tissue, the level of amputation must be reconsidered and amputation should probably be performed at a higher level. The future interface between stump and prosthesis should be considered when skin, muscle, and bone are excised. An excessively long fibula, inadequate beveling of the tibia, or an excessively long femur will interfere with the comfortable fit of the prosthesis, as will excessive soft tissue at the end of the stump. Although a more distal amputation offers superior potential for rehabilitation, an excessively long stump below the knee will cause alignment problems, impaired cosmesis, and discomfort with no functional advantage. On the other hand, if the remaining soft tissue and muscle are inadequate to cover the bony stump (especially

The average amputee is older and has several comorbid conditions, especially advanced atherosclerosis; therefore, the likelihood of potential complications after amputation is higher than that for younger, healthier patients who undergo other surgical procedures. Although amputation surgery is usually perceived as a failure, its goal is to improve the patient’s quality of life, and it may sometimes be lifesaving. Any postoperative complication experienced by amputees may decrease the chances of successful rehabilitation. Postoperative complications may be categorized as those related to the operative wound, those related to the prosthesis, and those related to pain syndromes.

Wound Complications Wound infection and wound breakdown are the most common complications after amputation surgery. The incidence of wound infection is directly related to the indication for amputation. The general incidence of wound infection is 10% to 30% but may be as high as 60% when the limb is infected (21). The use of prophylactic antibiotics is routine at most centers.

426

Klein and Lynn

Infections caused by gram-positive or gram-negative bacteria are usually treated with second- or thirdgeneration cephalosporins; coverage for anaerobes is added for patients with diabetes. The use of prophylactic antibiotics has been found to reduce the rates of wound infection and repeated amputation by as much as 50% (22). The question of whether antibiotics can adequately penetrate ischemic tissue has been answered by several studies that showed sufficient levels of antibiotics in the stump wound (22,23). In fact, in cases in which the level of antibiotics in the stump was not adequate, the rate of failed stump and the necessity for repeated amputation was much higher that for cases in which the antibiotic level was sufficient. This fact suggests that the problem was an incorrect choice of amputation level rather than antibiotic bioavailability (25). In cases of preexisting infection, antibiotic treatment should be guided by the specific preoperative cultures. If cultures are not available, treatment with broad-spectrum antibiotics against gram-positive, gram-negative, and anaerobic bacteria should be initiated. In severe cases of purulent infection, surgical drainage of the infected collection should be considered before the amputation is performed. In extreme situations, in which severe infection requires an emergent amputation, a guillotine procedure may be performed as a lifesaving procedure. Failure to heal is another common wound complication after amputation, especially among patients with a devascularized limb. In this situation, primary healing can be anticipated in only 20% to 50% of cases, and as many as 35% of those cases will require secondary amputation at a higher level (16,18,26,27). The most common reason for failure to heal is low oxygen delivery to the wound (28), primarily because of primary vascular insufficiency related to an incorrect choice of the level of amputation. The more distal the amputation level, the more limited the oxygen delivery to the tissues, and the greater the likelihood of wound breakdown (29). Other reasons for healing failure are increased pressure in the wound caused by undrained hematoma, compromised venous or lymphatic drainage, or sutures that are too tight, infection, immunosuppressive medications such as corticosteroids and metabolic disorders such as hypoalbuminemia or malnutrition.

amputation (8), the most common variable measured for determining quality of life is independent ambulation. Successful fitting of prosthesis should be expected after 75% of below-knee amputations and after 50% of above-knee amputations (31). More than 60% of those patients will still be able to use prosthesis three years after amputation (32). The main reason for failure in fitting a primary prosthesis is a technical problem related to inadequate shaping of the distal stump during the operation. Another reason is flexion contracture of the proximal joint, which is usually the result of a technical error (i.e., leaving a femur stump that is too short), failure to maintain the proper postoperative position of the stump, or failure to provide adequate physical therapy. Flexion contracture of more than 9 at the hip or 15 at the knee is considered a contraindication for the use of prosthesis. Inability to use the prosthesis after a successful primary fitting is usually due to stump shrinkage. Less common complications that might interfere with prosthesis use are calluses over weight-bearing bony prominences (usually because of the lack of soft-tissue coverage), bone spur of the distal stump, bone overgrowth in children, and osteoporosis of the stump bone caused by disuse. The combination of pressure and friction within the prosthesis socket can cause skin complications such as epidermoid cysts, which might become infected and are very likely to recur, even after a surgical removal. The most dramatic skin complication is verrucous hyperplasia. Proximal constriction in the socket without adequate distal tissue support causes extreme skin edema with wart-like appearance and occasional exudation. Surgical reshaping of the distal stump will gradually correct this problem. The general health of the patient can also affect independent ambulation. Cardiac, lung, or neurological pathology can result in inability to use the prosthesis because the patient cannot endure the excessive physical strain. A patient who has undergone unilateral below-knee amputation will require 10% more energy for ambulation than will a man of the same age who has not undergone amputation. The energy requirement for ambulation is increased by 50% for unilateral above-knee amputation and by 200% for bilateral above-knee amputation (33).

Problems with Function or with the Prosthesis

Pain

When evaluating the outcome of amputations, one must remember the general physical condition of the patient who requires amputation. Amputation is usually necessitated by diffuse atherosclerosis or poorly controlled diabetes. In 40% to 50% of cases, the patient will require a contralateral amputation within two years (8,30). Functional outcome after amputation depends on certain commonly measured variables. Although some series identify ability to independently perform activities of daily living as the main factor contributing to quality of life after

Although amputation may be the treatment for severe chronic and ischemic pain, several types of pain syndromes can affect the patient after amputation. The first and most common of these is immediate postoperative pain, which is no different from the pain that follows any other major surgical procedure. Usually, this pain is nonspecific and subsides during the first postoperative week. Carefully applied rigid dressings may decrease this pain by reducing local edema. Neuroma of the stump can cause severe localized pain during the weeks or months after

Chapter 35: Complications of Amputations

amputation. Neuroma is a universal nerve repair phenomenon and cannot be prevented. Usually, neuroma does not cause discomfort unless the area is subjected to excessive pressure or is not covered or supported by sufficient soft tissue. The pain is deep and dull in nature and may be resistant to conventional analgesic therapy (34). The best strategy for minimizing the clinical manifestations of neuroma is cutting the nerve deep in the muscle tissue during the primary operation. The tradition of ligating a mass of major nerves is of no value because it will not prevent the formation of neuroma. When amputations are performed after trauma, imbalanced sympathetic tone can cause reflex sympathetic dystrophy. This syndrome, originally termed causalgia, may cause burning pain along with mottled, cool skin and osteopenic bone in the stump. Lumbar sympathectomy for lower-extremity amputation and thoracic sympathectomy for upperextremity amputation may offer some relief, but the success rate varies. The most common postamputation pain syndrome is phantom pain. Almost all amputees experience phantom limb sensation, which is the feeling that the missing limb is still present. Some patients will experience phantom pain, which is a poorly localized pain that is burning, cramping, aching, or stabbing in nature (35). Three major characteristics define this syndrome: (i) pain that lasts for some time after the wound has healed; (ii) pain elicited by the activation of trigger zones; and (iii) pain that resembles the pain experienced preoperatively. Changes in somatic input, such as massaging the stump or treating it with ultrasound, may affect this pain (36). The likelihood of phantom pain is correlated with the site of the amputation; the higher the level of amputation, the more likely the phantom pain. Phantom pain is more likely after upperextremity amputations than after lower-extremity amputations. No single therapy or combination of therapies, medical or surgical, has proved successful in treating this disabling syndrome. The incidence of severe stump pain or phantom pain is reported to range from 30% to 85% (35,37,38), but some authors report that the likelihood of this disabling postamputation complication can be reduced to less than 5% by an early, aggressive rehabilitation program (39).

MORTALITY The high mortality rate after major amputation reflects the fact that most patients who require amputation suffer from multiple diseases. The estimation is that one-third of patients who undergo lower-limb amputation will die by the second postoperative year and two-thirds will die within five years (40). The

427

immediate operative and postoperative mortality rates range from 10% to 30%. The level of amputation is highly correlated with the mortality rate. The lower the level of amputation, the lower the mortality rate. Recent studies showed mortality rates of 2% for below-knee amputation and 9% for above-knee amputation (41,42). The mortality rate after bilateral amputation is 18%, and the mortality rate after above-knee amputation that followed a failed belowknee amputation is 21% (5). The reason that the mortality rate is higher after a higher-level amputation is probably related to the fact that these patients generally have more severe general atherosclerosis and an increased rate of cardiac complications. The amputation associated with the highest mortality rate (44%) is hip disarticulation. The mortality rate associated with this procedure is as high as 60% when it is required after trauma, probably because the victim has sustained a devastating amount of energy (43).

SUMMARY Since the end of the global military conflicts that characterized the 20th century, the most common indications for amputation have been peripheral vascular disease and diabetes. Patients undergoing amputation usually suffer from diffuse atherosclerosis or diabetes and from the end-organ dysfunction that accompanies these diseases. Thus, a high rate of postoperative complications can be anticipated. The surgeon must often make a difficult choice between the procedure that provides the best chance for rehabilitation and the procedure that carries the lowest complication rate. A judicious choice of the level of amputation, together with meticulous surgical technique and aggressive postoperative physical therapy, will give the patient the best chance for timely and successful rehabilitation. No other surgical procedure better illustrates the important rule of surgery: it is easier to prevent a complication than to treat it.

REFERENCES 1. Johns CHW. The Code of Hammurabi. 11th ed. Encyclopaedia Britannica, 1910–1911. 2. Wheeler HB. Myth and reality in general surgery. Bull Am Coll Surg 1993; 78:21–27. 3. Krupski WC, Nehler MR. Amputation. In: Way LW, Doherty GM, eds. Current Surgical Diagnosis and Treatment. 11th ed. Boston: McGraw-Hill, 2002:859–870. 4. Hansson J. The leg amputee. A clinical follow-up study. Acta Orthop Scand 1964; 10(suppl 69):1–104. 5. Dossa CD, Shepard AD, Amos AM, et al. Results of lower extremity amputations in patients with end-stage renal disease. J Vasc Surg 1994; 20:14–19. 6. Lind J, Kramhoft M, Bodtker S. The influence of smoking on complications after primary amputations of the lower extremity. Clin Orthop 1991; 267:211–217. 7. Cryer PE, Haymond MW, Santiago JV, Shah SD. Norepinephrine and epinephrine release and adrenergic

428

8. 9.

10. 11. 12. 13. 14. 15. 16.

17.

18. 19.

20. 21. 22.

23.

24.

Klein and Lynn

mediation of smoking-associated hemodynamic and metabolic events. N Engl J Med 1976; 295:573–577. Weiss GN, Gorton TA, Read RC, Neal LA. Outcomes of lower extremity amputations. J Am Geriatr Soc 1990; 38:877–883. Rosenberg N, Adiarte E, Bujdoso L, Backwinkel KD. Mortality factors in major limb amputations for vascular disease: a study of 176 procedures. Surgery 1970; 67:437–441. Bunt TJ, Manship LL, Bynoe RP, Haynes JL. Lower extremity amputation for peripheral vascular disease. A low-risk operation. Am Surg 1984; 50:580–584. Bunt TJ. Physiologic amputation for acute pedal sepsis. Am Surg 1990; 56(9):530–532. Harris JP, Page S, Englund R, May J. Is the outlook for the vascular amputee improved by striving to preserve the knee? J Cardiovasc Surg (Torino) 1988; 29:741–745. Ebskov B, Josephsen P. Incidence of reamputation and death after gangrene of the lower extremity. Prosthet Orthot Int 1980; 4:77–80. O’Dwyer KJ, Edwards MH. The association between lowest palpable pulse and wound healing in below knee amputations. Ann R Coll Surg Engl 1985; 67:232–234. Clyne CA. Selection of level for lower limb amputation in patients with severe peripheral vascular disease. Ann R Coll Surg Engl 1991; 73:148–151. van den Broek TA, Dwars BJ, Rauwerda JA, Bakker FC. A multivariate analysis of determinants of wound healing in patients after amputation for peripheral vascular disease. Eur J Vasc Surg 1990; 4:291–295. Aligne C, Farcot M, Favre JP, Alnashawati G, De Simone F, Barral X. Primary closure of below-knee amputation stumps: a prospective study of sixty-two cases. Ann Vasc Surg 1990; 4:143–146. Senkowsky J, Money MK, Kerstein MD. Lower extremity amputation: open versus closed. Angiology 1990; 41:221–227. Atesalp AS, Erler K, Gur E, Solakoglu C. Below-knee amputations as a result of land-mine injuries: comparison of primary closure versus delayed primary closure. J Trauma 1999; 47:724–727. Simper LB. Below knee amputation in war surgery: a review of 111 amputations with delayed primary closure. J Trauma 1993; 34:96–98. Desai Y, Robbs JV, Keenan JP. Staged below-knee amputations for septic peripheral lesions due to ischemia. Br J Surg 1986; 73:392–394. Norlin R, Fryden A, Nilsson L, Ansehn S. Short-term cefotaxime prophylaxis reduces the failure rate in lower limb amputations. Acta Orthop Scand 1990; 61:460–462. Kerin MJ, Greenstein D, Chisholm EM, Sheehan SJ, Kester RC. Is antiobiotic penetration compromised in the ischaemic tissues of patients undergoing amputation? Ann R Coll Surg Engl 1992; 74:274–276. Sonne-Holm S, Boeckstyns M, Menck H, et al. Prophylactic antibiotics in amputation of the lower extremity for ischemia. A placebo-controlled, randomized trial of cefoxitin. J Bone Joint Surg Am 1985; 67:800–803.

25. Mars M, Elson KI, Salisbury RT, Robbs JV. Do preoperative antibiotics reach the operative field in amputation surgery for peripheral vascular disease? A pilot study. S Afr J Surg 1990; 28:58–61. 26. Finch DR, Macdougal M, Tibbs DJ, Morris PJ. Amputation for vascular disease: the experience of a peripheral vascular unit. Br J Surg 1980; 67:233–237. 27. Keagy BA, Scwartz JA, Kotb M, Burnham SJ, Johnson G Jr. Lower extremity amputation: the control series. J Vasc Surg 1986; 4:321–326. 28. Kuhne HH, Ullmann U, Kuhne FW. New aspects on the pathophysiology of wound infection and wound healing—the problem of lowered oxygen pressure in the tissue. Infection 1985; 13:52–56. 29. Burgess EM, Matsen FA III, Wyss CR, Simmons CW. Segmental transcutaneous measurements of PO2 in patients requiring below-the-knee amputation for peripheral vascular insufficiency. J Bone Joint Surg Am 1982; 64:378–382. 30. Bodily RC, Burgess EM. Contralateral limb and patient survival after leg amputation. Am J Surg 1983; 146: 280–282. 31. Steinberg FU, Sunwoo I, Roettger F. Prosthetic rehabilitation of geriatric amputee patients: a follow-up study. Arch Phys Med Rehabil 1985; 66:742–745. 32. De Luccia N, Pinto MA, Guedes JP, Albers MT. Rehabilitation after amputation for vascular disease: a followup study. Prosthet Orthot Int 1992; 16:124–128. 33. Huang CT, Jackson JR, Moore NB, et al. Amputation: energy cost of ambulation. Arch Phys Med Rehabil 1979; 60:18–24. 34. Fisher GT, Boswick JA Jr. Neuroma formation following digital amputations. J Trauma 1983; 23:136–142. 35. Abramson AS, Feibel A. The phantom phenomenon: its use and disuse. Bull N Y Acad Med 1981; 57: 99–112. 36. Melzack R. Phantom limb pain: implications for treatment of pathological pain. Anesthesiology 1971; 35: 409–419. 37. Sherman RA. Published treatment of phantom pain. Am J Phys Med 1980; 59:232–244. 38. Sherman RA, Sherman CJ, Parker L. Chronic phantom and stump pain among American veterans: results of a survey. Pain 1984; 18:83–95. 39. Malone JM, Moore WS, Leal JM, Childers SJ. Rehabilitation for lower extremity amputation. Arch Surg 1981; 116:93–98. 40. Potts JR III, Wendelken JR, Elkins RC, Peyton MD. Lower extremity amputation: review of 110 cases. Am J Surg 1979; 138:924–928. 41. Rush DS, Huston, CC, Bivins BA, Hyde GL. Operative and late mortality rates of above-knee and below-knee amputations. Am Surg 1981; 47:36–39. 42. Kald A, Carlsson R, Nilsson E. Major amputation in a defined population: incidence, mortality and results of treatment. Br J Surg 1989; 76:308–310. 43. Endean ED, Scwarcz TH, Barker DE, Munfakh NA, Wilson-Neely R, Hyde GL. Hip disarticulation: factors affecting outcome. J Vasc Surg 1991; 14:398–404.

36 Complications of Hand Surgery Charles Eaton The Hand Center, Jupiter, Florida, U.S.A.

The primary goal of treating hand trauma is simply to avoid complications. In this chapter, complications are grouped as complications of missed diagnoses, complications of treatment, and complications of injuries.

COMPLICATIONS OF MISSED DIAGNOSES Missed Hand Injuries The best insurance against missed diagnoses related to hand injuries is eliciting an adequate history and performing a thorough physical examination.

History Because severe upper extremity injuries are frequently dramatic and are attended by emotional factors, it is usually best to elicit a history in a deliberate, orderly way. If possible, after hearing the story, the examiner should physically demonstrate the scenario of injury so that the patient can confirm the examiner’s understanding of the details, including the position of the extremity at the time of injury. If an injury involves machinery, the machinery should be described in enough detail to allow the examiner to visualize it. In simple mechanical terms, was the mechanism sharp or dull? Did it involve rotating blades, belts, or chains; heat, cold, or chemicals? Did the patient land on the palm with the wrist extended or on the dorsum of the wrist? Was the patient able to pull the hand out or was it trapped, requiring extrication? Was the bleeding pulsatile? One should not assume that all problems with the hand resulted from a single reported injury. Did the pain or numbness start immediately after the event or later? Has the hand been injured previously? Did these injuries occur recently or a long time ago? Were there prior problems with numbness, weakness, or pain? Attention to such details from the onset can avoid misguided treatment and false expectations. Additionally, hand injuries are a common starting point for personal injury litigation, and clear initial documentation of these points will

prevent needless difficulties later when an attorney becomes involved.

Physical Examination A working knowledge of anatomy usually allows assessment of an acute injury without touching the obvious site of injury. Sensory, motor, and vascular function distal to the injury can provide clues about the status of more proximal wounds. This gentle approach is clearly preferable to attempting to define the injury by probing or applying an instrument to a wound in the emergency room. Rapid Survey of the Hand

A focused, informative survey of the injured hand can be performed in approximately one minute. It is best to use a systems checklist (Table 1) when examining an injured hand. Such a checklist will review both objective and subjective findings. A focused examination of the median, ulnar, and radial nerves can be performed within a few seconds (Fig. 1). Indirect tenderness or pain that occurs when gentle percussion, traction, torsion, or bending stress is applied to the skeleton at a distance from the area of injury will occur when there is pathologic skeletal micromotion caused by fracture or ligament injury. Active unresisted motion may be limited but can confirm that tendons are in continuity and that innervation of the proximal muscles is intact.

Table 1 Subjective and Objective Findings Obtained by a Rapid Survey of the Hand Objective findings Skin: wounds, texture, turgor Vasculature: color, temperature, turgor, capillary refill, pulses Bone and joint: deformity, instability Muscle and tendon: posture, compartment turgor Subjective findings Perception of injury: pain, tenderness, apprehension of pain, weakness Peripheral nerve function: nerve-specific sensory islands; muscles with unique innervation Skeleton: tenderness at the site of injury Muscle and tendon: strength

430

Eaton

Figure 2 Allen’s type of testing may be performed on each digit to confirm patency of each digital artery. The finger is exsanguinated with distal to proximal massage (A, B, C), and then one of the digital arteries is released (D) and pattern of refill observed. This can be helpful as an adjunct test for evaluation of possible digital nerve injuries, because digital nerves are superficial to digital arteries and digital artery injury from a penetrating wound has a high likelihood of associated digital nerve injury.

Figure 1 Rapid assessment of the hand includes color, temperature, range of motion, tenderness convergence of fingertips toward the distal pole of the scaphoid to assess fracture malrotation, sensory and motor assessment of the median (radial index fingertip/abductor pollicis brevis), ulnar (ulnar small fingertip/first dorsal interosseous), and radial (dorsal first web space/extensor pollicis longus) nerves.

used to check relative finger posture during passive wrist flexion and extension (Figs. 4 and 5). Squeezing the mid-forearm will tighten the finger flexor tendons and mimic their active action. Rotation of the fingers may be suspected if the tips overlap, but if the fingertips are not adjacent during flexion, it is normal for all of them to converge toward the distal pole of the scaphoid, where the flexor carpi radialis tendon intersects the wrist flexion crease (Figs. 6 and 7). Contour abnormalities at joints or along long bones may indicate fractures or dislocations. Common contour changes resulting from displaced fractures include those due to distal radius fractures (Fig. 8), metacarpal neck fractures, and proximal phalanx fractures. Metacarpophalangeal or proximal interphalangeal joint dislocations alter flexor–extensor tendon tension balance and may cause unusual posture or positioning of the joints distal to the injury (Fig. 9). Bruising

Bone and Joint Assessment. Tips for Examining the Hand of an Unconscious Patient

Much information can be obtained in the absence of subjective findings by using the four categories of assessment listed below. Vascular Assessment. Comparing the color of the skin and nail beds of the injured hand to the color of those sites on the other hand can indicate arterial insufficiency (if the injured hand is paler) or venous insufficiency (if the injured hand is dark or purple). Allen’s test can be performed without patient participation by gently squeezing the palm while occluding the radial and ulnar arteries at the wrist, and then releasing one artery to assess the patency of the two main arteries and of the palmar arch. The digital Allen’s test is performed similarly; the examiner uses his or her fingertips to exsanguinate a finger from distal to proximal points and then releases one or the other side at the base of the finger (Fig. 2). Forearm compartment pressures can be measured with commercial kits or with materials available in any emergency room.

The posture of the fingers can indicate specific tendon injuries. Even if the patient is unconscious, if the tendons and phalanges are intact, the fingers should assume a cascade position of progressively more flexion of both proximal and distal interphalangeal joints, from the index to the small finger (Figs. 1 and 3). Tenodesis-induced motion of the fingers can be

Muscle and Tendon Assessment.

Figure 3 Finger posture changes following flexor tendon injury. Fingers at rest follow a natural flexion cascade, which is altered when either one or both flexor tendons have been cut. These patients show characteristic abnormal straightening of digits () due to proximal flexor tendon injuries (arrows): small finger superficialis (A), both tendons of ring finger (B), index profundus tendon (C), all tendons of all fingers, thumb spared (D).

Chapter 36:

Figure 4 Wrist tenodesis normally results in passive changes in finger posture: finger extension when the wrist is flexed, and finger flexion when the wrist is extended.

at a site away from an area of impact, such as bruising of the dorsal wrist after a fall on the outstretched palm, strongly suggests an underlying skeletal injury even when radiographic results are normal. Passive range-of-motion examination of the elbow, wrist, and fingers can be used to assess crepitation (which indicates an injury to the joint surface), resistance (which indicates swelling, subluxation, or dislocation), and instability (which indicates ligament injury).

Figure 5 This unconscious patient had a dorsal forearm laceration (straight arrow), which appeared superficial, but wrist tenodesis (curved arrows) demonstrated injury to the extensor tendons of the middle, ring, and small fingers (asterisks), which had incomplete extension with passive wrist flexion. This was confirmed in the operating room. Because of the obliquity of the laceration, the injury could not have been demonstrated by simple wound exploration in the emergency room.

Complications of Hand Surgery

431

Figure 6 In flexion, the fingertips converge toward a common point in the proximal palm, approximately the distal pole of the scaphoid (X). This can be helpful in quickly assessing for rotational malalignment of a metacarpal or phalangeal fracture in an unconscious patient.

The digital nerves are superficial to the digital arteries. Thus, an abnormal finding on a digital Allen test (Fig. 2) in the context of any palmar finger laceration strongly suggests an associated digital nerve injury because the zone of external injury must pass through the nerve before reaching the artery. Tactile adherence is assessed by sliding an object with a smooth surface across the palmar skin. A smooth surface such as a glass slide or the barrel of a shiny, smooth plastic pen will slide with much less resistance (adherence) over skin affected by nerve injury than over normal skin because recently denervated skin does not sweat. Normally, microscopic sweat droplets on the palmar skin create resistance to this motion. The wrinkle test makes use of the finding that recently denervated skin does not wrinkle upon prolonged

Nerve Assessment.

Figure 7 Rotational malalignment of metacarpal fractures may not be apparent with the fingers in extension. The rotation of this patient’s ring and small finger malunions is much less obvious with his fingers extended (left) than with attempted full flexion (right). Reduction of rotational malalignment should be checked in the acute setting by placing the fingers in full flexion.

432

Eaton

Figure 8 Acute deformity. The angulation of this fracture is obvious even in an unconscious patient. The fracture (above); The X–ray (below).

contact with water. In this test, the fingers are immersed in water (not in saline or another salt solution) for five minutes and are then inspected for wrinkling. The absence of wrinkling indicates denervation (Fig. 10). The mechanism behind this phenomenon is unknown. Missed Problems Elsewhere A dramatic hand injury can divert attention from a standard trauma systems evaluation. Life-threatening complications resulting from missed injuries are most

Figure 10 The wrinkle test may be used to assess sensory nerve function in the unreliable patient. For a period of months after nerve injury, denervated skin loses its ability to wrinkle when immersed in water. Here, failure of the small finger skin to wrinkle when immersed in water for 10 minutes (compared to normal adjacent ring finger) strongly suggests injury of both digital nerves of the small finger.

likely when a patient has sustained a traumatic amputation in a blunt trauma scenario, such as a traffic accident or a fall. Life-threatening central nervous system or thoracoabdominal injuries may be missed, as may proximal skeletal and brachial plexus injuries. An occult medical condition commonly accompanying hand injury is substance abuse; in one report, nearly half of patients requiring emergency room treatment for hand trauma tested positive for alcohol or another substance (1).

COMPLICATIONS OF TREATMENT The most common complication of any hand injury is stiffness, which results from the collaborative effects of inflammation, swelling, and immobility. Attempts at preventing stiffness are much more effective and worthwhile than later attempts at correcting established stiffness. Stiffness and other complications are less likely when the treatment follows priority-based guidelines.

Priorities

Figure 9 Complex dislocations of the thumb metacarpophalangeal joint present with an extended posture of the thumb, and occasionally a dimple or bruising at the palmar metacarpal head level. X-rays may show sesamoid interposition as is seen here. Such a finding increases the likelihood that open reduction will be necessary.

Management priorities are the same for severe and minor injuries: establish the extent of injury, remove the bad, reconstruct the good, involve the patient, and tailor the surgery to the patient (63). Severe upper extremity injuries with soft tissue loss result in shorter hospitalization periods and more rapid recovery when primary reconstruction is performed, even if such treatment requires primary microvascular freeflap surgery (76). One conceptual approach to organizing the initial management of severe hand injuries is to break down priorities as they relate to either healing or function.

Chapter 36:

Healing Priorities: Circulation, Skeleton, Closure Inadequate blood supply is the single most likely explanation for complications of delayed healing, fibrosis, and infection. Adequate blood supply is achieved by aggressive debridement, revascularization, and the use of vascularized flaps. Edema indicates inadequate lymphatic circulation and has the same ultimate effects as inadequate blood supply. Edema is best treated with elevation and active range-ofmotion exercises, when possible. Optimum bone and joint reconstruction goals are prompt anatomic reduction of injury and stable skeletal fixation with the least amount of additional soft-tissue disruption. Wounds should be closed and covered with mobile, well-vascularized soft tissue as quickly as possible. In the hand, stiffness, difficulty with use, and ultimate disability are directly related to the length of time required for wound healing.

Functional Priorities: Nerve, Joint, Muscle Nerve injuries should be approached aggressively because there is never a better time to evaluate injuries and to perform repairs, and the only satisfactory time at which partial nerve lacerations can be repaired is in the acute setting (Figs. 11 and 12). Passive rangeof-motion exercises have two purposes. The first purpose is preservation of the gliding function of the surfaces of the joints and tendons. This is achieved by early protected motion: all moving parts that can be moved safely are moved frequently, against no resistance and at the earliest opportunity. The second

Complications of Hand Surgery

433

purpose is maintenance of the physiological length of capsuloligamentous and muscular tissues. This is achieved by splinting the hand between exercises in the ‘‘protective position’’: interphalangeal joint extension, metacarpophalangeal joint flexion, and preservation of the thumb–index finger web space span (Fig. 13). Active range-of-motion exercises have the additional benefits of reducing edema, building strength, promoting bone healing, preventing dysfunctional patterns of disuse, and probably reducing the incidence of complex regional pain syndrome.

Complications of Bandaging Tight Dressings Finger dressings made from tubular gauze may produce ischemic pressure complications. Technical errors in application that can result in tubular gauze pressure complications include excessive longitudinal traction during application, a twist of more than a 90 during application, and rolled proximal edges of the dressing (21). Even minimally tight elastic dressings applied as part of a circumferential bandage may lead to progressive swelling, thereby aggravating all of the previously described ill effects of swelling of the injured hand. Swelling may hinder assessment and may delay surgery until it has been reduced by elevation and changed to a noncompressive dressing. A useful technique is to place multiple layers of circumferential gauze as the deepest portion of the bandage and then split them longitudinally before completing the bandage. This technique ensures that at least the deepest layer of bandage cannot exert circumferential pressure. Tight casts may result in local pressure sores, discomfort, and, in the worst scenario, vascular compromise and compartment syndrome. The risk of complications is highest when circumferential casts are applied after closed reduction of an elbow or forearm fracture on the day of injury. In this situation, the risk may be reduced by splitting (‘‘bivalving’’) the cast into two separate longitudinal sections immediately after application.

Inadequate Positioning

Figure 11 Forearm laceration (1A) with complete ulnar artery and partial ulnar nerve laceration (1B). Partial forearm posterior interosseous nerve laceration (2A) with primary fascicular repair (2B). (C) Local crush injury of proximal median nerve with central disruption, epineurium intact; excellent recovery after primary debridement and epineural repair. Exploration of digital (D) and median (E) neuromas late after partial injuries. When partial nerve function remains, resection and grafting is an unpredictable solution, and primary repair is preferable for partial nerve injuries.

Splints and other supportive dressings maintain a posture that may be helpful or detrimental. Often, splints fabricated for comfort in the emergency room restrain joints in positions that promote stiffness. Even splints intended to maintain the generic ‘‘protective position’’ may actually do just the opposite, a problem that may be confirmed only by radiography (Fig. 14).

Complications of Wound Care The goal of wound care is to maintain an environment that discourages excessive bacterial growth and encourages normal healing. Excessive bacterial growth occurs on moist, undisturbed surfaces and is a common problem in the interdigital web spaces of the

434

Eaton

Figure 12 Complications of foreign bodies. (A, B, C) A close-range shotgun blast of the palm, treated with local wound care, resulted in median neuropathy and finger stiffness. At exploration for reconstruction a year later, the entire plastic shotgun casing was found embedded in the proximal carpal tunnel. (D) Wood impaling the index finger ulnar digital nerve one month following injury. (E) Wood splinters lodged in the flexor tendon sheath. (F) Intraoperative fluoroscopy is essential for removal of radiopaque foreign bodies. The small piece of shrapnel lodged within the finger flexor tendon moved over a 1 in. excursion with finger motion.

immobilized hand and beneath occlusive bandages. Eventually, unchecked surface growth produces such high concentrations of organisms that the skin is invaded directly, producing maceration dermatitis. This condition may progress to cellulitis, but it can be stopped in the early stages by increasing the frequency of dressing changes and, when possible,

allowing the affected skin to dry. Allergic contact dermatitis may develop over the course of treatment with topical antibiotics or skin preparation formulas such as Mastisol1 (Fernade Laboratories, Inc., Michigan, U.S.A.) (Fig. 15). This complication can produce a confusing picture because inflammation associated with the reaction may be confused with infection.

Figure 13 The generic ‘‘safe position’’ for hand immobilization is intended to prevent the usual pattern of stiffness after hand disuse. The three features are maintenance of interphalangeal joint extension (1), metacarpophalangeal joint flexion (2), and first web-space span (3).

Figure 14 Safe hand position is sometimes difficult to maintain with splinting. The thickly padded palmar plaster splint applied in the emergency room is usually totally inadequate (left), but even a cast that appears correct may not be if the palmar flexion point is constructed distal to the center of the palm (right).

Chapter 36:

Complications of Hand Surgery

435

20 gauge) are used (45), during prolonged periods of cannulation, and when hypercoagulability is present (42). In the presence of ulnar artery occlusion, even a single radial artery needlestick for arterial blood gas determination, can precipitate acute hand ischemia (Fig. 16). Although uncommon, ischemia resulting in finger amputation has been reported after arterial monitoring of infants (60). Cutaneous Nerve Injury

Figure 15 Allergic contact dermatitis. Sensitivity to topical agents may develop after an initial exposure of 7 to 14 days, or earlier in patients who had been previously sensitized. (Left) Dermatitis developing from bacitracin was used to treat a superficial burn. As is frequently the case in application of topical antibiotics to treat or prevent infection, the allergic reaction was confused with infection, prompting further use of the offending agent. (Right) [Mastisol1 (Ferndale Laboratories, Inc., Michigan, U.S.A.)] was used to improve the attachment of surgical skin closure tape. Appearance seven days after surgery. Symptoms from severe reactions are best controlled with short-course high-dose systemic steroids.

The cephalic vein is frequently cannulated for intravenous access. It is closely related to the antebrachial cutaneous nerve in the proximal forearm and the branches of the superficial radial nerve in the distal forearm. Needlestick injuries can affect either of these nerves (33) and can lead to prolonged morbidity, although this complication is uncommon. Patients report a strong electrical paresthesia at the time of injury; this symptom should be taken as a sign of possible injury. Numbness or tingling lasting more than a day may indicate partial nerve injury and should lead to consideration of early exploration. Treatment options for chronic cases are the same as for any cutaneous neuroma and lingering symptoms are common. Extravasation Injuries

The early hallmarks of contact dermatitis are itching and tiny blisters accompanying the reaction.

Complications of Hand Procedures Tourniquet Palsy Tourniquet palsy occurs postoperatively in an average of 1 in 5000 cases; it is more commonly associated with microsurgical procedures than with other procedures (22). All nerves are usually affected to some degree, but the radial nerve is usually the most affected. Tourniquet palsy is more likely among patients with coagulation disorders or preexisting neuropathy, those who are thin and malnourished, those with systemic lupus erythematosus (22), and those with unintentionally high tourniquet pressures due to gauge failure (36).

Toxic Shock Syndrome Toxic shock syndrome is a rare complication but has been reported after elective reconstructive hand surgery (32).

Needlestick or Vascular Cannulation Injuries Radial Artery Catheterization

Acute ischemia of the hand may occur after radial artery catheterization if there is inadequate perfusion through the ulnar artery (42). This problem is more likely when ulnar artery perfusion is not confirmed by Allen’s test before catheterization, when cannulas of relatively large diameter (18 gauge as opposed to

Extravasation injuries (96–102) of the hand are common because the hand is commonly used for intravenous access. Local tissue necrosis has been reported after subcutaneous extravasation of chemotherapeutic agents, osmotically active substances, and tissue-toxic preparations such as injectable phenytoin. These injuries often exhibit delayed presentation, delayed healing, and prolonged morbidity, requiring reconstructive surgery if treated late. Serious limb growth disturbances may occur after extravasation or thrombosis in the neonatal period. Although not well described in the literature, tense hematomas associated with intravenous access of the wrist or dorsal hand may also result in tissue loss (Fig. 17). The outcome of extravasation injuries is best when they are recognized and treated early. Unfortunately, delayed presentation is still common because of the typically slow development of visible signs of injury. Treatment recommendations have varied over the years, but early treatment with soft-tissue infiltration and irrigation has the most consistent history of effectiveness. Local injection with hyaluronidase is helpful, but this drug is no longer available for use. Prevention appears to be the best approach; avoid the dorsal hand, anterior wrist, and antecubital fossa for infusion of tissue-toxic solutions because these locations are most likely to develop complications associated with extravasation.

Prior Axillary Lymphadenectomy Although it is common practice to instruct patients who have undergone mastectomy and axillary dissection to

436

Eaton

Figure 16 Iatrogenic vascular problems. Hand ischemia due to radial artery thrombus from blood gas puncture in a patient with undiagnosed ulnar artery occlusion (A, B). Median nerve compression from dialysis shunt crossing directly over the median nerve (C).

avoid manipulation or instrumentation of the hand, the risk of complications in this context has not been documented (88). Hand surgery ipsilateral to previous axillary dissection is probably safe.

Complications of Anesthesia Epinephrine in Digital Block Although traditional wisdom holds that the use of epinephrine in digital nerve blocks may result in digital gangrene, there are no reported cases of finger gangrene resulting specifically from the use of epinephrine with lidocaine for digital block, and the safe use of epinephrine has been reported (12).

Postoperative Ulnar Nerve Palsy Palsy of the ulnar nerve due to ulnar neuropathy at the level of the elbow is a recognized but poorly understood complication of surgery involving general anesthesia (15). The exact mechanism of this process remains unknown. Preventative measures—including protective positioning of the arm on the operative table, use of elbow pads, and avoidance of abduction, pronation, and elbow flexion—may reduce the incidence of this complication. Final outcome is unpredictable; both conservative and operative treatments have yielded mixed results.

Chapter 36:

Complications of Hand Surgery

437

Figure 17 Tense dorsal hand hematomas may result in necrosis of the overlying skin. This problem occurs most often in the context of anticoagulation. Emergent surgical evacuation may limit the extent of soft tissue loss, but the extent of the problem may be impossible to determine in the acutely ill and unstable patient. Skin graft or flap coverage is usually required and hand stiffness is a common outcome.

Brachial Plexus Block Anesthesia The incidence of postoperative dysesthesias after brachial plexus block anesthesia ranges from less than 2% (91,92) to 12% (17,89). Although rare, perineural fibrosis (90) and permanent neurologic injury (18) have also been reported after axillary block anesthesia.

COMPLICATIONS OF INJURY General Complications of Hand Injury Severe hand injuries are most often the result of crush or rotating blade mechanisms and are best treated by a hand surgery specialist (63). Such injuries usually involve all organ systems of the hand and are always associated with complications. The treatment principles and initial management that may be adequate for lesser injuries may be inadequate for a mangled hand (81). Intervention by a specialist reduces the duration and extent of disability and reduces the overall care requirements and cost of care associated with severe trauma to the extremity.

contractures may lead to progressive growth disturbances. Stiffness and contractures due to mechanical changes in joints and tendons, as discussed above, may develop independently.

Cosmetic Deformity The immature scar may be hypertrophic: thick, red, and raised. These changes usually resolve gradually over the course of a year, although the process may take longer for young children. Permanent visible deformity from hyperpigmentation, thin stretched scars over extensor surfaces, and tight scar bands across flexor surfaces may all be troublesome. Fingernail deformities are common after lacerations and crush injuries in the area of the nail bed. The most common problems are split nail resulting from nail bed injury and hook nail deformity resulting from loss of the tuft of the distal phalanx by a fingertip amputation. Such problems are sometimes unavoidable, but the best prevention is meticulous anatomic repair of nail bed lacerations. Once established, fingernail deformities may be difficult or impossible to correct.

Complex Regional Pain Syndrome Scar Contracture Contractures resulting from skin scarring are more likely to be a problem if the scars extend longitudinally across the flexor surface of a joint. In severe cases, scar contractures may develop over the first few weeks after injury, but in many cases they progress over the course of months. In the growing child, scar

Complex regional pain syndrome (Figs. 18 and 19)— previously known as reflex sympathetic dystrophy, algodystrophy, sympathetic maintained pain, and Sudeck’s atrophy—may develop after any hand injury, but is particularly common in association with nerve injury or irritation. This problem may occur spontaneously after major or minor injury. It may

438

Eaton

Figure 18 Reflex sympathetic dystrophy usually results in swelling, stiffness, disuse, and color and temperature changes. The entire hand is usually affected (A, B), less commonly a single digit (C). Dupuytren’s type contractures are common (D). This may follow elective surgery, but is most often seen after injury, such as a distal radius fracture (E).

cause spontaneous burning pain, hyperalgesia, swelling, vasomotor disturbances, and disuse, and it may be exacerbated by movement. Although there may be spontaneous resolution, most patients experience some degree of chronic symptoms such as pain, stiffness, and difficulty with normal use of the hand, despite all available treatment (103). The best treatment results require early recognition, aggressive medical therapy, and elimination of triggering phenomena. Medical therapy may involve sympathetic nerve blocks, gabapentin or other medications, and biofeedback. Triggers known to aggravate the condition include peripheral nerve irritation due to neuroma or compressive neuropathy, aggressive passive range-of-motion therapy, and dynamic hand splinting. The effects of complex regional pain syndrome may be far more disabling than the initial injury.

Dysfunction Patients may develop maladaptive patterns of use after injury, ranging from awkward positioning to

Figure 19 Reflex sympathetic dystrophy following multiple traumatic finger amputations.

complete disuse of the hand. This complication is often due to unconscious reflex protective mechanisms and may be difficult to correct. Extensor habitus refers to the tendency for the injured index finger or small finger to be held in extension. This unconscious posturing is powered by the independent extensor of the finger and is best treated by early recognition and buddy taping. Alien hand syndrome refers to complete disuse of the hand accompanied by the patient’s perception that the hand is ‘‘not mine.’’ Such problems may also be factitious, but labeling them as such does not improve the patient’s overall outcome.

Compartment Syndrome Compartment syndrome of the hand may develop after crush injury, reperfusion after fracture-related ischemia, intravenous injections, crush or blast injury (25), bleeding after fracture, arterial cannulation or regional surgery, or prolonged pressure on the hand or arm. The forearm is the most common site of compartment syndrome in the upper extremity. Compartment syndrome of the upper extremity is more likely to develop among patients with obtundation. Seriously ill children who receive multiple venous and arterial injections are also at particular risk. Treatment requires prompt recognition and decompression of intrinsic muscle compartments, as well as carpal tunnel release in selected cases (16). The late consequence of compartment syndrome of the upper extremity is Volkmann’s contracture (5,67), which involves both muscle contracture and local ischemic neuropathy. Ischemic muscle contractures respond poorly to nonoperative measures such as splinting; this condition requires an aggressive surgical approach using muscle slides, tendon lengthening, and tendon transfers similar to those used in the

Chapter 36:

treatment of upper extremity spasticity. Neurolysis is indicated for persistent nerve symptoms, but the outcome is unpredictable.

Complications of Specific Injuries Complications of Missed Complex Wounds Complex wounds are those that require additional procedures, such as radical debridement or flap wound closure. Severe Contamination

Severe contamination is common with missed complex wounds of the hand because the hand is so often physically exposed to contaminated mechanisms of injury. Bite Injuries Human Bite Injuries.

Human bite injuries to the hand most often occur as clenched-fist bite injuries, sustained when the fist strikes the mouth of another person during an altercation. The most common constellation of injuries is a skin laceration at the level of the metacarpal head accompanied by injury to the extensor tendon and the metacarpal head. This injury is usually sustained when the hand is in a clenched-fist position, but the patient

Complications of Hand Surgery

439

frequently does not present for treatment until the metacarpophalangeal joint is pulled into extension by dorsal hand swelling. This change in positioning places the soft tissue and bone injuries at an offset, giving the appearance that the injury is more superficial than it is (Figs. 20 and 21). Treatment requires a high level of suspicion, aggressive debridement, and intravenous antibiotics appropriate for a bite injury. Animal bites to the hand are most often dog or cat bites. These bites can lead to prolonged morbidity, particularly when there is a delay between injury and initial treatment (66). Dog bites are associated with soft-tissue crush injury and fractures. Cat bites are particularly dangerous in hand because the cat’s needlelike teeth can easily penetrate into joint spaces, tendon sheaths, and other deep compartments of the hand through a relatively innocuous skin wound.

Animal Bite Injuries.

Bites to the hand from insects such as brown recluse spiders may cause painful, slowhealing wounds, resulting in chronic functional deficits. The initial bite injury may be painless. When surgical excision is indicated, the results appear to be better when surgery is delayed until after the acute inflammatory process has subsided (103).

Insect Bites.

Figure 20 Clenched fist bite injuries of the metacarpal head or proximal phalanx are highly contaminated, often deeply penetrating wounds with bone involvement. They are at particular risk for the development of septic arthritis and osteomyelitis. This patient developed osteomyelitis of the proximal phalanx following such an injury, and received inadequate surgical and antibiotic treatment due to neglect. The image on the left shows a healed wound and a draining sinus, and arrows on the right show periosteal bone formation and erosions due to osteomyelitis. This eventually resulted in amputation.

440

Eaton

Figure 21 Complications of boxer’s fractures. Penetrating injuries of the metacarpal head from clenched fist bite injury mechanism are frequently delayed in presentation, and may rapidly progress to pyarthrosis and osteomyelitis. The patient may present with infection and a trivial appearing wound (A), masking a direct injury of the metacarpal head (B). Casual inspection of the wound in the emergency room may be misleading because the offset of soft tissue and joint penetration at the time of injury (C) is different than at the time of presentation when swelling results in metacarpophalangeal joint extension (D). Palmar angulation of the distal fracture fragment (E) may result in tendon imbalance and secondary flexion contracture of the proximal interphalangeal joint, usually to the same extent as angulation of the fracture malunion (F).

Rattlesnake bites to the upper extremity are associated with serious complications; at least one-third of patients experience complication such as local soft-tissue necrosis (the most common complication), coagulopathy, stiffness, loss of sensibility, and Volkmann’s contracture (43). Antivenin and steroids reduce the degree of swelling and hemorrhage but do not affect or prevent tissue necrosis, which may require operative treatment.

Rattlesnake Bites.

Chemical Burns Industrial Acid Burns.

The hand may suffer industrial acid burns when an inexperienced or careless worker splashes even small amounts of acid onto the fingers or hand. This type of injury can go undetected upon initial evaluation unless a careful history is obtained because visible signs of injury are often delayed (Fig. 22). Hydrochloric and hydrofluoric acids are used in industrial processing and may cause severe burns that do not manifest themselves for a day after exposure. Early recognition and treatment with topical, intravenous, or intra-arterial calcium gluconate can reduce both pain and the extent of tissue loss. White Phosphorus Burns. Workers may sustain white phosphorus burns as the result of handling military munitions, fireworks, and other industrial and agricultural products. Such injuries may result in deep, progressive burns and the systemic effects of multiple organ system failure. Although copper sulfate has been recommended as a specific antidote, the safest and most effective treatment is copious irrigation with water (105). Again, recognition of the nature of injury and immediate

treatment are essential for reducing long-term complications. Injection Injuries High-pressure Injection Injuries.

High-pressure injection of paint, sand, lubricating fluid, or other materials is uncommon, but important because such injuries are often missed in the accident ward. Typically, the patient has briefly placed the hand or fingertip over a pressure spray nozzle, thereby sustaining an injection of material into the soft tissues. Under pressure, this material tracks up tissue planes next to flexor tendons, nerves, and arteries

Figure 22 Topical hydrofluoric acid burns typically do not show visible evidence of injury for a day or two after exposure. By then, effectiveness of topical or systemic calcium treatment is diminished.

Chapter 36:

and through the named bursae and compartments of the hand and arm. Debris may be driven from the fingertip to the chest wall. The examiner may be misled by a small visible wound and (depending on the material injected) relatively few physical findings, and the patient may be discharged only to return within 24 hours because of worsening symptoms. Radiographs may show the presence of air, particulate debris, or pigment (in certain types of paint) in soft tissues. Treatment is emergency radical debridement (62). The pressure-injected material tends to track through the loose areolar tissue along longitudinal structures, and only careful debridement may allow preservation of all vital structures (Fig. 23). In contrast, late surgical treatment may require en bloc tumor-like excision of contaminated zones, or even amputation. Late results are worst when the injected material is either a petroleum-based solvent or a particulate material (sandblasting), when the tendon sheath is involved, and when there is wide proximal spread of the injected material (80). The injected material is not sterile and prophylactic antibiotic treatment is indicated. Poor perfusion in association with such injuries should be treated with primary amputation (80). Injection of pressurized aerosol fluorocarbon liquids such as that used in refrigerants may also result in deep frostbite injury. Intentional Injection Injuries. Injuries associated with the intentional injection of household cleaners, solvents, mercury, or illicit drugs may be difficult to evaluate because of the delusional or drug-seeking

Complications of Hand Surgery

441

Figure 24 Soft-tissue gas is a frequent benign finding accompanying open wounds immediately after injury, but in the presence of infection should always be assumed to be due to gas-forming organisms. Multiple organisms were isolated from the necrotizing infection in the antecubital area of this intravenous drug abuser.

nature of the patient. Radiographs may show particulate or metallic debris or evidence of gas-forming infection (Fig. 24). Factitious or Intentional Wounds

Factitious or intentional wounds to the hand are uncommon, but successful treatment is very difficult because of recurrence. Swelling, ulceration, and recurrent wound breakdown are common. Such wounds are most typical on the dorsum of the nondominant hand. Narcotic-seeking behavior may be part of the overall picture. The most important aspect of treatment is recognition, so that unnecessary, unsuccessful, or mutilating procedures may be avoided. Although the problem is psychiatric, psychiatric intervention may or may not be helpful, and confrontation or intervention is generally ineffective. Such patients may jump from doctor to doctor within a community, and it is wise to notify local colleagues when such problems with a patient are identified.

Complications of Obvious Complex Wounds Complex wounds are, by definition, prone to complications, even with ideal management. Common complex injuries of the hand are associated with predictable types of complications, which are listed below. Figure 23 Paint pressure injection injury. This patient presented with a relatively low pressure–injection injury of the middle finger with a small entrance wound (A). Titanium pigment in the paint was visible on X ray (B). Optimum treatment involved emergency meticulous radical debridement of all stained or damaged tissues, aided by use of the operating microscope (C, D). Late result with recovery of full sensation and range of motion (E, F).

Traumatic Amputations

Hand injuries involving traumatic amputation most often affect the fingers. The associated nerve injury always forms a neuroma, and the treating surgeon should trim the digital nerve ends away from the distal wound so as to lessen the chance of disabling scar

442

Eaton

tenderness. Dysesthesia is common and all patients should be provided with an early desensitization program that can be performed at home. Complex regional pain syndrome may be triggered and then maintained by tender finger stumps (Fig. 19); at first it may be difficult to distinguish this condition from the swelling, stiffness, tenderness, and avoidance that are always associated with the injury. Sensitivity to or intolerance of cold is a problem for most patients, but this condition usually improves after the first year. When more than the distal third of the phalanx is lost, a hook nail deformity will result, with the fingernail curving toward the palm and covering the distal fingertip. This and other variations of retained nail remnant may be avoided by careful total excision of the entire germinal matrix when the amputation is closed. Fingertip amputations are no less problematic than more proximal amputations, particularly when the critical contact areas used in pinching and fine manipulation are involved (Fig. 25). Amputations through the proximal phalanx often result in extensor habitus, described in the section General Complications of Hand Injury/Dysfunction. Metacarpophalangeal joint disarticulation of the index or small finger results in an easily traumatized and visibly prominent metacarpal head. Metacarpophalangeal joint disarticulation of the middle or ring finger results in a ‘‘hole in the hand,’’ through which small objects held in the cupped palm can fall. Treatment of either of these scenarios with removal of a metacarpal replaces the original problem with a narrowed palm and reduced torque grip strength.

Figure 26 Distal phalanx nonunions are uncommon and usually follow open injuries. Risk factors include inadequate reduction of displaced fragments, soft-tissue interposition, or bone loss, including bone loss due to surgical debridement. The top image shows the type of acute displaced fracture that is best treated with internal fixation. The bottom image shows an established nonunion following bone debridement and wound closure. Most fractures that require bone debridement would benefit from internal fixation. Late salvage of distal phalanx nonunions is technically demanding and requires precision bone grafting techniques.

Fingertip Injuries

Fingertip injuries other than amputation are associated with all of the painful and otherwise disabling complications of finger amputations. Nail deformities, tender scars, and nonunion (Fig. 26) are difficult treatment issues. Pediatric fingertip crush injuries are common, and severe injuries involving a sterile matrix laceration with a tuft fracture are frequently missed in children (78). These injuries require meticulous nail bed repair so as to avoid deformity. Foreign Bodies

Figure 25 Fingertip injuries that involve the ‘‘critical contact areas’’ of the thumb, index, and middle fingers pose particularly difficult challenges for reconstruction. Rotating blade saws commonly result in combined injuries such as shown here, impairing the ability to use the hand for fine manipulation.

Foreign bodies in the hand are likely to cause symptoms when they involve the distal phalanx (34). Removal of foreign bodies that are lodged entirely beneath the surface of the skin should be performed with tourniquet control and surgical anesthesia. Otherwise, a common result is that the area where the foreign body is lodged is incised, attempts at retrieval are unsuccessful, and the problem is compounded by the inflammation and scarring resulting from instrumentation. Foreign bodies are most likely to give rise to problems when they are composed either of organic materials (wood, plant thorn, etc.) or of highly contaminated materials. Phoenix date palm thorns frequently produce a chronic sterile

Chapter 36:

Complications of Hand Surgery

443

inflammatory reaction and require primary treatment with radical debridement and extensive synovectomy (93). Foreign-body entry points at the dorsal surfaces of the metacarpophalangeal or interphalangeal joints or at the palmar flexion creases of the fingers are at particular risk for contamination of tendons and deepspace infections. Chronic symptomatic foreign-body problems require tumor-like excision and synovectomy, not incision and removal (Fig. 12). Thermal Burn Injuries

Thermal burns to the upper extremity result in stiffness of the hand; this complication is best prevented by early active motion within two weeks of injury (65). This goal is difficult to achieve reliably because the depth of the burn may be difficult to assess, and areas that require skin grafting must be immobilized for at least one week after surgery. When possible, the goal is early definitive wound closure with fullthickness or tangential excision and skin grafts or flaps, followed by motion at the earliest possible opportunity. The ultimate disability in hand function is thought to relate to the time required to achieve wound closure, although this point is controversial (72). Burn injuries can cause lifetime problems that cannot be cured by any amount of surgery and therapy, and the surgeon must strive to promote realistic, achievable goals (70). Common complications are compartment syndrome (6,25); contractures (68) of web spaces and extensor and flexor surfaces; and hypertrophic scars and heterotopic ossification. Surface contact burns over the course of the brachial artery are rare but may lead to ischemic limb loss (4). Among children, burns to the hand more commonly involve an isolated contact burn to the palm, particularly among infants, that is sustained when a child grasps a hot object such as a curling iron and then grips even more tightly in response to pain. As for burns in other areas, excision and grafting are indicated if healing is expected to take more than three weeks, but in this instance contractures requiring additional reconstructive procedures are common (68). Pediatric hand burns have the most favorable outcome when managed in a specialty treatment program (77). Frostbite

A wide variety of early treatments are recommended for frostbite (106), but rapid rewarming is standard. Traditional management is observation and delayed amputation (Fig. 27). A bone scan may help distinguish unsalvageable from potentially salvageable regions. Early surgery may provide marginal tissue with a new blood supply and may preserve both the function and the length of the upper extremity. Electrical Injuries

Electrical injuries to the upper extremity may produce extensive deep-tissue injury, compartment syndrome,

Figure 27 Frostbite injuries most commonly affect the face, fingers, and toes. Viability may be estimated from bone scan, but even with this information, it is best to wait until a line of demarcation is clear. In the absence of infection, this process may take months.

delayed tissue necrosis, and delayed vascular thrombosis (110). Early exploration and decompression of deep compartments, vascular graft reconstruction of segmental defects, and early free microvascular flap reconstruction reduce the likelihood of amputation and shorten recovery times (85,108,109,111). Even with optimum treatment, however, long-term sensory loss is common and is an unsolved problem (107). Degloving Injuries

When the hand is caught in moving machinery, degloving injuries often result. If microvascular replantation of the degloved tissues is possible, this treatment approach probably yields the best final result, although achieving sensory recovery is difficult even with this technique (51). If replantation is not possible, efforts to salvage a crushed avulsed flap are usually unrewarding, and primary excision and resurfacing with a graft or flap (Fig. 28) are indicated so as to avoid a prolonged course of progressive flap loss, delayed healing, infection, and stiffness. Mangling Hand Injuries

The hand can be mangled by a wide range of mechanical injuries, usually involving all tissue components of the hand. Mechanisms include crush, blast, ballistic, traction, and avulsion injuries. All complications are possible and patients with these injuries are at particular risk of delayed healing, marginal wound necrosis (Fig. 29), infection (Fig. 30), delayed thrombosis, prolonged swelling, compartment syndrome (16,25), intrinsic muscle contractures (Fig. 31), nonunion (Figs. 32 and 33), stiffness, and lack of sensory

444

Eaton

Figure 28 Traumatic flaps have unpredictable vascularity, particularly when they are distally based, involve a crush mechanism, and involve the skin of the palm. This patient sustained a crush injury of the hand, resulting in distally based long, thin, palmar flaps (left, center). Primary wound healing and early recovery are most reliably achieved with excision and resurfacing areas of indeterminate viability. In this case, a full-thickness graft was applied to the palm (right).

recovery (51). The initial management plan is crucial, as outlined in the next section.

Complications Associated with the Treatment of Severe Hand Wounds

Figure 30 Crush and avulsion wounds typically have combined effects of indeterminate or inadequate vascularity and widespread contamination. Infection following such wounds is due to inadequate debridement, and primary wound closure increases the chance of marginal wound necrosis. This patient presented after primary closure of a dorsal hand crush–avulsion injury in which the extensor pollicis longus had been repaired. The wound margins became necrotic (A) and the extensor pollicis longus tendon underwent a progressive septic liquefaction necrosis (B). He was treated with wide debridement (C), and then was lost to follow-up. He performed his own wound care and healed uneventfully with the wound nearly closed one month after debridement (D).

Severe hand wounds may lead to many different complications that can add additional trouble to an already difficult situation.

Failure to Proceed with Primary Amputation

Figure 29 This series demonstrates marginal necrosis in a complex wound. This patient sustained multiple amputations in a power saw accident and underwent replantation of two of his fingers. Figures 1A and 1B show the original injury, and 2A and 2B show the appearance immediately after replantation. Figures 3A and 3B show peripheral necrosis of both proximal and distal wound margins. In this case, such conditions may lead to late failure of replantation due to thrombosis of the vascular repairs due to an inhospitable wound environment. Fortunately, secondary healing was uneventful and the replanted digits survived.

Figure 31 Intrinsic contractures. This patient sustained a crush injury of the hand, resulting in direct damage and late fibrosis of all of the intrinsic muscles. The thumb intrinsic muscles were extruded through burst wounds. Intrinsic contractures produce metacarpophalangeal joint flexion contractures, first web space contracture, and extension contractures of the interphalangeal joints, all evident here. Secondary swan-neck deformity of the fingers is also common. Correction requires extensive release of the intrinsic muscles, and then in cases such as this, first web reconstruction with a flap and opponensplasty.

It is difficult and emotionally stressful to decide when or whether to amputate a severely injured hand or digit, particularly when the part in question has at least the appearance of an existing blood supply.

Chapter 36:

Figure 32 High-velocity gunshot wounds result in a wide zone of softtissue injury. Fractures associated with high-energy injuries (A) are more likely to have complications of delayed union, nonunion, and hardware failure (B, C).

‘‘Saving’’ a mangled hand may simply burden the patient with a painful, useless extremity, resulting in a triumph of technique over judgment. One guide to making this decision is to ask the following question: If this injured extremity looked like this but involved complete amputation, would replantation be indicated? If the answer is clearly no, primary amputation should be strongly considered (Fig. 34). The best time to proceed with primary amputation for a mangled extremity is at the very first operation. If the surgeon realizes at the time of the first operation that the hand is unsalvageable but does not amputate, a precedent is set for false expectations and even greater disappointment than would otherwise be endured. The patient and the family see the bandage, conclude that the

Figure 33 Crush injuries are prone to delayed wound healing and stiffness with standard open reduction and internal fixation techniques. Percutaneous fracture pinning allows fracture stabilization without the addition of additional trauma from fracture site exposure.

Complications of Hand Surgery

445

Figure 34 The decision to proceed with primary amputation is the most emotionally difficult when there are apparently salvageable parts. This patient’s arm was trapped in a rotating wheel mechanism, resulting in combined crush and burn injuries of the forearm from above the elbow to below the wrist, and six hours of complete ischemic time prior to arrival at the trauma center. The hand was relatively spared. While technically possible to replant the hand at the level of the distal humerus, a prosthesis would be expected to provide better function and appearance, and the injury was converted to an above elbow amputation.

hand has been ‘‘saved,’’ and will find it much more difficult later to accept the fact that it has not. Although some patients with a saved mangled extremity may decide later to undergo elective amputation (59), most will be unable to make this decision even if the hand is a burden and is clearly inferior to a prosthesis.

Inadequate Debridement The single common denominator of wound healing complications such as infection, delayed healing, marginal necrosis, and wound breakdown is inadequate debridement. If the zone of injury can be determined with reasonable certainty, severe wounds should be radically de´brided with anticipation of the potential need for complex flap closure. Debridement should remove severely contaminated tissues and all ischemic tissues that cannot be revascularized, including crushed flaps, distally based flaps with a length-towidth ratio of more than one to two, and flaps that are obviously ischemic. Initial debridement should be performed under tourniquet control, and proper initial debridement of severe wounds involves en bloc tumor-like excision with a scalpel and saw, not a curette or irrigation, although these techniques may be used later. The skin of the palm has a primarily perpendicular rather than tangential vascular pattern, and traumatic palmar flaps should be considered for primary excision and alternate resurfacing because their vascularity is quite unreliable (Fig. 28).

Poor Timing of Wound Closure Traditionally, the timing of closure of severe hand wounds has been classified as primary (immediate), delayed primary (within two weeks), and secondary

446

Eaton

(after two weeks). Historically, delayed primary closure was recommended for military and other severe hand injuries. This recommendation is still appropriate when the only available wound closure technique is direct closure or closure with local flaps. However, the timing of wound closure using distant or microvascular free flaps follows different guidelines. The status of severe open wounds that are candidates for flap closure is classified as acute (before the appearance of granulation tissue; usually less than one week), subacute (after the appearance of granulation tissue but before dense scarring; usually one to four weeks), or chronic (usually after one month). Wounds that require flap closure are associated with the lowest complication rate (fewer flap failures, fewer postoperative infections, shorter hospitalization periods, fewer operations, and shorter overall period of disability) when closure is performed during the acute phase, and with the highest complication rate when surgery is performed during the subacute phase (27,76,83,84,86,87). Free flap reconstruction of burn injuries is associated with the lowest complication rate when it is used for the reexploration and reconstruction of healed, closed burn injuries (85).

Figure 35 Loss of axial pedicle flaps. Axial flaps, although generally reliable, still have a peninsular vascular anatomy, and the most distal extent of the flap may fail if the blood supply is compromised either by kinking or by tension of the flap. This last problem is most likely when a flap has been placed in a circumferential geometry, where swelling produces progressive tension along the circumference of the flap. (A, B) Pedicled inferior epigastric flap used to resurface the hand dorsum following hand replantation. The flap tip was lost, most likely due to unprotected tension and kinking at the juncture of the flap and the recipient site. (C, D) Reversed pedicled radial forearm flap tip lost due to combination of extension of flap harvest out of the most vascular zone and tension from circumferential wrapping of the flap around the hand.

Technical Failure of Complex Wound Closure Even with adequate debridement, avoidance of local flaps from the potential zone of injury, and careful planning, wound closure may fail. Skin grafts to the hand may be lost because of inability to provide adequate immobilization, and flaps may be lost when the complex wound dimensions exceed the capacity of the flap. Although free flaps tend to be successful or to be lost on an ‘‘all or none’’ basis, partial loss of a free flap may occur. Loss of a pedicled flap usually occurs at the exact point at which flap coverage is crucially needed (Fig. 35). However, even a perfectly designed and executed flap cannot obviate the effects of inadequate debridement or poor timing of wound closure.

(Fig. 29) if debridement is inadequate or if the surgeon in unable to distinguish viable from nonviable tissues in a wide zone of injury. The most common complication of successful replantation is stiffness due to tendon adhesions (14). Cold intolerance is uncommon after pediatric replantation but occurs after most adult replantations (14). Aesthetically disturbing fingertip atrophy occurs with nearly half of all replanted digits (14) because of the effects

Complications of Replantation All complications of complex hand wounds can occur after replantation, including tendon adhesions, tendon rupture, neuroma, and delayed healing. In addition, however, replantation carries the risk of a number of additional problems. Early vascular failure (Fig. 36) of replantation is influenced by the mechanism of injury and patient selection. Early failure is more common among smokers (69), when replantation occurs at a more distal level, and when the injury has involved crush or avulsion (79). After successful revascularization, venous problems are more likely than arterial thrombosis to result in loss of the replanted extremity (14,20). The crucial time for failure or successful salvage is the first four postoperative days (20). As is true for other wounds, replantation can result in marginal necrosis or interval gangrene

Figure 36 Digital replantation may fail in part due to damage from long periods of ischemia (left), or more commonly fail completely because of inability to maintain revascularization (right).

Chapter 36:

of incomplete reinnervation and, in some cases, the late effects of prolonged ischemia (Fig. 36). Lack of sensory recovery is more common when the patient is an adult, when both arteries have not been repaired (51), and when the injury involves avulsion (79). As with any other vascular repair, replantation may be associated with local vascular complications such as pseudoaneurysm (5), arteriovenous fistula (37), stricture, and late thrombosis. Delayed union, nonunion, or avascular necrosis may occur, particularly when the replantation is performed at the level of the phalangeal neck (44) because the phalangeal head is covered with cartilage and has a primarily intramedullary blood supply. Fractures or osteotomies through this level are prone to these complications even after procedures other than replantation (61). Prolonged incapacitation and multiple operations are typical; the average patient requires two or more additional procedures after replantation (52). Judgment as to the indications for replantation must include the realization that the poor results after replantation may be much more disabling than primary amputation. Functional outcome is significantly worse when replantation involves prolonged ischemia (Fig. 36) or injury in flexor tendon zone II (69).

Complications of Vascular Injuries Missed Vascular Hand Injuries Ring Avulsion Injuries

Ring avulsion injuries range from trivial skin lacerations to arterial or venous disruption to combined injuries in continuity to complete amputation. The zone of injury is usually greater than would be detected by casual inspection, and extremities with combined vascular and skeletal disruption injuries are often not salvageable despite the external appearance of a simple laceration (Fig. 37). When the ring is completely pulled off the finger in association with a circumferential finger wound, the distal soft-tissue envelope is usually severely injured, effectively turning the soft tissue sleeve inside out and irreparably damaging the distal part. For all but the most minor injuries, successful salvage with vein grafts and flaps is unlikely and, even when successful, often results in a stiff, insensate digit.

Complications of Hand Surgery

447

Figure 37 Ring avulsion injuries may appear deceptively normal because of color from blood trapped in the digit and active range of motion maintained by the long tendons. In this example, the finger appeared viable (A), despite a long zone of tissue disruption from the distal interphalangeal joint (B) to the base of the finger (C, D, E). In fact, the finger was only attached by the flexor tendons and digital nerves (F) and was unsalvageable.

ischemia or injury to the superficial radial nerve (7,41). The incidence of hand ischemia after radial artery harvest may be reduced by the use of preoperative color duplex scanning in addition to careful physical examination.

Partial Vascular Laceration

Partial vascular laceration is the most likely cause of persistent uncontrolled hemorrhage; such an injury may result in substantial bleeding. Persistent bleeding is better controlled by local pressure than by blind clamping, which may result in iatrogenic nerve injuries (Fig. 38). Late effects of partial vessel laceration include pseudoaneurysm, delayed hemorrhage, and delayed thrombosis. Iatrogenic Hand Injuries Related to Vascular Surgery Graft Harvest. Harvesting the radial artery for cor-

onary artery bypass procedures may result in hand

Figure 38 The drama of hemorrhage may lead to unwise maneuvers in the emergency room. Blind clamping of bleeding structures without proximal tourniquet control is risky and may compound injury. (A, B) This patient had inadequate control of bleeding despite multiple clamps placed by an inexperienced physician, who was successful in clamping the ulnar nerve and flexor carpi ulnaris tendon, but failed to control the large subcutaneous vein that was actually responsible for the bleeding. Continued hemorrhage is often due to partial vessel lacerations, as in this partial laceration of the cephalic vein (C) and this partial laceration of the radial artery (D). Ligation of structures without adequate visualization in the emergency room is also problematic, as in this patient who had inadvertent ligation of the ulnar nerve within Guyon’s canal (E).

448

Eaton

Nearly 2% of patients undergoing new angio-access procedures experience severe hand ischemia (11). This problem is more common among diabetic patients who have undergone multiple angio-access procedures (9) or who have diabetic neuropathy (24). Prompt recognition and treatment are crucial if tissue loss and permanent nerve injury are to be avoided. This problem should be suspected when finger pain, numbness, or nerve symptoms arise immediately after angio-access surgery. Optimum treatment options include ligation of the fistula, intraoperative duplex scanning– guided banding (10), or distal revascularization– interval ligation. Neurologic symptoms may arise even if critical ischemia cannot be demonstrated, and recovery of nerve function is unpredictable (24). Direct nerve compression may result from adjacent access materials (Fig. 16) or from hematoma around the side of a vascular suture line (35).

Dialysis Access.

Upper extremity ischemia has been reported as a steal phenomenon after axillofemoral bypass grafting, and as the result of emboli after thrombosis of an axillofemoral bypass graft (9,40).

Bypass Surgery.

Inappropriate Primary Call to the Vascular Surgeon

A common problem with hand injuries is an inappropriate primary call to the vascular surgeon. Upper extremity hemorrhage is usually best managed by an upper extremity surgeon. Time permitting, the ideal order of repair of the severely injured upper extremity is debridement, skeletal stabilization, musculotendinous repairs, nerve repairs, and then vascular repairs (Fig. 39), all under tourniquet control. Such an approach minimizes hemorrhage and allows the most precise primary repairs. Unfortunately, a common scenario is that the bleeding arm is first managed by a vascular surgeon, who does not provide definitive care of adjacent nerve and musculoskeletal injuries.

Severe Trauma: Order of Procedures Anticipated final outcome justifies attempted salvage? Yes

No

Soft tissue deficit?

Amputation

Yes No

General Complications of Upper Extremity Vascular Injuries After vascular injury to the hand, as elsewhere, ischemic gangrene, chronic ischemia, intrinsic contracture, traumatic aneurysm, arteriovenous fistula, thrombosis, and embolism may occur. Supracondylar fractures of the humerus may result in brachial artery compression or disruption, and postischemic reperfusion compartment syndrome of the forearm can follow restoration of arterial flow after either closed reduction or vascular repair. Fasciotomy should be considered if ischemic time exceeds two hours and should be performed if compartment pressures are elevated.

Complications of Treatment of Vascular Injuries Inappropriate Use of Techniques to Control Bleeding

The use of inappropriate techniques to control bleeding in the emergency room can add substantial injury. Nearly all bleeding in the upper extremity can be controlled by elevation and direct pressure. In the emergency room, the use of tourniquets should be limited to a few minutes at most, and tourniquets should ideally be used only by the surgeon who will provide the definitive surgical care. Inflating a tourniquet and then waiting for the hand surgeon to arrive in the emergency room is inappropriate and dangerous, and it limits future treatment options. Similarly, the use of local destructive intervention with clamps, ligature, or cautery (Fig. 38) by anyone other than the surgical specialist assuming final care of the patient should be strongly discouraged.

Decide now: Emergency flap Ectopic replantation Radical shortening

Vascular Injury? Yes

No Muscle vascularized? No

No

Yes Critically ischemic over 4 hours?

No

Yes

Enough time to revascularize (steps 1-5 below) with less than 4 hours ischemia? Yes

Decision to allow retained ischemic injured muscle (e.g. hand replant) or not (e.g. arm replant)

No

Proceed: A. Heparinize B. Shunt C. Reperfuse

Yes

Then

Proceed as applicable, in this order: 1. Debride under tourniquet control, extensile incisions 2. Wound irrigation/lavage 3. Skeletal realignment and stabilization 4. Fasciotomies if muscle ischemia time over 2 hours 5. Musculotendinous repairs adjacent to vascular injuries 6. Definitve vascular repairs 7. Confirm final plan for wound closure 8. Remaining deep soft issue repairs 9. Wound closure

Figure 39 Optimum management of severe extremity trauma involves an organized approach to avoid useless attempts at salvage, maximize the potential success of salvage, and minimize iatrogenic complications such as excessive blood loss.

Chapter 36:

In this situation, the vascular injury is repaired, often with a graft, and then the extremity surgeon is called in to complete the work. When the adjacent nerve and muscles are then repaired, the vascular ‘‘gap’’ that was thought to require a graft disappears, and the graft may need to be removed to avoid kinking due to redundancy. Similarly, performing only vascular repair, closing the wound, and referring the patient for secondary repair of adjacent structures may sacrifice the best opportunity for a precise primary repair of all structures in the most safe and efficient manner.

Complications of Tendon Injuries Complications of Missed Tendon Injuries Tendon injuries can be missed when either the patient or the initial examining physician fails to appreciate subtle findings. Partial Tendon Lacerations

A partial tendon laceration (Fig. 40) should be suspected when the patient apparently has full motion, but has pain when attempting to use the tendon against resistance. Consequences of partial tendon lacerations include delayed rupture, scarring with tendon adhesions, triggering, and weakness. Missed Injuries to the Extensor Mechanism of the Finger

Complications of Hand Surgery

449

at the proximal interphalangeal joint should be suspected when the patient has suffered a regional injury and has pain upon attempted extension against resistance, even if the patient has full, active motion without resistance. Missed Injuries to the Flexor Tendon of the Finger

Missed flexor tendon injuries are less common than missed extensor tendon injuries because of the change in the resting posture of the hand (Fig. 3). Isolated superficialis tendon injury with an intact profundus tendon produces only a subtle change in finger posture and is easily missed. Profundus tendon avulsion injuries (Fig. 41) are often unappreciated by the patient, who believes that the finger is simply ‘‘jammed’’ and delays medical evaluation until the best window of opportunity for treatment has passed. If there is substantial proximal retraction after profundus avulsion, the flexor tendon sheath fills with blood and within a matter of days shrinks enough so that reinsertion is either impossible or does not result in functional movement. Missed Injuries to the Extensor Tendon Injuries of the Dorsal Hand

Injuries to the extensor tendon of the dorsal hand may be missed because they often cause little initial functional deficit, either because of the action of adjacent

Injuries to the extensor mechanism of the finger may be missed because the broad expanse of the extensor mechanism can initially maintain posture until softening from the healing process allows the remnants of support to give way. Terminal tendon injuries at the distal interphalangeal joint and central slip injuries

Figure 40 Partial tendon lacerations should be suspected after penetrating wound injury when the muscle tendon unit functions with pain, or when there is pain when attempting motion against resistance. Missed partial tendon injuries may result in triggering, loss of motion from adhesions, or delayed tendon rupture.

Figure 41 Flexor profundus tendon avulsion injuries occur when the fingertip is suddenly pulled into extension while being actively flexed. The tendon may retract all the way into the palm at the metacarpal neck level (A). Lesser degrees of tendon retraction are more likely when the injury results in an avulsion fracture, which may lodge at the A2 or A4 flexor tendon pulley (B, C) or be minimally displaced (D). Retraction proximal to the proximal interphalangeal joint disrupts the vincular arteries, ischemic tendon damage, and a worse outcome.

450

Eaton

tendinous junctures or because only one of two (proprius and communis) tendons in the index or small fingers have been cut. Extensor pollicis longus tendon injuries may be missed because of trick motion through the action of the thumb intrinsic muscles on the thumb extensor mechanism; this action may allow interphalangeal joint extension to neutral despite a divided extensor pollicis longus tendon.

Common Complications of Tendon Injuries of the Hand Injuries to the tendons of the hand can result in complications such as stiffness, contractures, tendon rupture, recurrent adhesions, and weakness. The number and degree of complications depend on the exact level of injury. The worst results of flexor tendon injuries occur when the injuries are located in the flexor tendon sheath extending from the metacarpal head to the middle portion of the middle phalanx, referred to as ‘‘Zone II’’ or ‘‘no man’s land.’’ Even with ideal treatment, only about half of the patients with injuries at this level recover to experience good-to-excellent function, and fewer have a satisfactory outcome after staged flexor tendon reconstruction with a tendon graft (58). Quadrigia syndrome, or limited excursion of the middle, ring, and small fingers, occurs because of tethering connections between the profundus tendons of these fingers. This condition may follow a simple flexor tendon injury or may be the result of adhesions after amputation. The worst results of extensor tendon injuries occur when the injuries are located over the dorsum of the proximal phalanx or the proximal interphalangeal joint. Loss of proximal interphalangeal joint motion may take the form of a fixed contracture, a swan-neck deformity, or a boutonniere finger. Thin soft-tissue coverage and poor tolerance of any change in length contribute to the poor results associated with injuries at this level.

Complications of the Treatment of Tendon Injuries Tendon Adhesions

Adhesions are the most common problem after tendon repair. Rupture of a flexor tendon repair occurs in at least 4% of cases after primary flexor tendon repair in Zone II with postoperative controlled passive motion (58). Stiffness may be due to either adhesions or rupture, and it may be impossible to determine the nature of loss of motion, even with magnetic resonance imaging. Mallet Finger

Nearly half of the patients treated for mallet finger experience some type of treatment complication. Complications after surgery are more common, more serious (e.g., deep infection), and more frequently permanent than those arising from splinting alone (30).

Bowstringing

Bowstringing due to incompetence of the flexor tendon pulley system may follow injury or iatrogenic injury during efforts to expose, retrieve, and repair the tendon. External ring splints are commonly used to support the tendon pulley system, but have not been shown to be mechanically effective in preventing bowstringing. Staged Reconstruction of the Flexor Tendon

Reconstructing the flexor tendon in stages by using temporary silastic tendon spacers followed by tendon grafts carries all of the risks of primary tendon repair. In addition, staged reconstruction is more likely to result in flexion contractures and greatly extend the length of the necessary incapacitation.

Complications of Nerve Injuries Missed Nerve Injuries Partial Nerve Lacerations

Partial nerve lacerations may be missed because their presentation does not provide a full-blown picture of anesthesia or paralysis. Such injuries are best treated by primary repair (Fig. 11). Delayed or secondary exploration may result in additional nerve injury because it may be impossible to distinguish between healing tissue, scar tissue, and nerve tissue that is either functioning or has the capacity to heal. Late exploration of a healed partial nerve injury usually reveals an amorphous neuroma in continuity, and the only practical option may be to completely divide the nerve, excise the neuroma, and reconstruct the nerve with nerve grafts. This procedure may be difficult to justify when the patient has either retained or recovered partial nerve function. Motor Branch Injuries

Motor branch injuries are most often missed after wounds with a small entry site but deep penetration. The ulnar motor branch in the palm, the median motor branch in the palm, and the posterior interosseous nerve in the forearm may be injured without producing sensory loss, and these injuries may be missed by a casual survey.

Complications of Common Nerve Injuries Nerve injuries in the hand can lead to complications such as tender neuroma, paralysis, and incomplete sensory recovery. In addition, upper extremity nerve injuries usually produce some degree of cold intolerance and are a common trigger of complex regional pain syndrome. Dysesthesia and disuse of the hand may occur and are best treated with an aggressive desensitization and sensory reeducation program under the supervision of a hand therapist. Median nerve injuries result in a greater loss of hand function than ulnar nerve injuries because the critical contact areas of the hand are affected.

Chapter 36:

Complications of Hand Surgery

451

Complications of the Treatment of Nerve Injuries The treatment of nerve injuries may also result in complications, including failure due to repair under tension or repair within a poorly vascularized softtissue bed and contractures due to splinting for the relief of tension on a tight repair. Patients who have a wide zone of anesthesia must be instructed about self-protection from cuts and burns. Contractures resulting from paralysis are avoidable but must be anticipated and prevented with splinting: if median nerve palsy is left untreated, a contracture of the first web space will result, and ulnar nerve palsy will result in proximal interphalangeal joint contractures of the ring and small fingers.

Complications of Fractures and Joint Injuries Missed Fractures and Joint Injuries Scaphoid and Hook of Hamate

Scaphoid and hook of hamate fractures are commonly missed. These fractures are discussed below. Reversed Bennett’s Fracture

A reversed Bennett’s fracture is an intra-articular fracture of the base of the small finger metacarpal. This injury is usually associated with dorsal and proximal subluxation of the metacarpal shaft, which is caused by the unresisted action of the extensor carpi ulnaris tendon. In contrast to intra-articular fractures of the thumb metacarpal base (Bennett’s and Rolando’s fractures), which have a similar pathologic anatomy (Fig. 42) and good outcome with a variety of treatment techniques (64), reversed Bennett’s fractures are prone to chronic symptoms related to posttraumatic arthritis. These fractures are easily missed on plain anteroposterior and lateral radiographs, and their presentation is frequently delayed when they are sustained as boxing injuries.

Figure 42 Bennett’s and reverse Bennett’s fractures. These are fracture dislocations of the thumb and small metacarpal bases, respectively. They are each inherently unstable because of the resultant unopposed distracting forces of remaining muscle tendon attachments on the larger fracture fragment. The abductor pollicis longus displaces the thumb Bennett’s fracture, and the extensor carpi ulnaris displaces the small finger reverse Bennett’s fracture. Each of these requires at least temporary internal fixation to reliably maintain reduction.

Complications of Common Fractures and Joint Injuries Intra-Articular Fractures

Intra-articular fractures of the fingers frequently result in stiffness and functional impairment, particularly when they are sustained during childhood (49).

Phalangeal Neck Fractures

Fractures of the phalangeal neck may go unrecognized despite rotation or dorsal translation of the distal fracture fragments because alignment may look deceptively normal on routine posteroanterior radiographs. The rotated phalangeal neck fracture is unstable, prone to nonunion (61), and sometimes referred to as a ‘‘hangman’s fracture’’ (Fig. 43), not only because it involves the ‘‘neck,’’ but also because it is easy to miss in children and difficult to treat late. Missed Ligament Injuries

When ligament injuries are missed, it is usually because the patient downplays the extent of injury, only to seek evaluation later because of persistent symptoms. The most common missed ligament injuries are gamekeeper’s thumb and scapholunate ligament injuries, both of which are discussed below.

Figure 43 Phalangeal neck fractures are sometimes referred to as ‘‘hangman’s fracture,’’ either because the break is through the ‘‘neck’’ of the bone, or more importantly because the posterioranterior X-ray may be deceptively normal (1A, 2A), despite extension rotation of the distal fragment, as in these cases, rotated 30 (1B) and 70 (2B). Ninety-degree rotation may occur with little change on the PA view. Such fractures are prone to avascular necrosis of the distal part, even more so following later attempts at corrective osteotomy.

452

Eaton

Displaced articular fractures should be anatomically reduced and fixed whenever possible. Even minor degrees of malalignment are usually unacceptable. Long-term problems, including degenerative arthritis, are common, even with optimum initial care. Pathologic Fractures

Pathologic fractures of the hand are most commonly due to enchondroma involving one of the tubular bones. Complications of treatment are more likely in association with immediate rather than delayed treatment of the tumor (2), and the preferred management of pathologic fracture resulting from a benign tumor is to let the fracture heal and then initiate definitive treatment of the tumor. Complications of Phalangeal Fracture Distal Phalanx Fractures. Fractures of the distal pha-

lanx carry all of the complications associated with fingertip injuries, as previously discussed. Displaced fractures of the distal phalanx (Fig. 26) may give rise to nonunion if they are not reduced and provided with adequate internal fixation. Phalangeal neck fractures are discussed above. Phalangeal shaft fractures are affected to a much greater degree by associated soft-tissue damage and have an overall worse outcome than metacarpal fractures of similar magnitude. Poor functional outcome is common with phalangeal fractures that are open, comminuted, or associated with either significant soft-tissue injury or periosteal stripping, including periosteal stripping performed during open reduction (56,95) when there is associated nerve or tendon injury. Only about one in six displaced phalangeal fractures is stable after closed reduction (56), and redisplacement may occur after temporary fixation with Kirschner wires. Angulation causes a zigzag posture because of tendon imbalance, which results in joint contracture to a degree similar to the degree of proximal angulation (Fig. 44). Outcome is not improved when plateand-screw fixation is used instead of Kirschner wire fixation (54). On the basis of outcome studies, a strong argument can be made that all finger fractures should be referred to a surgeon with specialty training in hand surgery (55). Essentially all interphalangeal joint injuries are prone to complications because of the precision nature of the interphalangeal joints. It is common for the sprained proximal interphalangeal joint to be stiff, tender, painful, and swollen for 6 to 12 months after injury. Permanent joint enlargement and flexion contractures are common consequences of even a minor sprain or ‘‘jammed finger.’’ Mallet fracture dislocations (Fig. 45) of the distal interphalangeal joint should be distinguished from simple stable displaced mallet fractures because outcome after conservative management of such dislocations is

Phalangeal Joint Injuries Prone to Complications.

Figure 44 Unstable phalangeal fractures follow a predictable course regarding three complications. First, dorsal-palmar angulation of mid-shaft phalangeal fractures occurs due to asymmetric pull of muscle tendon units. Proximal phalanx fractures typically fall into a dorsal concave (‘‘apex volar’’) angulation. Second, failure of reduction often results in complete recurrence of fracture deformity. Third, tendon imbalance due to malunion in angulation results in joint contractures distal to the fracture equal in magnitude and opposite in direction as the malunion. The top radiograph demonstrates initial angulation of such a fracture. The fracture was reduced and stabilized with Kirschner wires, but redisplaced after the original surgeon removed the fixation. The bottom radiograph demonstrates the healed malunion, complete recurrence of the initial deformity, with a flexion contracture of the proximal interphalangeal joint equal in degree to the angle of malunion.

poor as the result of joint incongruity. Pure dislocations of the proximal interphalangeal joint (Fig. 46) are most commonly dorsal, are usually stable after reduction, and carry about the same outlook as a bad sprain of this joint. In contrast, palmar dislocations or dislocations with a lateral component are frequently unstable after reduction and are more prone to progressive contractures, angulation, and degenerative joint changes. Fracture dislocations of the proximal interphalangeal joint are usually dorsal with a small volar plate avulsion fracture. These dislocations are usually stable if the volar fracture fragment comprises less than one-third of the articular surface. In contrast, dorsal fracture dislocations in which the palmar fragment involves more than one-third of the joint surface, palmar fracture dislocations, and combined dorsal and palmar fractures (‘‘pilon fractures’’) are intrinsically unstable and cause persistent subluxation

Chapter 36:

Figure 45 Mallet fractures in which the smaller fracture fragment includes half or more of the joint surface may be unstable, resulting in palmar subluxation of the distal phalanx (top). However, this is not always the case, and the joint may remain stable despite a large avulsion fragment (bottom).

(Fig. 47). These injuries are extremely difficult to treat and may require internal and external fixation and cancellous or osteochondral grafting. In fact, these injuries may be unsalvageable.

Complications of Hand Surgery

Figure 47 Proximal interphalangeal joint fracture-dislocations are always unstable. Internal and/or external fixation are usually appropriate. Stiffness, flexion contracture, and posttraumatic arthritis are common consequences.

Metacarpal Fractures Prone to Complications. Metacarpal fractures heal in a fairly predictable manner, but nonunion is more likely when injuries involve a crush or blast mechanism. Gunshot injuries of the fingers frequently result in amputation, but similar injuries in the metacarpal area may produce surprisingly little nerve and tendon damage despite severe skeletal injury and risk of nonunion (Fig. 32). Multiple metacarpal fractures are often sustained in association with crush injuries, and the treating surgeon must decide between compartment decompression with wide exposure for open reduction and internal fixation or percutaneous fixation that can minimize additional injury (Fig. 33).

Complex dislocations are those in which intra-articular soft tissue interposition provides a block to reduction; they are also referred to as irreducible dislocations. These injuries most often involve the metacarpophalangeal joint of the thumb, and are usually associated with sesamoid interposition (Fig. 9). These dislocations usually require open reduction and must be recognized so that additional injury from overzealous attempts at closed reduction can be avoided. Rupture of the ulnar collateral ligament of the metacarpophalangeal joint of the thumb, also known as ski pole thumb or gamekeeper’s thumb, occurs when the thumb is forced into radial deviation. The extent of injury is frequently not appreciated by the patient, and delayed presentation is common. The results are better with

Metacarpal Joint Injuries Prone to Complications.

Figure 46 Proximal interphalangeal (PIP) joint ‘‘simple’’ dislocations are usually dorsal, and may be reduced by extending the finger to reproduce the deformity and then dorsal pressure on the base of the middle phalanx. These are usually stable following reduction. Volar dislocations represent a more serious injury, are less commonly easily reduced, and more commonly unstable following reduction. Although often passed off as a ‘‘jammed finger,’’ PIP dislocations commonly result in 6 to 12 months of swelling, PIP flexion contracture, and permanent joint enlargement.

453

454

Eaton

Figure 48 Scaphoid fractures usually occur as an isolated fracture (A) but may be part of a larger injury complex (B). Oblique (C), comminuted (D), and proximal (I) fracture lines are prone to nonunion. Nonunion may be radiographically subtle (E), cystic (F), or hypertrophic (G). The majority of scaphoid nonunions progress to a pattern of radioscaphoid and midcarpal arthritis referred to as scaphoid nonunion advanced collapse or ‘‘SNAC wrist’’ (H, I).

acute ligament repair than with late reconstruction (53), and arthrodesis may be indicated. Carpal Injuries Prone to Complications. Scaphoid Fractures. Scaphoid fractures (Fig. 48) are prone to healing problems because of the combination of poor perfusion of the proximal fracture fragment and strong forces across the fracture site because of normal wrist mechanics. Scaphoid fractures may heal in malunion (‘‘humpback deformity’’), but delayed union and nonunion are much more common and difficult problems. Left untreated, scaphoid nonunions naturally progress to a characteristic pattern of wrist arthritis, initially involving the radioscaphoid and capitolunate joints, referred to as scaphoid nonunion advanced collapse, or ‘‘SNAC wrist.’’ Unstable, displaced, or proximal fractures are prone to nonunion even with prolonged casting and should be considered for early open reduction because the outcome of surgery is more likely to be satisfactory for acute unstable or displaced fractures than for unstable or displaced nonunions. For unstable or displaced nonunions, open reduction and fixation with bone graft and screws is associated with a failure rate as high as 40% (46). Scapholunate Ligament Injuries.

Injuries to the scapholunate ligament result from the same mechanism of injury as scaphoid fractures. Like scaphoid fractures, these injuries may not be apparent on initial radiographs. Dynamic scapholunate dissociation may be obvious only on kinematic or stress-deviation radiographs. Conversely, bilateral benign congenital scapholunate diastasis may be confused

Figure 49 Scapholunate dissociation may require dynamic stress views to demonstrate (A, B). The natural progression of scapholunate dissociation is a pattern of wrist arthritis involving the radioscaphoid and capitolunate joints, referred to as scapholunate advanced collapse or ‘‘SLAC wrist’’ (C, F). Very proximal scaphoid avulsion fractures (D, E) are mechanically similar to scapholunate ligament injuries.

with an acute injury if both sides are not compared. Left untreated, scapholunate dissociation naturally progresses to a characteristic pattern of wrist arthritis, initially involving the radioscaphoid and capitolunate joints, referred to as scapholunate advanced collapse, or ‘‘SLAC wrist’’ (Fig. 49) (94). Treatment options include partial wrist fusion, proximal row carpectomy, and a variety of softtissue ligament reconstruction procedures. Capsulodesis procedures appear to be more successful than tendon graft procedures, although no current soft-tissue procedure reliably corrects scapholunate diastasis that is visible on radiographs (48). The outcome of injuries associated with scapholunate dissociation or partial ligament disruption is better than that for injuries resulting in complete disruption with a static instability pattern (48). Perilunate dislocations and fracture dislocations (Figs. 50 and 51) are severe wrist injuries that usually result in some degree of permanent wrist stiffness, even with ideal management. These injuries may not be appreciated on casual inspection; the most common report of inadequate evaluation is ‘‘something just isn’t right.’’ These injuries require open reduction and internal fixation and frequently require carpal tunnel release for acute traumatic neuritis.

Perilunate Dislocations and Fracture Dislocations.

Hook of Hamate Fractures. Hook of hamate fractures are often difficult to demonstrate on plain radiographs, and additional evaluation and management may be indicated on the basis of clinical suspicion (Fig. 52). Fractures of the hook of the hamate rarely heal with conservative treatment. The problem may mimic a variety of other problems, including carpometacarpal or capitohamate joint disorders.

Chapter 36:

Figure 50 Lunate dislocations may be missed in the emergency room through failure to recognize the typical states such as the ‘‘spilled teacup’’ sign on the lateral wrist X-ray (1A), and hand abnormalities that are not dramatic on the PA film (1B). It is unusual for the lunate to be displaced enough to be obvious to the untrained eye as in these examples of complete volar (2A, 2B) and dorsal (3A, 3B) lunate dislocations.

Figure 51 Perilunate fracture-dislocations represent high-energy injuries, which are usually best treated with provisional closed reduction and then definitive open repair. Unreduced radiographs (A, B) are usually difficult to interpret. Many variations exist, including trans-scaphoid transcapitate perilunate fracture-dislocation (C1), trans-scaphoid perilunate fracturedislocation (C2), transradial trans-scaphoid perilunate fracture-dislocation (C3), and transradial perilunate fracture-dislocation (C4). Traction films are helpful in defining the exact pathology, as in this trans-scaphoid perilunate fracture-dislocation (D).

Complications of Hand Surgery

455

Figure 52 Hook of hamate fractures have a high incidence of nonunion (A, B, C). The hook acts as a pulley for the profundus tendon to the small finger, and rough surfaces created by nonunion may result in tendon rupture (D).

Problems with this fracture include flexor tendon rupture because of abrasion against the fractured hook area. Tendon rupture is a serious complication, often resulting in permanent disability despite multiple operations and extensive therapy. Surgery to remove the fractured hook and inspect the tendons and nerves is indicated to minimize the risk of complications. Forearm Fractures Prone to Complications. Distal radius fractures account for about one of every six fractures and three of four forearm fractures seen in the emergency room. They are most common in children of both sexes between the ages of 6 and 10 years and in women between the ages of 60 and 69 years. These fractures may be classified according to a number of schemes, but no existing scheme correlates well with final functional outcome (73). A large number of operative and nonoperative treatment options have been recommended, many of which appear to give comparable results. Operative treatments include external fixation, percutaneous pinning, open reduction, and any combination of these. Poor final outcome is more likely when the fracture is initially very displaced, when the distal radioulnar joint is involved, when the radiocarpal joint is comminuted, when there is residual shortening of more than 2 mm, or when there is dorsal angulation of more than 15 . The outcome of closed reduction of intra-articular distal radius fractures is satisfactory in about four of five cases (47). However, redisplacement will occur with about one of three closed reductions, and the final outcome when redisplacement occurs and repeated closed manipulations are required will be good or excellent for only one of three fractures (47). There are conflicting reports regarding the importance of final fracture alignment on function, but one can make the argument that malunion should

456

Eaton

Figure 53 Distal radius malunions are common after closed reduction of an unstable fracture pattern, resulting in dorsal (left) or volar (center) angulation, shortening and loss of radial inclination (right). Functional outcome correlates poorly with radiographic changes.

be avoided (Fig. 53) because secondary surgery for distal radius malunion is successful in only three of four cases (71). Nonunion (Fig. 54) is uncommon but is more likely after severely displaced fractures because of the possibility of pronator quadratus or other soft-tissue interposition. Complex regional pain syndrome (Fig. 18) and finger stiffness occur to some degree in as many as one of three patients. Loss of motion is also common but unpredictable. Tendon rupture (19) may occur early or late, with either open or closed reduction, and may be related to fracture displacement, hardware irritation (Fig. 54) (13), or ulnar head prominence. Median or ulnar nerve compression may develop early or late after this fracture.

Posttraumatic arthritis is most common among young adults and is seen on radiographs of two out of three young patients, years after injury. Fortunately, the evidence on radiographs does not correlate well with the degree of symptoms, and many of these patients have no symptoms. Compartment syndrome of the forearm may develop in association with emergency reduction and stabilization of a distal radius fracture with a circumferential cast. However, compartment syndrome of the forearm may develop after a high-energy injury distal radius fracture even in the absence of circumferential cast or bandages (26) and may develop up to 48 hours after the initial injury (23). Men less than 50 years old are at particular risk of distal radius fractures (31), probably because they are more likely to suffer high-energy injuries. If clinical examination is unreliable, as in patients with obtundation or those whose symptoms may be masked by narcotics in hospital observation, repeated measurement of compartment pressures may be indicated during the first two days after injury. Carpal instability may develop, either as a discrete ligament injury or as a result of changes in the angle of the radiocarpal joint. Nonunion of associated ulnar styloid fractures is common and is usually painless. Prolonged recovery (6–12 months) is typical, as are long-term subjective symptoms such as pain, fatigue, and loss of grip strength. Such symptoms are reported by about half of the patients with an injury not related to compensation, in about four of five adult patients under the age of 45, and in essentially all patients with a compensation-related injury. Nevertheless, on average, three of four patients experience a satisfactory functional result after distal radius fracture.

Fractures of both bones of the forearm in an adult may result in a variety of problems. Complications are more common and prognosis is worse for displaced fractures and for open fractures. On the average, nondisplaced fractures take six to eight weeks to heal, and displaced fractures take three to five months. Satisfactory functional end results may be expected in about 8 of 10 patients with nondisplaced fractures and in about half of those with displaced fractures. As many as half of the patients will have obvious loss of forearm pronation, which may or may not be functionally significant. Loss of forearm rotation is most likely when fractures occur in the middle third of the forearm. Synostosis may lock the forearm in a fixed position of rotation. Nonunion occurs in as many as one of 10 patients (Fig. 55). Nonunion related to technique is more likely when semitubular plates are used, or when fewer than six cortices are engaged on each side of the fracture. Early protected motion appears to improve the odds of satisfactory final motion. Internal or external fixation is usually indicated for open or very unstable fractures; the surgeon must accept the

Fractures of Both Forearm Bones.

Figure 54 Distal radius nonunions are uncommon, but usually symptomatic due to progress of angulation and symptoms from distal radioulnar joint disruption. This patient developed progressive angulation and hardware failure following dorsal plating of a Galeazzi type metaphyseal distal radius fracture-dislocation (A, B). The original surgeon should have used a more sturdy plate and should have captured more cortices proximally. At exploration, soft-tissue interposition was found in the fracture line, and the extensor pollicis longus tendon was on its way to an attritional rupture, positioned behind the plate (C). The fracture was reduced, stabilized with a 3.5-mm plate and cerclage, and the distal ulna was used as bone graft (D).

Chapter 36:

Figure 55 Example of inadequate fixation. This patient underwent open reduction and internal fixation of a displaced both bone forearm fracture (A). Technical errors by the first surgeon include use of semitubular plates, which have inadequate rigidity for this type of fracture, and failure to achieve purchase of at least four cortices on each side of the fracture (B). Each of these errors increases the chance of nonunion. This patient developed nonunion with progressive angulation of the radius (C), which required repeat fixation with bone graft and compression plates (D).

risk that postsurgical infection may occur in as many as 1 of 20 patients. Proximal forearm fractures are associated with a variety of problems, including nonunion, nerve and tendon injuries, and synostosis. One-fifth to one-half of the patients can be expected to have significant permanent loss of forearm rotation. Open treatment of acute fracture or nonunion may be complicated by additional nerve injury or synostosis, more likely when injuries are open or classified as high energy. Synostosis, or cross-union between the radius and ulna, is much more common with proximal than with distal forearm fractures, occurring in about 1 of 15 patients with proximal fractures. Synostosis is more likely among children, when fractures are open fractures, when surgical access to both forearm bones is achieved with a single incision, and when the injuries are high energy. The results of surgery for correction of synostosis are poor when surgery is performed less than one year or more than three years after injury; even under ideal conditions, only one in five patients can be expected to regain as much as 50 of forearm rotation.

Fracture dislocations of the longitudinal forearm (Fig. 56) include three special combinations of injury: Galeazzi fracture-dislocation, Monteggia fracturedislocation, and the Essex-Lopresti lesion (52). Galeazzi fracture-dislocation refers to a fracture of the shaft of the radius associated with dislocation of the distal radioulnar joint. Monteggia fracturedislocation refers to fracture of the ulna with dislocation of the radial head. Each of these

Longitudinal Forearm Fracture Dislocations.

Complications of Hand Surgery

457

Figure 56 Combined fractures include three special combinations of injury: the Essex-Lopresti lesion (A), Monteggia fracture-dislocation (B), and Galeazzi fracture-dislocation (left and C). Galeazzi fracture-dislocation refers to a fracture of the shaft of the radius associated with dislocation of the distal radioulnar joint. Monteggia fracture-dislocation refers to fracture of the ulna with dislocation of the radial head. Each of these fracture-dislocation patterns is best treated with open fracture reduction and closed treatment of the dislocation. Essex-Lopresti lesion refers to longitudinal disruption of the radioulnar interosseous membrane and proximal migration of the radius, associated with fractures involving the proximal radioulnar joint, the distal radioulnar joint, or both sites. The most common presentation of Essex-Lopresti is associated with radial head excision for fracture, resulting in ulnocarpal impingement syndrome. Treatment is controversial. When diagnosed acutely in the context of an unreconstructable radial head fracture, Essex-Lopresti justifies use of a temporary radial head implant. Late surgical options include ulnar shortening osteotomy or the developing technique of ligament reconstruction with a tendon graft.

fracture-dislocation patterns is best treated with open fracture reduction and closed treatment of the dislocation. Essex-Lopresti lesion refers to longitudinal disruption of the radioulnar interosseous membrane and proximal migration of the radius; this lesion is associated with fractures involving the proximal radioulnar joint, the distal radioulnar joint, or both sites. The most common presentation of Essex-Lopresti lesion is associated with radial head excision for fracture, which results in ulnocarpal impingement syndrome. Treatment is controversial. When diagnosed acutely in the context of an unreconstructable radial head fracture, EssexLopresti lesion justifies the use of a temporary radial head implant. Late surgical options include ulnar shortening osteotomy or the developing technique of ligament reconstruction with a tendon graft. Radial Head Fractures. Fractures of the radial head often appear to be isolated injuries but are associated with distal radial ulnar joint pathology due to proximal migration of the radius, elbow arthritis, and loss of elbow motion. Early excision of radial head fractures is associated with a substantial

458

Eaton

complication rate, including proximal migration of the radius, which occurs to some degree in most patients (3). Efforts should be made to reconstruct rather than excise a fractured radial head. Skeletally Immature Forearm Fractures and Dislocations. ‘‘Isolated’’ radial head fractures in children are often associated with some degree of plastic deformation of the ulna, or ‘‘plastic’’ Monteggia fracture. Chronic pediatric radial head dislocation associated with plastic deformation of the ulna is frequently unrecognized, and late treatment requires open reduction and ulna osteotomy (74).

Complications of the Treatment of Fractures and Joint Injuries The complications associated with the treatment of fractures and joint injuries have been covered in previous sections. The most common of these are nonunion (Figs. 32, 55, and 57); infection or fracture related to external fixation (Fig. 57); arthrofibrosis and capsuloligamentous contractures; osteomyelitis (Fig. 58); tendon adhesions or rupture (Fig. 54); hardware prominence, exposure, or related fracture (Figs. 59 and 60); and complex regional pain syndrome (Fig. 18).

Figure 58 Osteomyelitis is uncommon in the hand, usually requiring the combination of severe contamination and crush injury as in this patient who was kicked by a horse and developed osteomyelitis after being treated for open metacarpal fractures. Draining wound (A) radiographs suggestive of septic non union (B), gross pus found at exploration (C).

because the dense fibrous compartments within the hand mask swelling and contour changes associated with a deep abscess. The diagnosis is based on suspicion, with the caveat that throbbing hand pain that keeps the patient awake at night and is associated with any other signs of infection indicates deep hand abscess until proven otherwise. Missed severe contamination has been discussed in the previous section on complications of missed complex wounds.

Complications of Infection Missed Diagnosis of Infection

Herpetic whitlow, a viral skin infection of the finger pulp, is commonly misdiagnosed as an abscess, felon, or paronychia. Diagnosis is suggested by a prodrome of pain and early signs of tiny vesicles and itching. Treatment by incision and drainage only prolongs recovery and should be avoided if possible. Missed deep infections (Fig. 61) of the hand are possible

Figure 57 Complications of external fixation. External fixers used for distal radius fracture fixation may result in sensory nerve injury and symptomatic neuroma, pin tract infections, osteomyelitis, and fracture through the pin site. (A) Healed index of metacarpal fracture following external fixation in a patient with osteogenesis imperfecta. (B) Pin loosening and periosteal reaction in another patient. (C, D) Index metacarpal fracture through pin site and complete loss of reduction in another patient.

Figure 59 Complications of hardware exposure. (A) Exposed tension band wire years following proximal interphalangeal joint arthrodesis. (B) Kirschner wire penetration of nail bed. The original surgeon should not have accepted this pin position, which resulted in a permanent split-nail deformity. (C) Improper choice of length for the Kirschner wire. Hardware should be either placed entirely beneath the skin or allowed to protrude well beyond the skin entrance site. Cutting pins off at skin level greatly increases the chance of pin tract infection.

Chapter 36:

Complications of Hand Surgery

459

Unsatisfactory results are more likely when hand infections involve anaerobes, Eikenella corrodens, or human bites (57). Quantitative cultures are the single

most sensitive and specific predictor of infection after microvascular free-flap reconstruction of complex extremity injuries and should be a routine part of this form of treatment. Complex wounds that contain more than 1000 organisms per cubic centimeter at the time of free-flap closure should be treated with a second surgical procedure involving flap elevation, repeated debridement, and closure (82). Atypical infections (57) may involve subcutaneous tissues or, more commonly, tendon sheath spaces. Mycobacteria species, most commonly Mycobacterium marinum, produce slowly progressing hand infections. Deep-space involvements with either typical or atypical infection usually follow puncture wounds that contaminate tendon sheath compartments or joint spaces. The most vulnerable areas in which apparently trivial wounds can contaminate deep spaces are the flexion creases of the fingers and the extension creases on the dorsum of the fingers. Diabetic hand infections, particularly in patients with diabetic chronic renal failure, are common, are frequently severe, and often result in tissue loss. Hand infections among such patients are frequently more severe than clinical examination indicates, and the surgeon must consider early extensive surgical debridement of the entire zone of inflammation (50). Gram-negative infections are common, and amputation is a common consequence. Pyarthrosis and septic arthritis of the small joints of the hand are more likely to be associated with a poor outcome if they occur more than 10 days after injury or are associated with severe trauma (75). The

Figure 61 Wound infection following carpal tunnel surgery. This patient developed worsening pain one week after opening carpal tunnel release. She was seen by her original surgeon, who was unimpressed by her physical findings and dismissed her as being narcotic seeking. She sought a second opinion, and at exploration later that day, was found to have pus throughout the carpal tunnel and tracking deep to the flexor tendons into the retroflexor space (Parona’s space). Deep wound infections are uncommon following elective clean hand surgery, and the signs of a deep-space infection may be subtle, requiring a high degree of suspicion for diagnosis.

Figure 62 Scaphoid nonunion eventually leads to progressive wrist arthritis in many patients. Uncommonly, nonunion may be the site of infection. This patient had undergone open reduction of a scaphoid fracture, and by history had a pin tract infection of one of the Kirschner wires used for scaphoid fixation, requiring early removal by the first surgeon. He presented over a year later with an abscess at the site of the prior pin tract infection, tracking deep to an infected scaphoid nonunion. Infection control involved radical debridement, including a proximal row carpectomy.

Figure 60 Following wrist arthrodesis, the distal screw hole securing the plate to the metacarpal (left) formed a stress riser which resulted in a symptomatic metacarpal fracture (right).

Complications of Infections and Infections Prone to Complications

460

Eaton

Figure 63 Infection of implants may arise late due to hematogenous seeding from distant infection. This patient with diabetes and rheumatoid arthritis on immunosuppressive therapy developed a plantar abscess following an unrecognized injury (left). She presented with infection of a long-standing functioning index metacarpophalangeal silastic joint spacer (middle). Organisms cultured from the site (right) were the same as those present in the foot. Implant infection is more likely when there is a permanent space where motion occurs adjacent to the implant, as is the case with silicone rubber joint replacements.

most common scenario for small-joint infection of the hand involves clenched fist bite injury (Fig. 21). Hematogenous seeding resulting in implant infection (Figs. 62 and 63) is an uncommon but catastrophic problem justifying the administration of prophylactic antibiotics during high-risk procedures for patients who have implants such as silastic joint spacers, which maintain a permanent open space around the implants. Tetanus may develop after hand injuries (38) and is most common in the context of parenteral drug abuse. More commonly, deep softtissue infections from parenteral drug abuse are polymicrobial and may present as gas-forming infection (Fig. 24), necrotizing infection, or suppurative thrombophlebitis (39). Treatment requires excision of the involved area, wide drainage, repeated debridement, and appropriate parenteral antibiotics.

REFERENCES 1. Hutchinson DT, McClinton MA, Wilgis EF, FriskMillner N. Drug and alcohol use in emergency hand patients. J Hand Surg (Am) 1992; 17(3):576–577. 2. Ablove RH, Moy OJ, Peimer CA, Wheeler DR. Early versus delayed treatment of enchondroma. Am J Orthop 2000; 29(10):771–772. 3. Leppilahti J, Jalovaara P. Early excision of the radial head for fracture. Int Orthop 2000; 24(3):160–162. 4. Colville RJ, Berry RB. A small contact burn injury requiring upper limb amputation. Burns 2000; 26(7): 656–658. 5. Gerard F, Obert L, Garbuio P, Tropet Y. An unusual complication of reimplantation of the hand: a false aneurysm of the radial artery anastomosis. Apropos of a treated case. Chir Main 1999; 18(2):149–152. 6. Gundes H, Cirpici Y, Sarlak A, Muezzinoglu S. Prognosis of wrist ganglion operations. Acta Orthop Belg 2000; 66(4):363–367.

7. Arons JA, Collins N, Arons MS. Permanent nerve injury in the forearm following radial artery harvest: a report of two cases. Ann Plast Surg Sep 1999; 43(3):299–301. 8. Rijbroek A, Vermeulen EG, Slotman BJ, Wisselink W, Rauwerda JA. Radiation-induced arterial disease. Ned Tijdschr Geneeskd 2000; 144(8):353–356. 9. Sessa C, Pecher M, Maurizi-Balzan J, et al. Critical hand ischemia after angioaccess surgery: diagnosis and treatment. Ann Vasc Surg 2000; 14(6):583–593. 10. Shemesh D, Mabjeesh NJ, Abramowitz HB. Management of dialysis access-associated steal syndrome: use of intraoperative duplex ultrasound scanning for optimal flow reduction. J Vasc Surg 1999; 30(1):193–195. 11. Tordoir JH, Leunissen KM. Arterial perfusion disorders of the hand in 9 patients with arteriovenous fistula for hemodialysis. Ned Tijdschr Geneeskd 1999; 143(21):1093–1098. 12. Wilhelmi BJ, Blackwell SJ, Miller J, Mancoll JS, Phillips LG. Epinephrine in digital blocks: revisited. Ann Plast Surg 1998; 41(4):410–414. 13. Bell JS, Wollstein R, Citron ND. Rupture of flexor pollicis longus tendon: a complication of volar plating of the distal radius. J Bone Joint Surg Br 1998; 80(2):225–226. 14. Schwabegger AH, Hussl H, Ninkovic MM, Anderl H. Replantation in childhood and adolescence. Longterm outcome. Unfallchirurg 1997; 100(8):652–657. 15. Stahl S, Norman D, Zinman C. Postoperative ulnar nerve palsy of the elbow. Harefuah 1997; 133(11):533–535, 590. 16. Ouellette EA, Kelly R. Compartment syndromes of the hand. J Bone Joint Surg Am 1996; 78(10):1515–1522. 17. Pearce H, Lindsay D, Leslie K. Axillary brachial plexus block in two hundred consecutive patients. Anaesth Intensive Care 1996; 24(4):453–458. 18. Stark RH. Neurologic injury from axillary block anesthesia. J Hand Surg (Am) 1996; 21(3):391–396. 19. Bonatz E, Kramer TD, Masear VR. Rupture of the extensor pollicis longus tendon. Am J Orthop 1996; 25(2):118–122. 20. Janezic TF, Arnez ZM, Solinc M, Zaletel-Kragelj L. One hundred sixty-seven thumb replantations and revascularisations: early microvascular results. Microsurgery 1996; 17(5):259–263. 21. Giandoni MB, Vinson RP, Grabski WJ. Ischemic complications of tubular gauze dressings. Dermatol Surg 1995; 21(8):716–718. 22. Landi A, Saracino A, Pinelli M, Caserta G, Facchini MC. Tourniquet paralysis in microsurgery. Ann Acad Med Singapore 1995; 24(4 suppl):89–93. 23. Simpson NS, Jupiter JB. Delayed onset of forearm compartment syndrome: a complication of distal radius fracture in young adults. J Orthop Trauma 1995; 9(5):411–418. 24. Hye RJ, Wolf YG. Ischemic monomelic neuropathy: an under-recognized complication of hemodialysis access. Ann Vasc Surg 1994; 8(6):578–582. 25. Naidu SH, Heppenstall RB. Compartment syndrome of the forearm and hand. Hand Clin 1994; 10(1):13–27. 26. Denolf F, Roos J, Feyen J. Compartment syndrome after fracture of the distal radius. Acta Orthop Belg 1994; 60(3):339–342. 27. Stanec Z, Skrbic S, Dzepina I, et al. High-energy war wounds: flap reconstruction. Ann Plast Surg 1993; 31(2):97–102. 28. Atabek U, Spence RK, Alexander JB, Pello MJ, Camishion RC. Upper extremity occlusive arterial

Chapter 36:

29.

30. 31. 32. 33. 34. 35.

36. 37. 38. 39. 40. 41.

42.

43. 44. 45. 46.

47.

disease after radiotherapy for breast cancer. J Surg Oncol 1992; 49(3):205–207. Farina C, Schultz RD, Feldhaus RJ. Late upper limb acute ischemia in a patient with an occluded axillofemoral bypass graft. J Cardiovasc Surg (Torino) 1990; 31(2):178–181. Stern PJ, Kastrup JJ. Complications and prognosis of treatment of mallet finger. J Hand Surg (Am) 1988; 13(3):329–334. Stockley I, Harvey IA, Getty CJ. Acute volar compartment syndrome of the forearm secondary to fractures of the distal radius. Injury 1988; 19(2):101–104. Smith PA, Hankin FM, Louis DS. Postoperative toxic shock syndrome after reconstructive surgery of the hand. J Hand Surg (Am) 1986; 11(3):399–402. Yuan RT, Cohen MJ. Lateral antebrachial cutaneous nerve injury as a complication of phlebotomy. Plast Reconstr Surg 1985; 76(2):299–300. Morgan WJ, Leopold T, Evans R. Foreign bodies in the hand. J Hand Surg (Br) 1984; 9(2):194–196. Reinstein L, Reed WP, Sadler JH, Baugher WH. Peripheral nerve compression by brachial artery-basilic vein vascular access in long-term hemodialysis. Arch Phys Med Rehabil 1984; 65(3):142–144. Aho K, Sainio K, Kianta M, Varpanen E. Pneumatic tourniquet paralysis. Case report. J Bone Joint Surg Br 1983; 65(4):441–443. Matsubara J, Seko T, Ohta T, et al. Arterio-venous fistula formation after hand replantation. Jpn J Surg 1983; 13(3):207–210. Thorne FL, Kropp RJ. Wound botulism: a lifethreatening complication of hand injuries. Plast Reconstr Surg 1983; 71(4):548–551. Sears N, Grosfeld JL, Weber TR, Kleiman MB. Suppurative thrombophlebitis in childhood. Pediatrics 1981; 68(5):630–632. Bandyk DF, Thiele BL, Radke HM. Upper-extremity emboli secondary to axillofemoral graft thrombosis. Arch Surg 1981; 116(4):393–395. Nunoo-Mensah J. An unexpected complication after harvesting of the radial artery for coronary artery bypass grafting. Ann Thorac Surg 1998; 66(3): 929–931. Rodriguez Montalban R, Martinez de Guerenu Alonso MA, et al. Acute ischemia of the hand as a complication of radial artery catheterization. Apropos of 2 cases following abdominal sarcoma surgery. Rev Esp Anestesiol Reanim 2000; 47(10):480–484. Grace TG, Omer GE. The management of upper extremity pit viper wounds. J Hand Surg (Am) 1980; 5(2): 168–177. Snelling CF, Hendel PM. Avascular necrosis of bone following revascularization of the thumb. Ann Plast Surg 1979; 3(1):77–87. Davis FM. Radial artery cannulation: influence of catheter size and material on arterial occlusion. Anaesth Intensive Care 1978; 6(1):49–53. Gupta A, Risitano G, Crawford RJ, Burke FD. The ununited scaphoid: prognostic factors in delayed and nonunions of the scaphoid. Hand Surg 1999; 4(1):11–19. Leung F, Ozkan M, Chow SP. Conservative treatment of intra-articular fractures of the distal radius -factors affecting functional outcome. Hand Surg 2000; 5(2):145–153.

Complications of Hand Surgery

461

48. Saffar P, Sokolow C, Duclos L. Soft tissue stabilization in the management of chronic scapholunate instability without osteoarthritis. A 15-year series. Acta Orthop Belg 1999; 65(4):424–433. 49. Leclercq C, Korn W. Articular fractures of the fingers in children. Hand Clin 2000; 16(4):523–534. 50. Gunther SF, Gunther SB. Diabetic hand infections. Hand Clin 1998; 14(4):647–656. 51. Adani R, Busa R, Castagnetti C, Castagnini L, Caroli A. Replantation of degloved skin of the hand. Plast Reconstr Surg 1998; 101(6):1544–1551. 52. Idler RS, Steichen JB. Complications of replantation surgery. Hand Clin 1992; 8(3):427–451. 53. Arnold DM, Cooney WP, Wood MB. Surgical management of chronic ulnar collateral ligament insufficiency of the thumb metacarpophalangeal joint. Orthop Rev 1992; 21(5):583–588. 54. Pun WK, Chow SP, So YC, et al. Unstable phalangeal fractures: treatment by A.O. screw and plate fixation. J Hand Surg (Am)1991; 16(1):113–117. 55. Davis TR, Stothard J. Why all finger fractures should be referred to a hand surgery service: a prospective study of primary management. J Hand Surg (Br) 1990; 15(3):299–302. 56. Pun WK, Chow SP, So YC, et al. A prospective study on 284 digital fractures of the hand. J Hand Surg (Am)1989; 14(3):474–481. 57. Dellinger EP, Wertz MJ, Miller SD, Coyle MB. Hand infections. Bacteriology and treatment: a prospective study. Arch Surg 1988; 123(6):745–750. 58. Strickland JW. Results of flexor tendon surgery in zone II. Hand Clin 1985; 1(1):167–179. 59. Brown PW. Sacrifice of the unsatisfactory hand. J Hand Surg (Am) 1979; 4(5):417–423. 60. Green JA, Tonkin MA. Ischaemia of the hand in infants following radial or ulnar artery catheterisation. Hand Surg 1999; 4(2):151–157. 61. Al-Qattan MM. Phalangeal neck fractures in children: classification and outcome in 66 cases. J Hand Surg (Br) 2001; 26(2):112–121. 62. Zyluk A, Walaszek I. Results of treatment for highpressure injection hand injuries. Chir Narzadow Ruchu Ortop Pol 2000; 65(4):367–374. 63. Germann G, Karle B, Bruner S, Menke H. Treatment strategy in complex hand injuries. Unfallchirurg 2000; 103(5):342–347. 64. Bartelmann U, Dietsch V, Landsleitner B. Fractures near the base of the first metacarpal bone–clinical outcome of 21 patients. Handchir Mikrochir Plast Chir 2000; 32(2):93–101. 65. Tredget EE. Management of the acutely burned upper extremity. Hand Clin 2000; 16(2):187–203. 66. Grant I, Belcher HJ. Injuries to the hand from dog bites. J Hand Surg (Br) 2000; 25(1):26–28. 67. Lanz U, Felderhoff J. Ischemic contractures of the forearm and hand. Handchir Mikrochir Plast Chir 2000; 32(1):6–25. 68. Barret JP, Desai MH, Herndon DN. The isolated burned palm in children: epidemiology and long-term sequelae. Plast Reconstr Surg 2000; 105(3):949–952. 69. Waikakul S, Sakkarnkosol S, Vanadurongwan V, Unnanuntana A. Results of 1018 digital replantations in 552 patients. Injury 2000; 31(1):33–40. 70. Salisbury RE. Reconstruction of the burned hand. Clin Plast Surg 2000; 27(1):65–69.

462

Eaton

71. Flinkkila T, Raatikainen T, Kaarela O, Hamalainen M. Corrective osteotomy for malunion of the distal radius. Arch Orthop Trauma Surg 2000; 120(1–2):23–26. 72. van Zuijlen PP, Kreis RW, Vloemans AF, Groenevelt F, Mackie DP. The prognostic factors regarding longterm functional outcome of full-thickness hand burns. Burns 1999; 25(8):709–714. 73. Gliatis JD, Plessas SJ, Davis TR. Outcome of distal radial fractures in young adults. J Hand Surg (Br) 2000; 25(6):535–543. 74. Kemnitz S, De Schrijver F, De Smet L. Radial head dislocation with plastic deformation of the ulna in children. A rare and frequently missed condition. Acta Orthop Belg 2000; 66(4):359–362. 75. Boustred AM, Singer M, Hudson DA, Bolitho GE. Septic arthritis of the metacarpophalangeal and interphalangeal joints of the hand. Ann Plast Surg 1999; 42(6):623–628. 76. Schwabegger AH, Anderl H, Hussl H, Ninkovic MM. Complex hand injuries. Importance of primary repair with free flaps. Unfallchirurg 1999; 102(4):292–297. 77. Sheridan RL, Baryza MJ, Pessina MA, et al. Acute hand burns in children: management and long-term outcome based on a 10-year experience with 698 injured hands. Ann Surg 1999; 229(4):558–564. 78. Giddins GE, Hill RA. Late diagnosis and treatment of crush injuries of the fingertip in children. Injury 1998; 29(6):447–450. 79. Aziz W, Noojin F, Arakaki A, Kutz JE. Avulsion injuries of the thumb: survival factors and functional results of replantation. Orthopedics 1998; 21(10):1113–1117. 80. Lewis HG, Clarke P, Kneafsey B, Brennen MD. A 10-year review of high-pressure injection injuries to the hand. J Hand Surg (Br) 1998; 23(4):479–481. 81. Pechlaner S, Hussl H. Complex trauma of the hand. Orthopade 1998; 27(1):11–16. 82. Breidenbach WC, Trager S. Quantitative culture technique and infection in complex wounds of the extremities closed with free flaps. Plast Reconstr Surg 1995; 95(5):860–865. 83. Godina M. Early microsurgical reconstruction of complex trauma of the extremities. Plast Reconstr Surg 1986; 78(3):285–292. 84. Melissinos EG, Parks DH. Post-trauma reconstruction with free tissue transfer—analysis of 442 consecutive cases. J Trauma 1989; 29(8):1095–102; (discussion 1102–1103). 85. Shen TY, Sun YH, Cao DX, Wang NZ. The use of free flaps in burn patients: experiences with 70 flaps in 65 patients. Plast Reconstr Surg 1988; 81(3):352–357. 86. Arnez ZM. Immediate reconstruction of the lower extremity—an update. Clin Plast Surg 1991; 18(3):449–457. 87. Scheker LR, Langley SJ, Martin DL, Julliard KN. Primary extensor tendon reconstruction in dorsal hand defects requiring free flaps. J Hand Surg (Br) 1993; 18(5):568–575. 88. Dawson WJ, Elenz DR, Winchester DP, Feldman JL. Elective hand surgery in the breast cancer patient with prior ipsilateral axillary dissection. Ann Surg Oncol 1995; 2(2):132–137. 89. Hartung HJ, Rupprecht A. The axillary brachial plexus block. A study of 178 patients. Reg Anaesth 1989; 12(1):21–24. 90. Regnard PJ, Soichot P, Ringuier JP. Local complications after axillary block anesthesia. Ann Chir Main Memb Super 1990; 9(1):59–64.

91. Stan TC, Krantz MA, Solomon DL, Poulos JG, Chaouki K. The incidence of neurovascular complications following axillary brachial plexus block using a transarterial approach. A prospective study of 1,000 consecutive patients. Reg Anesth 1995; 20(6):486–492. 92. Fanelli G, Casati A, Garancini P, Torri G. Nerve stimulator and multiple injection technique for upper and lower limb blockade: failure rate, patient acceptance, and neurologic complications. Anesth Analg 1999; 88(4):847–852. 93. Cahill N, King JD. Palm thorn synovitis. J Pediatr Orthop 1984; 4(2):175–179. 94. Watson HK, Ballet FL. The SLAC wrist: scapholunate advanced collapse pattern of degenerative arthritis. J Hand Surg (Am). 1984; 9(3):358–365. 95. Duncan RW, Freeland AE, Jabaley ME, Meydrech EF. Open hand fractures: an analysis of the recovery of active motion and of complications. J Hand Surg (Am) 1993; 18(3):387–394. 96. Kumar RJ, Pegg SP, Kimble RM. Management of extravasation injuries. Aust N Z J Surg 2001; 71(5):285–289. 97. Cicchetti S, Jemec B, Gault DT. Two case reports of vinorelbine extravasation: management and review of the literature. Tumori 2000; 86(4):289–292. 98. Raley J, Geisler JP, Buekers TE, Sorosky JI. Docetaxel extravasation causing significant delayed tissue injury. Gynecol Oncol 2000; 78(2):259–260. 99. von Heimburg D, Pallua N. Early and late treatment of iatrogenic injection damage. Chirurg 1998; 69(12):1378–1382. 100. Fullilove S, Fixsen J. Major limb deformities as complications of vascular access in neonates. Paediatr Anaesth 1997; 7(3):247–250. 101. Hayes AG, Chesney TM. Necrosis of the hand after extravasation of intravenously administered phenytoin. J Am Acad Dermatol 1993; 28(2 Pt 2):360–363. 102. Loth TS, Eversmann WW Jr. Extravasation injuries in the upper extremity. Clin Orthop 1991; (272):248–254. 103. Zyluk A. The sequelae of reflex sympathetic dystrophy. J Hand Surg Br 2001; 26(2):151–154. 104. DeLozier JB, Reaves L, King LE Jr., Rees RS. Brown recluse spider bites of the upper extremity. South Med J 1988; 81(2):181–184. 105. Eldad A, Wisoki M, Cohen H, et al. Phosphorous burns: evaluation of various modalities for primary treatment. J Burn Care Rehabil 1995; 16(1):49–55. 106. Su CW, Lohman R, Gottlieb LJ. Frostbite of the upper extremity. Hand Clin 2000; 16(2):235–247. 107. Tredget EE, Shankowsky HA, Tilley WA. Electrical injuries in Canadian burn care. Identification of unsolved problems. Ann N Y Acad Sci 1999; 888: 75–87. 108. d’Amato TA, Kaplan IB, Britt LD. High-voltage electrical injury: a role for mandatory exploration of deep muscle compartments. J Natl Med Assoc 1994; 86(7): 535–537. 109. Teot L, Griffe O, Brabet M, Gavroy JP, Thaury M. Severe electric injuries of the hand and forearm. Ann Chir Main Memb Super 1992; 11(3):207–216. 110. Bongard O, Fagrell B. Delayed arterial thrombosis following an apparently trivial low-voltage electric injury. Vasa 1989; 18(2):162–166. 111. Wang XW, Bartle EJ, Roberts BB. Early vascular grafting to prevent upper extremity necrosis after electric burns: additional commentary on indications for surgery. J Burn Care Rehabil 1987; 8(5):391–394.

PART VII Neurosurgical Complications

37 Postoperative Pain Management Edward Lubin, Michael J. Robbins, and Raymond S. Sinatra Department of Anesthesiology, Yale University School of Medicine, New Haven, Connecticut, U.S.A.

Pain is among the most common of complaints that patients make to health practitioners, and yet it remains poorly treated. Patients in pain historically have been victimized by disdain, underdosing of analgesics, and unnecessary physical and emotional suffering. We, practitioners, must exercise our responsibility to control pain most especially in the care of those patients recovering from major surgery and trauma. Why is this so? First, here we know pain is bound to occur, and yet in no case is pain control more frequently neglected. This irony is compounded insofar as persistent pain in these acute settings is by and large preventable. It is certainly more easily managed at this stage than at any other. Moreover, the costs of not treating are too high: the experience of pain sets off a cascade of physiologic and emotional events, the impact of which is felt by the patients and their family well beyond the passing of the acute stimulus (1,2). This cascade of events may facilitate the development of a chronic pain syndrome, a condition about which we have incomplete knowledge, and for which in many instances we have little to offer. Finally, the memory of pain often overshadows the initial insult in its long-term impact on the patient, a phenomenon that has led to the worldwide application of techniques to preempt both the experience and the memory of pain (3). The purpose of this chapter, then, is to present the current thinking about management of pain in the care of the trauma patient in the acute postoperative setting, and to describe our most recent breakthroughs in the treatment of chronic pain that may well develop later. The new treatments in both these phases include the introduction of more potent analgesics, more efficient means of their administration, and neuroaugmentative techniques based on a clearer picture of the mechanism of the development of chronic pain. All these efforts have dramatically improved the safety and efficacy of postoperative pain management, and have reduced the incidence and severity of persistent pain.

PATIENT AND CAREGIVER VARIABLES INFLUENCING PERIOPERATIVE ANALGESIC EFFECT The clinical inadequacy of traditional pain management is a continuing problem. More than 75% of adult patients, treated with on-demand doses of narcotic, continue toexperience moderate to severepain (4).What are we doing wrong? Analgesic underadministration has been related to a variety of factors, foremost of which is the physicians’ inadequate or erroneous pain assessment, likely stemming from their general lack of clinical experience and training in pain management. Few physicians have detailed knowledge of opioid pharmacology; the majority of them commonly underestimate the range of effective doses while, at the same time, overestimate analgesic duration and the risk of overdose. That nurses generally administer as little as 25% of the prescribed dose compounds the patient’s problem further (5). Practitioners, as a rule, administer opioid analgesics on a milligram per kilogram basis, and yet there is no evidence linking body weight to individual dose requirement. Age appears to be one of the most important variables in determining dose response and the degree of pain relief achieved following administration of opioid analgesics. Advancing age is generally associated with greater risk of unrelieved pain (6). While earlier studies (7) show reductions in pain perception in the elderly, more recent work shows little difference in pain intensity and perception with age (8). Thus, despite our knowledge of enhanced opioid sensitivity and decreases in opioid consumption with advancing age, these concepts must be viewed with healthy skepticism when treating the elderly in the postoperative period (9). The site, extent, and duration of surgery have a dramatic influence on both the intensity of postoperative pain and analgesic requirements. Thoracotomy, upper abdominal, and flank procedures require the most painful incisions, while laparoscopic, breast, and pelvic surgery are associated with lower pain intensity. The importance of the role of sex upon opioid

464

Lubin et al.

dose requirements is unclear. Epidemiologic studies show that many painful states parse out with greater female than male prevalence, particularly head and neck, musculoskeletal, and autoimmune diseases, and visceral conditions, while males show prevalence in complaints related to, for example, pancreatic disease, paratrigeminal syndrome (Raeder’s syndrome), brachial plexus neuropathy, and posttraumatic and cluster headaches (10). Merskey and Bogduk (10) have compiled an exhaustive epidemiologic list of sex differences in painful states; the neurobiological underpinnings of these remain unclear. Bias is known to exist insofar as women are more likely to report pain as a symptom, and in general are more eager to seek medical care (11). Sex hormonal influence, gender roles, and tools for clinical assessment may all play a part in these differences. Let us look at pain from the patient’s perspective. Pain is defined by the International Association for the Study of Pain as ‘‘an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage’’ (12). It is foremost an experience, with all of the personal and subjective character this implies. It is emotional, which at once defines it out of the realm of average practitioners’ expertise, unless they are particularly sensitive to the patient or have some psychiatric training. Finally, it is unpleasant, and therefore its relief, above other medical or surgical problems that may be present, takes primacy in the patient’s mind. Much of our knowledge of the neurophysiologic basis of pain is derived from animal studies. However, by our definition mentioned above and for clinical purposes, pain is an individual, and a human, phenomenon. Reaction to pain is arguably a conditioned behavior that reflects the values of a given individual and culture. The patient’s coping strategies must therefore not be ignored when fashioning postoperative pain treatment. These strategies are rooted in cultural differences, learned behavior, and one’s unique perception of pain and disability. It is difficult, and occasionally unwise, to generalize about the pain response of an individual or an entire group of people. On the other hand, an appreciation of such conditioning is in order, if it will help us understand someone’s pain, and thereby deliver better care. Responses to pain will vary, but may be divided into two very broad categories: stoic, which are those responses in which the patient expresses minimal discomfort verbally; and emotive, which are those we associate with patients who are vocal in their response to, and their demands to be relieved from, pain. Highly aggressive and angry patients also tend to get, and consume, more medication than patients whose coping styles are more passive. The phenomenon of ‘‘catastrophizing,’’ wherein patients focus on excessively negative thoughts, is a potent example of such a cognitive coping strategy with important ramifications for care (13,14). The successful clinician will treat pain based on the patient’s informed self-report,

carefully factoring in subjective variables, and thus foster an atmosphere of mutual trust. We have noted that undertreatment of pain is rooted in lack of knowledge. A more insidious cause for both undertreatment and lack of familiarity with appropriate treatment of pain is the clinician’s hesitancy to use opioids, out of both concern for the development of addiction, and the (not unwarranted) fear of recrimination by oversight agencies such as the federal government. This issue has been taken up with great care by Portenoy (15,16) with respect to cancer pain. It is regrettable that the relatively freer use of opioids in the postoperative period, as well in the emergency room setting, is based on our expectation of short-term use of these agents by the patient, or, as a manifestation of further disingenuousness, the brevity of the encounter between the practitioner and the patient. Many of our colleagues are only too happy to leave matters of addiction, diversion, or social stigma of opioid prescription use to those health care personnel with a more protracted relationship with the patient. This approach skirts our responsibility to understand the neurobiologic bases of pain and addiction, and to familiarize ourselves with the advantages and dangers of the drugs we use on a daily basis. Physicians also tend to limit opioid administration in patients with an ongoing history of substance abuse. However, more recent and enlightened thinking has helped to deliver well-supervised opioid and nonopioid analgesic therapy to patients with alcohol, cocaine, and heroin addictions (17,18). Patients with a history of chronic pain and opioid dependence often require larger amounts of opioid to compensate for the development of tolerance. Tolerance is a normal and predictable change of physiologic state, probably due to opioid receptor downregulation and enhanced drug metabolism and elimination, in which higher doses are needed to produce effects formerly achieved at lower doses. Physicians must therefore take into account both baseline opioid requirements, as well as that needed to control acute postsurgical pain. Just as neglecting aspects of the patient’s cultural, social, and medical history may lead to opioid underdose, underestimating the patient’s compromised physiologic state may lead to overdose. Decrements in cardiac, hepatic, and renal function are often associated with significant alterations in the volume of distribution, clearance, and excretion of most analgesic agents. For analgesics having high hepatic uptake and clearance, reductions in hepatic blood flow are accompanied by proportional decrements in the overall extraction rate and prolonged pharmacological effects. Agents that undergo biotransformation or are eliminated by the kidneys may produce serious adverse events in patients with renal failure unless dose adjustments are made (19). In the postsurgical setting, the physical and emotional responses to poorly controlled pain are undesirable (Table 1). In addition to ethical and humanitarian concerns for minimizing pain and suffering,

Chapter 37: Postoperative Pain Management

Table 1 The Acute Injury Response: Potential Benefits After Traumatic Injury Vs. Disadvantages in Controlled Postsurgical Settings Beneficial effects after traumatic injury

Adverse effects in patients recovering from surgery

Maintenance of intravascular Hypertension, hypervolemia, increased risk volume and mean of arterial pressure hemorrhage, stroke Maintenance of cardiac Tachycardia, arrhythmias, myocardial output and cerebral ischemia, perfusion congestive heart perfusion failure Enhanced hemostasis Hypercoagulable state, increased risk of arterial and deep venous thromboses Substrate mobilization, Hyperglycemia, negative nitrogen balance enhanced energy production Immobilization, minimizing Reduction in respiratory volume and flow further tissue injury rates Learned avoidance hypoxia, Anxiety, fear, demoralization, prolonged pneumonia convalescence

the physician wishes to avoid the pain-related anxiety, sleeplessness, and release of catecholamines and other stress hormones, all of which may have deleterious effects upon postsurgical outcome. This is particularly true in elderly or critically ill populations (19,20).

ANATOMY AND PATHOPHYSIOLOGY Pain may be defined as the conscious awareness of tissue injury (20,21). Therefore, understanding the basic anatomy and physiology of how this injury is translated to conscious awareness of pain is essential to effective management. Pain perception can be divided into two major components. The sensory-discriminative component describes the location and quality of the stimulus. The affective-motivational component underlies suffering and emotional components of pain and is responsible for learned avoidance and other behavioral responses (21,22). This latter component will be discussed in the clinical sections throughout the chapter. The first component is pain perception (nociception), which reflects the activation of nociceptors following thermal, mechanical, or chemical tissue injury, afferent transmission to the spinal cord, and relay via dorsal horn to higher cortical centers. Tissue injury causes cellular disruption, resulting in escape of intracellular potassium ions (Kþ) and the release of bradykinin (BK), serotonin (5-HT), and prostaglandin (PG), all of which are potent activators of the peripheral endings of nociceptive neurons (23). BK and PG stimulate release of substance P, which in turn sensitizes additional nociceptive neurons at sites adjacent to the injury (23). This process of recruitment and sensitization of peripheral nerve endings underlies hyperalgesia, an altered state of sensibility in which the intensity of pain sensation induced by noxious stimulation is greatly increased. Peripheral nerve endings are for the most part free nerve endings of myelinated A-delta and unmyelinated C-fibers (20). These fibers are the peripheral extension of bipolar sensory neurons, and are

465

classified according to their responses to mechanical, thermal, and chemical stimulation, and to the conduction velocity of their axonal fiber (21). The cell bodies of these peripheral afferents are found in the dorsal root ganglia. The central component projects to the dorsal horn of the spinal cord, where they synapse upon the second-order sensory neurons, the majority of which coalesce to form the spinothalamic tract, the principal ascending spinal pathway for pain. There are three types of this dorsal horn neuron, based on its location in the dorsal horn and on the source of its input. Wide–dynamic range neurons are found in laminae I to VI (principally in IV–VI) and receive signals from low- and high-threshold mechanoreceptors and polymodal C-fibers (23). As their name implies, these neurons receive multiple inputs and increase their signal frequency (intensity) over a wide range of externally applied mechanical deformation and heat. The other two types of dorsal horn spinothalamic neuron are both called nociceptive specific (21). Both types are found mainly in the marginal layer of the dorsal horn. The first of these receives input from mechano- and thermoreceptors and responds primarily to firm pressure or pinch, while the second, receiving input exclusively from A-delta highthreshold mechanoreceptors, fires in response only to noxious cutaneous mechanical stimulation (21,22). The spinothalamic tract synapses primarily in the thalamus, in its ventral portion, but sends branches to the spinal reticular formation and the periaqueductal gray matter (20). Third-order sensory neurons carry nociceptive information from the thalamus to cortex. Because of the numerous synapses and areas of connection, there is potential for the nervous system to modulate the transmission of the painful stimulus at many levels. This modulation can take the form of molecular memory, peripheral and central sensitization, descending pain inhibitory pathways, the neuroendocrine response, and the elaboration of endogenous opioid transmitters.

Molecular Memory Molecular memory began as a concept in the study of memory function in Aplysia in the 1980s, when it was discovered that oncogenes could be found in stimulated neurons. We can now expand this concept to include a wide variety of changes in the environment of the neuron, including protein-binding characteristics, receptor surface density, intracellular calcium concentration, and N-methyl-D-aspartate (NMDA) receptor function. These molecular changes lead to changes in the firing pattern which individual neurons demonstrate in response to pain stimulation.

Central Sensitization Central sensitization can in turn be defined as the result of modulation of molecular memory. In general, it is

466

Lubin et al.

the ‘‘priming’’ of the central nervous system (CNS), such that incoming pain messages to the system generate subsequent responses out of proportion to the original stimulus, and non-nociceptive stimuli thus generate ‘‘pain’’ responses. The phenomenon of central sensitization was hypothesized by Woolf and Chong (25), who developed an animal model in 1983 to investigate and subsequently prove the hypothesis. Following the discovery, central sensitization has been documented as a cause of somatosensory hypersensitivity observed in patients after surgical trauma (26). Central sensitization begins with the induction of activity in peripheral C-fibers, which can be activated by mechanical, thermal, or chemical stimuli. During the initiation phase of central sensitization, pain is induced by a variety of peripheral substances such as potassium and hydrogen ions, histamine, 5-HT, BK, and PG E2, all found in the ‘‘inflammatory soup’’ in peripheral tissue after trauma. Glutamate is the most ubiquitous of the CNS neurotransmitters; thus as was inevitable, its role in the transmission of pain has been elucidated (24). The NMDA receptor, one of glutamate’s principal binding sites, in turn has been found to play a crucial role in the development of central sensitization. Indeed, the principal component of central sensitization is mediated by the activation of presynaptic NMDA receptors, located on the central terminals of C-fibers. This component of central sensitization is marked by increased intracellular levels of calcium, which promote excitability of dorsal horn neurons. Theoretically, then, central sensitization can be forestalled by NMDA receptor antagonism, voltage-dependent calcium ion channels, as well as G-protein–coupled receptors such as P NK1 and mGluR (25,26). This exciting lead in pain relief, examined in detail in animal studies, has not panned out in the treatment of pain in humans. Those NMDA blockers currently available in clinical practice, such as ketamine, amantadine, memantine, and dextromethorphan, and the experimental drug CHF3381, all show some clinical effect, but their efficacy is far less than that which would be predicted by theoretical and experimental work (27).

Peripheral Sensitization and Neuroendocrine Responses Nociceptive mediators, released during tissue injury, activate primary afferent neuronal endings, and increase regional nociceptive sensitivity to further tissue damage and painful stimulus. This primary hyperalgesia is worsened by ambulation, incentive spirometry, or physical therapy and leads to an increased dynamic or ‘‘effort-dependent’’ pain. Nociceptive impulses also impact upon and alter the activity of hypothalamus and adrenal cortex and medulla. These changes, termed the neuroendocrine or ‘‘stress’’ response to injury, are characterized by an increased secretion of catabolic hormones such as

cortisol, glucagon, growth hormone, and catecholamines. Innumerable chemical mediators participate in this response, and many novel mediators recently have been discovered. Nerve growth factor, first purified by Levi-Montalcini and colleagues in 1987, is the prototypical neurotrophic factor, and has been shown to play an exciting functional role as neuromodulator, coordinating inflammatory and neuroendocrine responses, with increased endogenous levels in acute inflammation as well as chronic pain states (28). Such alterations mediate enhanced mobilization of substrate, hyperglycemia, and a negative nitrogen balance. This catabolic response leads to muscle wasting, impaired immunocompetence, and decreased resistance to infection. Related to this is the concept of sympathoadrenal activation. Surgical injury is associated with marked increases in plasma epinephrine and norepinephrine concentrations. Increased sympathetic tone has been associated with an increased risk of perioperative myocardial ischemia in patients with poorly compensated coronary artery disease. Severe pain is commonly associated with an impaired ability to ambulate and decreased venous flow. Catecholamines, angiotensin, and other factors associated with surgical stress increase platelet–fibrinogen activation, while surgical manipulation in and around the pelvis may damage venous conduits, diminishing blood return from the lower extremity. These factors underlie Virchow’s triad of venous stasis, hypercoagulability, and endothelial injury, which increases the risk of clot formation, deep vein thrombosis, and pulmonary embolus (28). Humoral and neurologic alterations in and around the site of injury may be responsible for increased postoperative discomfort and disability. Continued activation of nociceptors secondary to neural compression, stretch, infection, hematoma, and edema can explain ongoing or progressive worsening of acute pain as well as prolonged disability and impaired rehabilitation. In these settings, continued periosteal and muscle irritation may initiate reflex motor responses leading to spasm and myofascial pain. Heightened reflex activity in sympathetic efferent fibers results in vasoconstriction and continued nociceptor sensitization. Alterations in blood flow and efferent outflow may be responsible for persistent pain syndromes.

PAIN SERVICES AND THERAPEUTIC OPTIONS FOR POSTOPERATIVE ANALGESIA Nearly 25 million surgeries are performed annually in the United States, and most require some form of pain management. Major goals in pain management are to (i) reduce the severity of acute postoperative or posttraumatic pain in both adult and pediatric populations; (ii) introduce new and potentially more effective methods of providing analgesia; (iii) educate patients to effectively communicate increases in pain

Chapter 37: Postoperative Pain Management

intensity; (iv) teach caregivers about the need for prompt evaluation and treatment; and (v) by these efforts reduce the number of pain-associated complications, improve outcome, and shorten hospital stay. Postsurgical pain service is generally multidisciplinary, with anesthesiologists, surgeons, nurses, pharmacists, and nursing assistants playing important roles. Pain intensity has increasingly become recognized as the ‘‘fifth vital sign’’: that is, a variable (approaching zero) associated with homeostasis, to be measured frequently and maintained at a predetermined optimized value. Pain intensity may be reliably measured using visual analog scale scores, verbal pain scores, or ‘‘Oucher’’ type picture scales for very young, elderly, and mentally impaired patients (29). These scores are incorporated in the patient’s chart on a regular basis to gauge overall analgesic effectiveness and quality assurance. Continued staff education and the use of standardized orders significantly reduce the number of nurse calls for assistance.

PARENTERALLY AND ORALLY ADMINISTERED ANALGESICS A cautionary note on PRN: although ‘‘as-needed’’ (or PRN) intramuscularly and orally administered analgesic regimens remain the mainstay of acute pain management, a large number of papers have documented inadequacies associated with such therapy (3). A major deficiency relates to the timing of the analgesic dose because patients often wait too long to seek pain relief and request medication, and staff may not be able to immediately deliver it. A second problem associated with PRN dosing relates to its ineffectiveness in maintaining therapeutic plasma concentrations (30). When intramuscular (IM) analgesics are administered every three to four hours, concentrations in plasma may equal or exceed minimal effective analgesic concentration only 30% of the intervening time. The provision of pain medication on a traditional PRN every-three-to-four-hour dosing schedule involves an elaborate sequence of events, which inevitably delays administration, resulting in repetitive cycles of increasing pain. Because pain is usually not treated as an emergency, the length of time that the patient waits for an analgesic is dependent principally upon the nursing workload at the time of the request. Once the level of pain is deemed significant to warrant treatment, the nurse must then take the time-consuming steps (i.e., medication signout, preparation, etc.) necessary to administer the dose. These steps delay the onset of effective relief and worsen pain-induced anxiety, helplessness, and sleep deprivation. Because the dose administered is often relatively large and absorption is erratic and prolonged, the initial analgesic effect is often followed by sedation and some degree of respiratory depression (31).

467

Oral Analgesics Oral administration of analgesics offers a safe, simple, and cost-effective method of controlling postoperative pain that should always be considered in patients tolerating oral diet and experiencing moderate discomfort. Oral dosing is best employed during the rehabilitation phase following surgery; however, alterations in gastrointestinal function and perfusion that follow exposure to general anesthesia and traumatic injury markedly reduce the reliability and effectiveness of such therapy, particularly during the emergent phase (29). Oral opioids, including morphine, meperidine, hydrocodone, and oxycodone, and compounded opioid preparations containing acetaminophen, aspirin, or ibuprofen (e.g., oxycodone in Percocet1, Tylox1, Percodan1, and Combunox1; hydrocodone in Vicodin1, Lortab1, and Lorcet1), provide effective relief for patients complaining of moderate to severe pain. (While oxycodone is available as an uncompounded oral product, hydrocodone is not; an upward titration of hydrocodone is therefore limited by associated acetaminophen increase.) Orally administered morphine and meperidine are poorly absorbed and undergo significant enterohepatic metabolism. For this reason, onset is delayed, duration is less predictable, and dose requirements are high, perhaps two to three times higher than parenteral requirements. Oxycodone and hydrocodone have higher oral effectiveness, as they are more reliably absorbed and are less likely to undergo first-pass hepatic metabolism. Sustained-release oral opioid preparations are available for morphine (MS-Contin1) and oxycodone (Oxycontin1). These offer wider dosing schedules, which is convenient for the patient and can reduce hospital labor. These preparations avoid the frequent peaks and trough plasma levels that lead to the cycles of euphoria, dysphoria, and sedation at the peak, and discomfort at the trough. This greater analgesic uniformity allows for 8 to 12 hours of pain relief per dose, and is ideally suited for patients who have opioid-responsive chronic pain, or who are engaged in prolonged physical rehabilitation. Over the last few years, Oxycontin has been increasingly prescribed for pain control in acute postsurgical settings. Personal communication to R. Sinatra determined the dose relationship between oral Oxycontin and intravenous (IV) patient-controlled analgesia (PCA) morphine in patients recovering from general, gynecological, and orthopedic surgery. They found that the initial dose of Oxycontin needed to maintain effective pain control was only 1.3 times higher than the prior day’s dose of morphine. We have found that the relationship is closer to 1:1; that is, if, on the previous day, the patient required 40 mg of morphine, the initial dose of Oxycontin is 40 mg/day, or 20 mg every 12 hours. Oxycontin 20 to 40 mg every 12 hours may also be utilized as an effective transitional analgesic in patients previously treated with

468

Lubin et al.

epidural analgesics during the first 24 to 48 hours following surgery (16). Oral Oxycontin is started either one to two hours following discontinuation of the epidural infusion or as soon as the patient notices a slight increase in pain intensity. Onset of analgesia is noted within 30 to 60 minutes, and highly uniform pain relief is maintained for 10 to 12 hours. The early introduction of Oxycontin is generally well tolerated, and allows the duration of epidural therapy to be shortened from 72 hours to 24 to 48 hours, all the while minimizing the need for parenteral analgesics (32). The safety and effectiveness of Oxycontin versus doses of immediate-release oxycodone given PRN and ‘‘by the clock’’ was evaluated in patients recovering from anterior cruciate ligament repair (31). Patients treated with sustained-release formulation required less drug and experienced superior pain control than individuals treated with oxycodone. They also benefited from a reduction in opioid-related side effects such as nausea, vomiting, sedation, and sleep disturbance. Patients treated with oral immediate- and sustained-release opioid preparations are always at risk for ileus and constipation, mandating that such therapy be supplemented with a bowel regimen that includes stool softeners, bulk laxatives, and occasional enemas. Less potent analgesics may be prescribed in patients with mild to moderate pain, and may be given alone, or as supplements to opioid therapy or regional neural blockade. These include tramadol, a weak noncontrolled opioid, Ultram ER, a tramadol extended-release oral preparation, tramadol compounded with acetaminophen (Ultram1), and a variety of nonsteroidal anti-inflammatory drugs (NSAIDs). Rofecoxib (Vioxx1), celecoxib (Celebrex1), and valdecoxib (Bextra1) represent a newer, and potentially safer, class of NSAID, termed the cyclooxygenase-2 (COX-2) inhibitors. These agents selectively block COX-2, thereby inhibiting PG synthesis following tissue injury. Unlike other NSAIDs, however, they do not block COX-1, which constitutively maintains platelet function and gastric mucosal integrity. (In the fall of 2005, both Vioxx and Bextra became unavailable in the United States; Vioxx was removed by the Food and Drug Administration (FDA) over concerns of excessive cardiovascular risk; Bextra was withdrawn voluntarily by its manufacturer over concerns related to the very uncommon occurrence of Stevens–Johnson syndrome. The degree of hazard posed to the public by these drugs, compared to their potential benefit, is yet unclear; it is entirely possible that by publication of this writing, both agents would have returned, at some recommended dose level, to the American market.) Meloxicam (Mobic1) is another NSAID, which is similar in this regard, as it has relative specificity to the COX-2 isoform. Following long-term use of these drugs, the incidence of gastric ulcer is similar to that observed with placebo, and significantly lower than that observed with nonselective NSAIDs. Despite this increase in safety, the

COX-2 inhibitors should not be given to patients with active bleeding ulcer. COX-2 is also expressed in the kidney: here the enzyme is inducible in response to salt restriction and hypovolemia, and participates in renin release. Thus, COX-2–specific agents are contraindicated in prerenal azotemia. The COX-2 inhibitors have no effect on bleeding time, and their safety and effectiveness have been demonstrated in several postorthopedic surgical models. For example, rofecoxib 50 mg/day for five days is as effective as naproxen sodium in reducing pain intensity and opioid dose requirements in patients recovering from orthopedic surgery (26). There has been only one comparison of rofecoxib versus celecoxib for postoperative pain management (28). In this study, rofecoxib 50 mg was associated with a more prolonged duration and greater reduction in morphine consumption than was Celecoxib1 200 mg (31).

Parenteral Analgesics Morphine remains the standard opioid analgesic for control of orthopedic injuries, as it effectively blocks musculoskeletal and visceral pain. Onset of analgesia occurs within five minutes after IV, and 15 minutes following IM administration, while duration ranges from two to four hours, depending upon dose and site of administration. Administration of morphine may release histamine and has been associated with hypotension and biliary colic. Meperidine (Demerol1) and hydromorphone (Dilaudid1) are useful alternatives in patients intolerant of morphine’s adverse effects. Meperidine’s parenteral potency is one-tenth that of morphine, while its duration of effect is only twothirds as long. Doses exceeding 600 mg/day may result in seizures secondary to accumulation of its metabolite, normeperidine. Hydromorphone is approximately five times as potent as morphine, but has a more rapid onset of analgesia and lower incidence of adverse effects. Fentanyl is a potent analgesic advocated for use in patients with marked hemodynamic instability or individuals highly tolerant to opioid analgesics. Bolus doses (50–200 mg) and IV infusions (50–200 mg/hr) of fentanyl are particularly useful in patients who remain intubated. The most serious complications associated with parenteral opioids, as with oral preparations, include constipation, increasing sedation, and progressive respiratory depression. The rapid onset and increased potency of parenterally delivered opioids, however, mean that the practitioner must be more vigilant in their use. Ketorolac (Toradol1) is a potent NSAID available in parenteral form (32). Ketorolac reduces pain intensity by nonselectively reducing PG synthesis at the peripheral site of injury, as well as in pain-processing circuits in the CNS (32,33). Ketorolac in doses of 30 to 60 mg is as potent as 10 mg of morphine, and is particularly useful in managing posttraumatic musculoskeletal pain. Although it is not associated with excessive sedation or respiratory depression, major

Chapter 37: Postoperative Pain Management

side effects, including increased risk of hemorrhage, gastric ulceration, and renal toxicity, limit its usefulness during the acute and emergent phases following injury. To minimize these complications, ketorolac dose should be limited to 7.5 to 15 mg every 6 hours for a maximum of 48 hours. Injectable COX-2 inhibitors and IV acetaminophen derivatives that offer greater safety and equivalent analgesic effects as ketorolac will soon be available for perioperative use.

INTRAVENOUS PATIENT-CONTROLLED ANALGESIA PCA allows patients to titrate small doses of pain medication in amounts proportional to a perceived pain stimulus. The technique avoids cycles of excessive sedation and ineffective pain control observed with ‘‘by the clock’’ and PRN IM dosing, and limits variation related to inappropriate pain assessment on the one hand and unpredictable drug absorption on the other (32). Patients control the dose frequency (within prescribed time limits) and thereby correct for individual differences in pain perception and pharmacokinetics and delays in administration. It must always be borne in mind that while the patient is titrating the drug, it is being done within the careful limits set by the administering physician; i.e., patient self-titration should not be confused with self-administration. Furthermore, PCA prescription is not an alibi for inadequate patient assessment and monitoring. Commercially developed PCA systems incorporate microprocessors that allow the patient to interact with an infusion pump connected to the established IV line. A patient activates the pump by pressing a button connected to the apparatus. A preprogrammed amount of opioid (incremental bolus dose) is then administered over 10 to 30 seconds, and a preset ‘‘lockout’’ time interval begins, within which a second dose will not be delivered. A prolonged lockout interval or inadequate incremental bolus may diminish analgesic effectiveness (34). Conversely, too large incremental dose increases the number of treatment failures related to intolerable side effects. Patients inevitably find themselves titrating pain against sedation, excessive nausea, or other side effects. Interestingly, patients report that they are usually willing to accept some amount of pain in order to have a clear sensorium (30,32,34). One of the keys to successful initiation of PCA is the administration of an opioid ‘‘loading’’ dose. This first dose provides baseline plasma concentrations of analgesic, which can then be augmented by patientinitiated boluses. In general, morphine (5–15 mg) or hydromorphone (0.5–3 mg) is titrated to patient comfort for this purpose. While morphine remains the standard and most widely administered PCA analgesic, it does have drawbacks, including delay to peak analgesic effect, sedation, and histamine release (35). Other opioids employed for IV-PCA, and suggested dosage and lockout intervals, are presented in Table 4. More sophisticated PCA devices incorporate a continuous (basal) rate infusion in addition to

469

patient-activated bolus doses on demand. Patients receiving basal opioid infusions should be monitored carefully, as the continuous delivery of opioid bypasses the inherent safety of PCA, and may result in progressive sedation. Conversely, some newer PCA devices incorporate a ‘‘safety setting,’’ whereby a three- or four-hour limit on total drug delivery may be set. While superficially appealing, this ‘‘safety setting’’ technique reincorporates all the drawbacks of PRN, i.e., cyclical blood levels of opioids, while eliminating the very advantage of the PCA technique: that is, patient self-titration of the drug. Using a ‘‘safety setting’’ encourages lax dose, lockout interval, and basal adjustments in an environment of false security, and does nothing to establish a PCA dose schedule that can be converted to a postoperative oral (P.O.) regimen. Furthermore, safety, as has been stated, is established in patient assessment and monitoring. The most appealing advances in PCA technology are those which allow the nurse and physician to record the number and times of dose demand against the doses delivered. This provides the most useful data in adjusting delivery parameters to steady opioid blood levels and thereby optimizes therapy (32,34). The two most common reasons for patient dissatisfaction and failure of the PCA method of treatment are inadequate analgesia and nausea and vomiting. In large measure, both of these complications can be treated well with patient education. It must be said that PCA therapy must begin with the patient’s complete understanding of the self-titration method. Countless cases of poor results with PCA therapy stem from patients simply not understanding how to push the button, and when. Patients must be trained to treat their pain before it overwhelms them, for example, just prior to physical therapy, or before any form of movement that might increase discomfort. Similarly, the patient should know to avoid the button during periods of relative comfort. If the patients have intolerable side effects, they may call the nurse or the pain service to ask for help or another medication; they should be instructed not to abandon the PCA device in the meantime. Finally, concerned relatives (and nurses!) should never push the PCA button for the patient (32,34).

SPINAL OPIOID ANALGESIA The administration of opioid analgesics into the intrathecal or epidural space, termed spinal opioid analgesia, is perhaps the most powerful method of controlling pain in many clinical settings (35,36). Following an epidural injection, a portion of the analgesic crosses the dura to enter the cerebrospinal fluid (CSF). Intrathecal (or ‘‘spinal’’) injection administers this medicine directly to the CSF. Opioid molecules then only bind to receptors in dorsal horn, effectively blocking pain transmission at the first synapse in the CNS. Intrathecal and epidural opioids provide greater

470

Lubin et al.

analgesic potency than similar doses administered parenterally. In general, epidural doses of hydrophilic opioids such as morphine and hydromorphone exhibit the greatest potency, while doses of highly lipophilic opioids behave similar to IV (37–39). A second advantage noted with spinal opioids is the ‘‘selectivity’’ of analgesic effect, which is maintained in the absence of motor or sympathetic blockade (38,39). A single intrathecal bolus (0.25–0.5 mg) or multiple epidural boluses (2.5–5 mg) of morphine are commonly used for the control of pain following trauma and lower extremity orthopedic surgery (34). Doses are usually administered via spinal needles or epidural catheters inserted at lumbar interspaces. In general, analgesic onset is appreciated after 30 to 60 minutes, and peak effect at 90 to 120 minutes. Duration ranges from 12 to 24 hours. Twenty-four hour milligram analgesic requirements are reduced to about one-tenth the parenteral dose. While this method is effective and simple, the major drawback is the abrupt rise in CSF opioid concentrations following each epidural or spinal bolus, which may result in a high incidence of occasionally serious adverse effects.

Continuous Epidural Analgesia Infusions of epidural opioids, opioids diluted in local anesthetics, and local anesthetics alone have been advocated as methods to control postoperative pain (40,41). Continuous infusion permits analgesia to be more precisely titrated to the level of pain stimulus and rapidly terminated if problems arise. This avoids the high peak concentrations that follow intermittent epidural boluses and reduces the risk of delayed respiratory depression (39–41). Continuous infusion technique also provides greater therapeutic versatility because shorter-acting opioids and dilute local anesthetic solutions may be administered. Continuous infusions of morphine (Duramorph1, Astramorph1) and hydromorphone offer effective epidural analgesia for patients recovering from a variety of surgical procedures (39–41). Epidural infusions of morphine mixed with dilute bupivacaine provide excellent analgesia, are associated with an extremely low incidence of serious adverse events, and may be safely administered to patients recovering on routine postsurgicalwards(42).Hydromorphoneinfusion compares favorably with morphine. Over 90% of patients receiving continuous lumbar epidural infusions of lowdose hydromorphone for postthoracotomy analgesia report either no pain or only mild discomfort. Hypoventilation, pruritus, and nausea are milder than that observed with equipotent doses of epidural morphine. Patients on continuous epidural infusions of hydromorphone experience effective pain relief, with less sedation and pruritus, than patients on continuous infusions of morphine at equivalent doses (43). Fentanyl is commonly administered as continuous epidural infusion because of its rapid onset, and short duration facilitates analgesic titration. Epidural

fentanyl dose requirements are high (40–70 mcg/hr) and are often equivalent to that required with IV infusions of fentanyl. Improved analgesic effectiveness may be achieved by combining epidural fentanyl with dilute solutions of bupivacaine (44,47). Epidural infusion of local anesthetics, especially bupivacaine and ropivacaine, offers reliable and effective analgesia in a segmental fashion for patients recovering from lower extremity vascular and orthopedic surgery or trauma. Local anesthetic infusions are particularly useful in patients who are sensitive to the side effects of opioid analgesics. Local anesthetic therapy is associated with sensory-motor and sympathetic blockade. Hypotension and impaired micturation, however, occur more frequently with epidural local anesthetics than with opioids.

Patient-Controlled Epidural Analgesia Patient-controlled epidural analgesia (PCEA) offers higher analgesic efficacy and lower dose requirement than IV-PCA, while providing greater control and patient satisfaction than either single doses or continuous infusions of epidural opioids. This technique combines the control and titratability of patient-controlled administration with the higher analgesic efficacy associated with neuraxial analgesia (45). Morphine’s latency to peak effect and risk of delayed onset respiratory depression represent undesirable characteristics for PCEA, therefore hydromorphone and more lipophilic opioids such as fentanyl, which offer rapid onset and greater titratability, have become the agents of choice in this setting (46,48). Epidural dose requirements for single bolus techniques, continuous epidural infusion, and patientcontrolled epidural infusion are presented in Table 9.

ADVERSE EVENTS AND CONTRAINDICATIONS OF NEUROAXIAL ANALGESIC TECHNIQUES Epidural and intrathecally administered opioids are associated with a number of annoying and occasionally serious adverse effects including pruritus, nausea, urinary retention, somnolence, and respiratory depression (41,42,45). Treatment protocols have been developed which can decrease the incidence and severity of side effects and improve patient safety while maintaining effective analgesia. The presence of side effects should be assessed frequently and treated quickly in order to minimize morbidity and patient dissatisfaction. Pruritus and nausea are the most common side effects associated with epidural or spinal opioids; however, respiratory depression is the most feared complication (41–43,45,46,49). Respiratory depression following epidural or intrathecal morphine occurs at two different intervals (46,49). An early mild phase observed soon after administration is followed by delayed depression occurring between 8 and 12 hours

Chapter 37: Postoperative Pain Management

later. Risk factors for respiratory depression are listed in Table 8. The speed with which epidural opioidinduced respiratory depression develops is not sudden, but slowly progressive, and is generally preceded by nausea, vomiting, and increased sedation (46). Vigilant nursing observation, and documentation of inadequate respiratory effort, slow respiratory rate, or unusual somnolence represent the best form of monitoring (46). Prophylactic naloxone infusions (400 mg/L) at 100 to 125 ml/hr have been advocated to reduce the risk of opioid-induced respiratory depression in elderly or debilitated patients while maintaining effective analgesia (47,48,51). Naloxone infusions effectively reduce the incidence and severity of other adverse effects including pruritus, nausea, and urinary retention (47). Relative contraindications to neuraxial opioid analgesia include spinal fracture, infection at the insertion site, septicemia, coagulopathy, and treatment with low-molecular weight heparinoids. Epidural placement requires assessment of coagulation status, and the absence vertebral fractures, instability, and neural deficit. Catheters should not be inserted in patients with consumptive or drug-induced coagulopathy unless the underlying cause is corrected. There is concern about the safety of epidural catheter placement in patients receiving using anticoagulant-based prophylaxis of deep venous thrombosis. In 1997, the FDA issued an advisory about the potential risk of epidural hematoma in patients receiving regional (spinal or epidural) anesthesia and low-molecular weight heparin. The American Society of Regional Anesthesia and Pain Medicine has published guidelines regarding safe use of anticoagulants in patients undergoing neuraxial anesthesia and analgesia.

NEURAL BLOCKADE FOR ACUTE PAIN MANAGEMENT Peripheral neural blockade minimizes exposure to opioids and are ideally suited for patients sensitive to opioid-induced ileus and bowel obstruction. Other indications include avoidance of opioid-induced ventilatory depression, particularly in patients with underlying pulmonary disease. Infiltration techniques employ injections of local anesthetic at the site of surgery and offer up to 12 hours of postoperative analgesia. These techniques require surgical infiltration of concentrated local anesthetic solutions into skin, subcutaneous tissues, and into the joint capsule. Continuous infiltration techniques typically employ 20-gauge multiport catheters to deliver local anesthetic (usually dilute bupivacaine) under the skin and muscle layers of an incision. In addition, catheters may be placed in the iliac crest (donor bone graft site), pleural space, femoral nerve fascial compartment, and brachial plexus. Additional analgesia may be safely provided with IV-PCA (32,34).

471

POSTTRAUMATIC PAIN Pain management in the trauma patient has been divided into three phases: emergent, acute, and rehabilitative. During the emergent phase, primary attention must be given to stabilizing the patient’s respiratory and cardiovascular status; thereafter the intensity of pain may be assessed and carefully controlled. Primary analgesic therapy includes IV titration of opioids in doses that provide pain relief while not compromising the patient’s hemodynamic status or obscuring the diagnostic process. The acute phase begins with admission to the intensive care facility and ends with transfer to the medical/surgical ward. Acute phase pain management is determined not only by the primary injury, but also by associated medical-procedural interventions. For instance, not only does the noxious periosteal and ligamentous pain of a rib fracture need to be addressed, but also that associated with chest tube insertion or thoracotomy. The same premise is evident in the following upper abdominal injuries; patients suffering blunt abdominal trauma may require an exploratory laparotomy for diagnosis and treatment of acute abdomen. IV-PCA provides useful pain control following major orthopedic trauma and extensive soft-tissue injuries. If there are no contraindications to regional analgesia, IV-PCA may be supplemented with intercostal, brachial plexus, or interpleural blocks. Continuous epidural analgesia should be considered the technique of choice in patients recovering from multiple rib fractures and flail chest. In general, epidural morphine may be administered at lumbar sites while hydromorphone and fentanyl are infused via thoracic catheters. Dilute concentrations of bupivacaine (0.03–0.1%) are added to the epidural infusate in hemodynamically stable patients. Manipulative procedures including external bone fixation, wound debridement, burn eschar excision, and dressing changes are quite painful pain and generally require greater amounts of analgesic than is routinely prescribed. During these procedures, IV-PCA may be supplemented with rapid-acting, shortduration opioids (fentanyl 1–5 mg/kg or equivalent doses of alfentanil), or ketamine (1–3 mg/kg) (32,34). The rehabilitative phase begins when the patient is transferred from the intensive care unit to the medical/surgical ward and ends with full recovery. Patients are expected to move out of bed, ambulate with increasing frequency, and participate in physical therapy. As gastrointestinal (GI) function returns and diet is advanced, IV-PCA and epidural analgesia may be discontinued and oral pain medications substituted. Baseline (resting) pain may be relieved with timed-release preparations of morphine and oxycodone, or methadone. Increases in pain intensity associated with physical therapy may be controlled with morphine or oxycodone given immediately prior to the procedure. During the rehabilitative phase, NSAID may be used to augment timed-release oral

472

Lubin et al.

opioids, while tricyclic antidepressants, clonidine, and sympathetic blockade may be employed to control persistent neuropathic and sympathetically mediated pain (32,34).

NEW IDEAS IN PAIN MANAGEMENT Preemptive Analgesia In the late 1980s, Wall (52) proposed the concept of ‘‘preemptive preoperative analgesia,’’ suggesting that analgesic intervention is most effective when it is made in advance of the pain stimulus rather than in reaction to it. The possibility that preemptive intervention produces analgesic effects that long outlast the pharmacological duration of the agents or techniques employed, suggests important new avenues of treatment of the trauma and postsurgical patient (53). Patients treated with femoral nerve block prior to arthroscopic knee surgery require 50% less opioid analgesic during the first 24 hours of recovery than individuals receiving a similar block at completion of the procedure. Preemptive analgesic benefits have also been observed with epidural opioids. A single dose of epidural fentanyl (4 mcg/kg), administered prior to thoracotomy incision, is more effective in reducing postoperative pain scores and IV-PCA morphine requirements in the day following surgery than a similar epidural fentanyl dose administered during surgery (54). Conversely, Figure 25 illustrates why singletreatment preemptive therapy may be insufficient for the management of surgical pain beyond the immediate postoperative period. Woolf et al. (1,2) has proposed that optimal preemptive analgesia should be continuous and perioperative in nature, and should include preoperative initiation as well as intra- and postoperative maintenance of therapy. To date, only a few studies have compared pre- and postoperative initiation of continuous preemptive treatment (55). In one study (55), there was no clinically significant difference in postoperative pain during continuous epidural bupivacaine–morphine infusion initiated either before or following completion of colon surgery. Its authors speculate that the study protocol may not have prevented central sensitization because complete preemptive afferent blockade was not achieved. While the majority of clinical studies have examined relief of postsurgical pain and acute disability, preemptive analgesia may also provide longer term convalescent-rehabilitative benefits and prevent or minimize the severity of persistent pain syndromes (56). Patients recovering from back-fusion surgery, in which donor bone is taken from the iliac crest, may develop chronic periosteal pain that persists for months to years after the operation. Infusion of bupivacaine via an iliac crest catheter attenuates the intensity of acute postoperative pain and appears to minimize development of chronic sensitivity (56). Neuralgias, phantom limb pain, and deafferentation syndromes

are common after amputation. Recent evidence indicates that preemptive analgesia provided by perioperative epidural conduction blockade can prevent the development of chronic stump and phantom limb pain in patients recovering from below the knee amputation (Fig. 26) (56).

Multimodal Analgesia Complete abolition of postsurgical pain (pain prevention) is difficult to achieve with a single drug or analgesic technique (55). In an effort to minimize single agent dose requirements and the potential toxicity associated with reliance on one agent, ‘‘balanced’’ or multimodal analgesic regimens have been advocated. It is thought that effective pain relief may be achieved by the additive or synergistic activity of two or more analgesics. By reducing the amount of each drug administered, the incidence and severity of potentially serious side effects may be diminished. Multimodal analgesic therapy employs a variety of agents which interfere with noxious transmission and pain perception at different levels within the peripheral and CNS. Recommendations for multimodal anesthetic and perioperative analgesic dosing are presented in Table 2.

PAIN CONTROL AND POSTSURGICAL OUTCOME Do efforts to minimize postoperative pain and associated stress responses result in improved postsurgical outcome? The cost of PCA and epidural infusion devices is considerably higher than traditional forms of analgesia. Studies are under way to determine whether optimal postsurgical analgesia can decrease morbidity and duration of hospital stay, thus offsetting cost. Patients who benefit most from intraspinal opioid analgesia are those recovering from extensive surgical procedures where parenteral opioid dose requirements are high. Therapeutic gains are dramatic in patients with underlying cardiovascular and pulmonary disease, whereas they are less obvious in healthy individuals recovering from minimally invasive procedures. Optimally administered epidural analgesia can suppress release of catecholamines, maintain hemodynamic stability, reduce myocardial oxygen requirements, improve respiratory function, and facilitate physical therapy. Such therapy has also been shown to reduce mortality, hospital stay, and overall cost. These desirable outcomes outweigh the greater invasiveness and potential side effects associated with epidural placement and indwelling catheters (32). Epidural infusions of local anesthetic (0.25–0.5% bupivacaine) can suppress sympathoadrenal and neuroendocrine responses accompanying surgical trauma. Suppression is most effective after lower abdominal and extremity procedures (56). Epidural conduction blockade has been shown to significantly reduce thromboembolic complications in patients

Chapter 37: Postoperative Pain Management

473

Table 2 Dosing Guidelines for Intravenous Patient-Controlled Analgesia Opioid

Concentration

Dose

Bolus

Dose Interval (min)

Morphine Meperidine

1 mg/m 10 mg/ml

3–10 mg 25–50mg

0.5–1.5 mg 5–15 mg

6–8 6–8

Hydromorphone Oxymorphone

0.2 mg/ml 0.1 mg/ml

0.5–1 mg 0.3–1 mg

0.1–0.3 mg 0.1–0.2 mg

6–8 6–8

Fentanyl

20 mcg/ml

30–100 mcg

10–20 mcg

5–6

recovering from hip surgery (57). In a study of critically ill patients recovering from major surgery, Yeager et al. (58) noted that patients treated with epidural morphine benefited from significant reductions in cardiac and respiratory failure and incidence of major infections when compared with individuals administered IV opioids. Seventy-six percent of patients in the general anesthesia parenteral opioid group (19/25) developed some form of organ failure versus 32% (9/28) in the epidural anesthesia–analgesia group. Similar results were reported in patients undergoing major vascular surgery and randomized to receive epidural anesthesia–analgesia or general anesthesia with parenteral opioids for postoperative pain relief. Patients in the general group experienced greater postsurgical morbidity, in particular cardiovascular and infectious complications (32,34,56).

CHRONIC PAIN Despite our best efforts at controlling pain in the perioperative period, the patient’s discomfort may linger long past the resolution of the initial physical trauma. Chronic pain (pain lasting more than three to six months beyond the initial insult) is one of the greatest challenges in medical practice today. It may be the result of inadequate treatment of the acute pain, or the peculiar physical and emotional makeup of the patient. In some instances, the pain is due to traumatic or perioperative nerve injury, with the subsequent development of neuropathic pain, a type of chronic pain that is singularly resistant to common therapies. The following sections include a description of the common perioperative neuropathies, and a brief discussion of the development and treatment of neuropathic pain.

Perioperative Neuropathies Perioperative nerve injury is a considerable source of patient injury and liability in both anesthesia and surgical practice. Neuropathies secondary to surgery are infrequent, however, they can be potentially debilitating in surgical patients. According to the American Society of Anesthesiologists Closed Claims Project, perioperative nerve injury is the second most common class of injury, accounting for 15% of all claims

Rate

Comments

0.5–1.5 mg/hr Major abdominal/orthopedic surgical pain; slow onset Not recommended Useful for visceral pain, limit dose to 600 mg/24hrs 0.1–0.3 mg/hr Rapid onset, minimal side effects 0.1–0.2 mg/hr Rapid onset, best for severe pain, high incidence of nausea/vomiting 10–20 mcg/hr Rapid onset, short duration of effect,

(60). Many of these neuropathies are avoidable and are usually associated with inappropriate patient positioning in the operating room. The most common nerves affected are the peripheral nerves such as the ulnar and sciatic nerves. However, more centrally located nerves, such as the brachial plexus and the lumbosacral nerve roots, can also be affected. Additionally, and of particular interest, there are epidemiologic and anatomic studies to suggest that factors other than intraoperative malpositioning may contribute to the development of neuropathies. Nevertheless, the occurrence of mechanical nerve injury in connection with anesthesia and surgery is probably more common than generally believed, and by no means all cases are reported to the patient injury claims department. Both incorrect positioning of the patient on the operating table and pressure from retractors and other instruments can contribute to the occurrence of such injuries. Both neurological and neuropsychological procedures should be used to localize the injury, particularly for the purpose of insurance assessment. Most important of all, however, is prevention of injury. Risk minimization is dependent on the observation of meticulous routines in surgery units, and clear division of staff responsibilities.

Common Perioperative Neuropathies Upper Extremity Carpal Tunnel Syndrome.

Median neuropathy at the wrist is the most common nerve entrapment syndrome. Hundreds of papers have been written to describe the condition and its diagnosis and treat-

Table 3 Factors Which Increase the Risk of Spinal Opioid-Induced Respiratory Depression Drug-related factors Hydrophilic opioids Excessive dose Large volume of injectate Excessive dose frequency Intrathecal administration Concomitant administration of parenteral opioids Patient-related factors Age greater than 60 years Debilitated individuals Coexisting respiratory disease Raised intrathoracic pressure Shock-wave lithotripsy Trendelenberg position

474

Lubin et al.

Table 4 Dosing Guidelines for Epidural Opioid Analgesia Morphine Lumbar catheters: Administer 3–8 mg bolus 2–4 mg bolus followed by 2–4 mg bolus followed Ketorolac 15–30 mg q 6 hr incisions below T8 in 10 ml preservative infusion (50 mcg/ml at 10–15 infusion (50 mcg/ml at 8-Epidural bupivacaine Thoracic catheters free saline every 8–24 ml/hr, lumbar catheters), 10 ml/hr, lumbar; 4 ml/hr 0.1% or less for upper abdominal hours as clinically 4–8 ml/hr, thoracic catheters) thoracic) PCEA bolus dose and thoracic surgery indicated 1–2 q 15 min, 4 hr limit 30–50 Hydromorphone Lumbar catheters: 0.5–1.5 mg bolus every 0.5–1.5 mg bolus followed by 0.5–1.5 mg bolus followed by Ketorolac 15–30 mg q 6 hr incisions below T8, 6–10 hours infusion (10 mcg/ml at 10–15, infusion (10–20 mcg/ml at 8–10 Epidural bupivacaine thoracic catheters for lumbar catheters), (20 mcg/ml ml/hr, lumbar; 4–6 /hr 0.1–0.05% or less upper abdominal and at 4–8 ml/hr, thoracic thoracic) PCEA bolus dose thoracic surgery catheters) 1–2 q 15 min, 4 hr limit 30–50 ml Meperidine Lumbar catheters: 50–75 mg bolus every 50–75 mg bolus followed by 50–75 mg bolus followed by Ketorolac 15–30 mg q 6 hr incisions below T8, 4–6 hours infusion (100 mcg/ml at 10–15, infusion (100 mcg/ml at 8–10 Epidural bupivacaine 0.1% thoracic catheters for lumbar catheters), (4–8 ml/hr ml/hr, lumbar; 4–6 /hr upper abdominal and thoracic catheters. thoracic) PCEA bolus dose thoracic surgery 1–2 q 6–10 min, 4 hr limit 30–50 Fentanyl Lumbar catheters: 50–100 mcg bolus 50–100 mcg bolus followed 50–100 mcg bolus followed by Ketorolac 15–30 mg q 6 hr incisions below T12, every 2–4 hr infusion (5 mcg/ml at 10–15, infusion (5mcg/ml at 8–10 Epidural bupivacaine 0.05thoracic catheters for (not recommended) lumbar catheters), (4–8 ml/hr ml/hr, lumbar; 4–6 /hr 0.1% or less almost everything else thoracic catheters. thoracic) PCEA bolus dose 30–50 Sufentanil Lumbar catheters: 20–30 mcg bolus 20–30 mcg bolus followed 20–40 mcg bolus followed by Ketorolac 15–30 mg q 6 hr incisions below T12, every 2–4 hr infusion (1–2 mcg/ml at 10–15, infusion (1–2 mcg/ml at 8–10 Epidural bupivacaine 0.05thoracic catheters for (not recommended) lumbar catheters), (4–8 ml/hr ml/hr, lumbar; 4–6 /hr 0.1% or less almost everything else thoracic catheters. thoracic) PCEA bolus dose 30–50
Dependent on age, physical status, height, extent of surgical dissection, and so on. Note: The bolus amount for each agent may be used as a starting hourly basal infusion rate.

ment (61). While it is a common outpatient nerve disorder, its prevalence as a perioperative injury is low in comparison. Ulnar Neuropathy. The ulnar nerve is the single most common site of perioperative peripheral nerve injury, constituting 28% of all perioperative nerve claims. Approximately 50% of all patients with ulnar neuropathy have deficits that persist for more than a year. External compression of the ulnar nerve in the cubital tunnel is usually regarded as the most important cause of anesthesia-related injury; however, detailed study of individual cases has not provided strong correlation. In a large-scale multivariate analysis, Warner et al. (69) isolated several independent risk factors for ulnar neuropathy, including male gender, high or low body mass, and prolonged hospital stay. How these factors speTable 5 Advantages of Epidural Patient Controlled Analgesia Vs. intravenous patient-controlled analgesia Superior pain relief Reduced drug requirement Reduction in drug-related side effects Shorter hospitalization Vs. continuous epidural opioid infusion Patient self-adjustment Reduced hourly infusion requirement Accommodation for changes in pain intensity (i.e., ambulation) Reduced anxiety, increased patient control

cifically relate to intraoperative mechanisms of injury still remain unclear. Anatomic studies reveal differences in the elbow anatomy of men and women (degree of adipose tissue and thickness of the retinaculum in the cubital tunnel). However, the current understanding of the role that these differences play in the genesis of nerve injury has not evolved beyond speculation. An identifiable mechanism of injury has been reported in only 9% of cases in the Closed Claims Project database. Although the true mechanisms of anesthesiarelated ulnar neuropathy remain undefined, it has been assumed that external pressure exerted against the nerve is the likely etiologic factor. Several reports suggest that the ulnar nerve is more susceptible to ischemia than the radial or median nerve. Many authors acknowledge that ulnar neuropathy remains a clinical entity for which we still have minimal understanding of cause-and-effect relationships (62), nor whether it is always a preventable complication (64). Accumulating evidence suggests ulnar nerve injury can occur at any time during the course of hospitalization. Prielipp et al. made an important breakthrough in the current study of perioperative or old or nerve injury using a pressure-sensitive mat to determine how changes in arm position affect the external pressure transmitted to the ulnar groove. Supination of the forearm produces the least amount of pressure at the ulnar groove, pronation produces the most, and

Chapter 37: Postoperative Pain Management

475

a neutral forearm precision results in an intermediate value. The findings of this study are consistent with inferences from basic anatomic considerations. Their study also used somatosensory-evoked potentials to explore the relationship between sensory changes and the electophysiologic changes that occur when external pressure is applied to the ulnar nerve. Interestingly, half of other volunteers failed to perceive any sensory changes despite marked abnormal changes in somatosensory-evoked potential signal strength. Therefore, this suggests that the patient’s report is of limited value for the early detection of compressive nerve injury. Although a humbling reminder that we have limited understanding of the relationship between perioperative care and the generation of peripheral nerve injury, the study has practical application for clinical decision-making.

be difficult to distinguish from a peripheral ulnar neuropathy. In the lateral decubitus position, compression plays a predominant role in injury when the plexus is compressed against the thorax by the humeral head. Anatomical variations of the thoracic outlet and the presence of an extra rib on the seventh cervical vertebrae can also be predisposing factors. Brachial plexus neuropathy associated with procedures that are performed on patients in a steep head-down position and who are restrained with shoulder braces can be caused by direct compression or stretch of the brachial plexus. Theoretically, stretch of the plexus may occur when the head is turned contralateral and when the ipsilateral arm is abducted with elbow flexion. Fortunately, the frequency of perioperative brachial plexus neuropathy seems to be low in patients who are positioned prone.

Stretch of the brachial plexus is particularly induced by arm abduction, external rotation, and posterior shoulder displacement (60). Conditions that may predispose to this include extension and lateral flexion of the head to one side with the patient in the supine position, in addition to abduction and external rotation of the arm by allowing it to drop away from the side of the body. Considerable stretch of the brachial plexus roots can also occur from extreme in abduction of the arm with the hand resting above the head. Patients who are placed in the prone position may also be vulnerable to stretch of the brachial plexus (67). Stretch or compression of the brachial plexus has also been described with upward movement of the clavicle secondary to sternal retraction and median sternotomy (63,64). Brachial plexus nerve injury during the sternal retraction is most common during the internal mammary artery dissection (65,66). The mechanism is assumed to be an asymmetric retraction of the rib cage displacing the upper rib cage that may stretch or compress the C8 through T1 nerve trunks. The trunks form the major contribution to the ulnar nerve and therefore brachial plexus neuropathy may

Trunk Postherniorrhaphy Pain/Ilioinguinal Neuropathy.

Brachial Plexus Neuropathy.

Chronic pain after inguinal herniorrhaphy is a well-known complication that can be debilitating. The reported incidence of chronic groin or inguinal pain after hernia repair varies from 0% to 62% after one year depending on the studies reviewed (70–72). Between 6% and 16% of patients across the studies report moderate to severe pain one year after surgery with a greater incidence in those with recurrent repair. Eleven to sixty-two percent report some degree of pain in the groin area with 10% reporting persistent pain that interferes with activity two years postoperatively. Several distinct types of chronic pain have been reported. The most common pain appears to be somatic and localized to the common ligamentous insertion to the pubic tubercle. The second is neuropathic and referable to possible injury to the ilioinguinal or genitofemoral nerves either at the site of surgery or due to encroachment of scar tissue. The third pain type is visceral and related to ejaculatory pain. Persistent numbness can be common in the distribution of the branches of the ilioinguinal or iliohypogastric nerves in up to 24% of patients.

Table 6 Multimodal Analgesia: Clinical Applications In the setting of General Anesthesia: Premedication-analgesic base consisting of moderate to long duration opioid such as morphine (0.05–0.1 mg/kg) or hydromorphone 2. An additional 0.1– 0.2 mg/kg morphine dose administered immediately prior to anesthetic induction (the administration of preoperative and induction doses of benzodiazepines, and induction agents should be reduced), 3. Preincisional administration of ketamine (0.2 mg/kg and NSAID (rofecoxib 50 mg, Celecoxib 200 mg or ketorolac 15–30 mg) unless contraindicated by patient status or surgical procedure) 4. preincisional neural blockade; either wound and fascial infiltration, peripheral nerve block or plexus block, 5. IV PCA started in the PACU as soon as the patient is considered alert enough to appropriately utilize such therapy, 6. Maintenance oral rofecoxib 50 mg/day X5days, ketorolac 7.5–15 mg slow IV q 6 hr for 24–48 hrs. In the setting of epidural/spinal anesthesia or epidural-light general anesthetics : 1. To reduce local anesthetic-opioid doses requirements (increase neuroaxial specificity) epidural catheters should be inserted at interspaces adjacent to the site of surgery. 2. Epidural induction with 2% lidocaine or 0.5–0.75% bupivacaine in doses sufficient to block afferent input from dermatomal sites of surgical incision and deeper dissection. Utilize lower concentrations and total dose of local anesthetic in elderly patients or individuals at risk for intraoperative hypovolemia. 2. Preincisional administration of epidural opioid (morphine, hydromorphone, fentanyl) prior to incision, 3. Preincisional administration of NSAID (ketorolac 15–30 mg) and ketamine (0.2 mg/kg for epidurallight general patients) if not clinically contraindicated, 4. Opioid-bupivacaine infusion initiated during surgical closure or upon arrival in the PACU, 5. Addition of Patient controlled epidural dosing when deemed appropriate (refer to Table 26-5), 6. Maintenance of ketorolac 7.5–15 mg IV q6hr for 24–48 hrs.

476

Lubin et al.

Treatments for persistent pain range from local anesthetic injection to remedial surgery and appear to have varying results. Heise and Starling (73) report that remedial surgery with coincident neurectomy affords more favorable results in 60% of patients than mesh removal alone. En bloc resection of the scar tissue has also been suggested by some authors as a potential treatment. The true incidence of chronic pain after hernia repair remains to be documented. Patients treated for recurrent hernia and those with a high initial pain score and persistent pain four weeks after repair are more likely to report chronic pain one year postoperatively. Mesh repair has become very popular. Its tension-free properties should lead to less acute postoperative pain, however, in randomized studies it has not proven superior to other open repair techniques with respect to acute pain or the use of analgesics (74,75). Nerve injury at the time of operation may be an important mechanism and it appears that remedial surgery with neurectomy may afford acceptable results. Lower Extremity

Vascular, general surgical, orthopedic, or gynecologic procedures performed either inside the abdominal cavity or on the lower extremities may result in nerve injury. Motor and sensory neuropathy in the lower extremity is a well-recognized potential complication of procedures that can either be related to the surgery itself or to surgical positioning of the patient. Injury resulting in lower extremity neuropathy can virtually occur after any invasive procedure outside of the spinal column from the L1 vertebral level down. The true incidence of neurologic injury is likely underreported and one can imagine that anywhere a surgical scalpel crosses a nerve, where retractors are held, and any position that results in direct pressure or vascular compromise to a peripheral nerve can lead to neurologic consequences. Common Peroneal Nerve Injury. Sciatica. The sciatic nerve is especially at risk if the patient is thin, the operating room table hard, the operation long, and when the opposite buttock is elevated as in the hip-pinning position. Smoking during the preoperative period has also been associated with the development of neuropathies. External rotation of the flexed thigh in the lithotomy position may damage the nerve by stretch (76). Several studies have suggested that many factors other than inappropriate intraoperative care and positioning may contribute to the risk of lower extremity nerve injuries (77–79). In a retrospective review of the frequency and type of motor neuropathies that occurred in 198,461 consecutive patients who underwent procedures in the lithotomy position, it was found that the nerves most often involved are the common peroneal nerve (81%), sciatic nerve (15%), and femoral nerve (4%) (80).

In view of the fact that prolonged duration in the lithotomy position as well as patient-specific factors is involved in the development of neuropathy, use findings suggest that not all neuropathies should be assumed to be the result of inappropriate positioning. Femoral Neuropathy.

Unlike many other neuropathies in which the anesthesia provider or inappropriate positioning is implicated, those injuries that involve the femoral nerve and its branches are thought to result from inappropriate placement of abdominal wall retractors and direct compression of the nerve during surgery (81). Here, it is assumed that the retractor used for an abdominal surgical approach to the pelvis places continuous pressure on the iliopsoas muscle, thereby stretching the nerve. An alternate explanation is that retractor compression causes nerve ischemia by occluding the external iliac artery or its branches that penetrate the nerve as it passes through the muscle. It has also been speculated by Rosenblum et al. (82) that self-retaining retractors placed in the abdomen are more likely than hand-held retractors to result in femoral neuropathy, as these devices exert a continuous pressure against the tissues and the surrounding structures. In addition to retractors, certain patient factors may be associated with femoral neuropathy in the perioperative period. An extremely thin body habitus and smoking are associated with lower extremity neuropathy after surgery. Additionally, type II diabetic patients are susceptible to compression neuropathy, as they are insulin resistant and the nerve tissue is dependent on glycolysis.

Identification of Neuropathic Pain The symptomatic complaint of nerve-related pain in the face of a new motor deficit makes for easy diagnosis. Identification of neuropathic pain may not be straightforward. First, neuropathic pain may take on a variety of characteristics, many of which may be difficult for the patient to describe. This further confuses the clinician and delays the correct diagnosis. Second, when faced with the complaint of neuralgia without accompanying motor deficit, the physician must entertain a higher index of clinical suspicion. Reliance on knowledge of the anatomy of nerve contributions from the various dermatomes and myotomes is the key when taking a detailed history of the patient’s symptomatic complaint in order to correctly identify where the pathology lies. Frequently, this will require serial examinations of the patient. The subjective complaints on these occasions will include new descriptive characteristics if central sensitization or wind-up occurs. Patients with neuropathic pain may exhibit persistent or paroxysmal pain that is independent of an inciting stimulus. This stimulus-independent pain is usually described as shooting, lancinating, and

Chapter 37: Postoperative Pain Management

burning, and may depend on activity of the sympathetic nervous system. The pain complaint may also take on the characteristic of a deep ache, constriction, or band-like sensation, or the feeling of tissue expansion or swelling. Stimulus-evoked pain is a common component of peripheral nerve injury and has two key features which we have previously encountered: hyperalgesia and allodynia. Hyperalgesia is an increased response to a suprathreshold noxious stimulus and is the result of abnormal processing of nociceptor input. Allodynia is the sensation of pain elicited by a typically non-noxious stimulus. The treatment of neuropathic pain can be difficult as multiple etiologic mechanisms may be involved. These mechanisms can include an alteration in the number and type of sodium channel receptors along the axon, expression of alpha-adrenergic receptors that renders the nerve sensitive to circulating catecholamines and norepinephrine released from postganglionic sympathetic nerve terminals, disruption of the fiber input into the dorsal horn of the spinal cord responsible for inhibitory pathways, and a myriad of additional mechanisms responsible for chemical and receptor changes in the spinal cord. Unfortunately, there is no treatment known which can neither prevent the development of neuropathic pain nor reliably control established neuropathic pain. Although laboratory investigation and quantitative sensory testing in patients have advanced our knowledge of the mechanisms that produce neuropathic pain, we are lacking in the sensitive and specific diagnostic tools that are needed to reveal the particular pathologic process that is responsible for pain in a particular patient. The responsibility rests with the physician to use the techniques of history, physical examination, and extant diagnostic tools in an appropriate way to identify and treat neuropathic pain.

Diagnosis of Neuropathic Pain The diagnosis of neuropathic pain lies completely in the patient’s subjective complaint of pain, the physical examination, and the clinical suspicion of the physician. Electromyographic (EMG) testing is unreliable for the diagnosis of neuropathic pain, as it tests the integrity only of A-beta fibers, and does not test the function of A-delta or C-fibers, the elements that are responsible for pain transmission. However, as the A-beta fibers are involved in the activation of segmental inhibitory pathways, an abnormal EMG may signify loss of inhibitory control by this pain fiber population. It bears emphasizing that a ‘‘normal EMG’’ does not exclude the diagnosis of neuropathic pain. Other clues, such as the minimal response to opioid medications or a clinical response to empiric therapy with medications directed at the treatment of nerve pain, make the diagnosis in retrospect. The physical examination of the patient complaining of pain, as is true for any neurologic investigation, should be conducted in a thorough,

477

logical manner. Therefore, it should begin with light touch or stroke of the physician’s palpating hand over the affected body region, and progress in a layer-bylayer palpatory fashion; i.e., by subsequently applying more pressure so as to palpate the deeper structures. If pain is elicited by light stroke or mild deformation of the overlying skin, the physical examination should be conducted to identify the total surface area involved that may represent a dermatomal or an irregular shaped peripheral nerve distribution. A stocking-glove distribution may signify a more central neurologic process. An underlying mechanism of injury such as local soft-tissue trauma or more central neurologic process should be sought after. In many cases, the exact cause will go unidentified. Physicians skilled in anatomy and regional anesthesia have the option of ‘‘compartmentalizing’’ a particular nerve or group of nerves with the administration of local anesthetic to aid in the diagnosis. This can be performed as a simple office procedure, or may require the use of fluoroscopy in an operating suite. The underlying cause should be identified and treated accordingly; however, medication management directed toward palliation of symptoms should not be withheld in the meantime.

Treatments for Neuropathic Pain As a general rule, if neuropathic pain is suspected, medications known to be effective in the treatment of nerve pain should be considered. Strategically selecting any particular agent that would benefit the patient can be difficult due to multiple factors (2,3). However, each situation is different and necessitates a careful assessment and identification of the particular neuropathy involved. Neuropathy may be sensory or motor and, in general, sensory lesions are more frequently transient when compared with motor lesions. If the symptoms consist of numbness or tingling only (that is, without pain), conservative management would be appropriate, and treatment would not require the use of any medications. It is important both to inform the patient that such neuropathy often resolves postoperatively (84) and to instruct the patient to avoid positions that might compress or stretch the involved nerve. A full neurological examination should be performed and documented in the medical record. If the neuropathy has a motor component, a neurologist should be consulted immediately. EMG studies may not be revealing within the first 14 to 21 days after the initial injury. Medication Trials Tricyclic Antidepressants.

The tricyclic antidepressants were so named because they are all three-ringed structures with more or less substitution at the middle ring. They are further classified as secondary or tertiary amines based on the terminal substitution of the side chain. The antidepressant medications are generally well absorbed after oral administration. They undergo

478

Lubin et al.

extensive first-pass hepatic metabolism and are highly bound to serum proteins. Their elimination half-lives are long and they frequently have active metabolites. These drugs are generally oxidized by hepatic system and conjugated for excretion. Elimination occurs via urine and feces. The antidepressant medications exert their actions by altering monoamine neurotransmitter activity at the level of the synapse; specifically, they block presynaptic uptake of 5-HT and norepinephrine. It appears that neurotransmitter modulation is the probable mechanism. Review of multiple studies indicates that the antidepressants alone or in combination with other medications can be beneficial to the patient suffering from neuropathic conditions. Therapy with antidepressant medications should begin with the lowest dose, to be taken one to two hours before sleep. The medication can be titrated to clinical effect, and any sign of toxicity should prompt a serum level determination. Antidepressants can cause a number of side effects such as sedation, dry mouth, and orthostatic hypotension. Abrupt discontinuation of antidepressants can lead to sleep disruption, characterized by vivid and colorful dreams that are usually not dysphoric. The presenting signs of antidepressant overdose are related to the extremes of their sedative effects and their influence on the cardiac conduction system. This is frequently accompanied by significant hypotension that may be unresponsive to IV fluid management. These medications are used as one part of a comprehensive approach to chronic pain. They have demonstrated analgesic activity, are nonaddicting, and have a myriad of beneficial pharmacologic actions.

The anticonvulsants are a heterogeneous group of drugs used in pain management. Numerous agents, including valproic acid, carbamazepine, clonazepam, gabapentin, topiramate, and tiagabine, have been used in the treatment of neuropathic pain. The mechanisms of action of the drugs for this purpose are not well understood and are probably unique to each drug. They are generally believed to have effects on sodium, calcium, and potassium flux, and some have effects on gamma-amino-butyric acid (GABA) activity. Anticonvulsant medications, and in particular carbamazepine, are considered to be the treatment of choice in trigeminal neuralgia; however, they are used in a variety of neuropathic conditions, with varying levels of success. Newer agents such as topiramate, tiagabine, and zonisamide are receiving clinical attention in antineuralgic therapy because as a group they are more easily titrated (frequent blood levels are not necessary) and have more favorable side effect profiles. Unfortunately, apart from trigeminal neuralgia and its responsiveness to carbamazepine, pharmacotherapy for neuropathic pain has been less than ideal, and it effects are

Anticonvulsants.

unpredictable due to the multiple underlying mechanisms that may be involved. Although effective for most pain, oral opioids may be only partially effective in the treatment of neuropathic pain. As a result, many patients self-medicate in a desperate attempt to obtain pain relief, with limited success and much grief. Patients often finish their opioid medication prescriptions early, running the risk of becoming unwelcome and stigmatized by the medical profession. Unremitting pain of this sort has emotional, psychological, and behavioral consequences, and the administration of opioids in these circumstances may be useless or worse. Because of this, and because of the practitioner’s confusion and incomplete understanding of neurologic anatomy and the process of central sensitization, the patient is frequently regarded as malingering or somatisizing. This is not to say that there is no place for opioids in the treatment of chronic, nonmalignant pain. Interestingly, opioids given via the epidural or subarachnoid route behave very differently with respect to onset, duration, and side effects than the same drugs given systemically. Therefore, pain unresponsive to systemic opioids may respond to those same opioids given centrally. In addition, by administering the agent centrally, the dose of opioid can be reduced a 100- or a 1000-fold, reducing many unwanted effects that were unavoidable in doses given systemically.

Opioids.

Advanced Pain Therapies Spinal cord stimulation for the clinical control of pain was introduced in 1967 by Shealy et al. (89), based on the expectation of effect predicted by the gate control theory of pain, published in 1965 by Melzack and Wall (90). The basic premise of Melzack and Wall’s theory is that the stimulation of sensory input carried by the large myelinated A-beta fibers, by their earlier arrival at the spinal ‘‘gate,’’ suppress or modify information carried by the carrying nerve fibers carrying nociceptive information, such as A-delta and C-fibers. Shealy et al. speculated that if the A-beta fibers of the dorsal columns were electrically stimulated, reception of painful information would be inhibited. They presented the first clinical evidence for electrical stimulation in the treatment of pain. Many theories subsequently have been presented in the literature that postulate the mechanisms underlying the efficacy of spinal cord stimulation. Basically, spinal cord stimulation appears to influence the release of neurotransmitters and neuromodulators, and to play an inhibitory role at sympathetic efferent neurons, which have been shown to promote the maintenance of neuropathic pain. These effects of spinal cord stimulation are seen at both spinal and supraspinal sites. While the exact mechanism of spinal cord stimulation’s effects remains obscure, it is clear that the treatment

Chapter 37: Postoperative Pain Management

479

Table 7 Pain Amenable to Spinal Cord Stimulation

Summary

Sympathetically maintained pain Causalgia Reflex sympathetic dystrophy Arachnoiditis Perineural fibrosis/failed back surgery Radicular pain Peripheral vascular insufficiency Phantom limb pain Deafferentation pain Postherpetic neuralgia Peripheral neuropathies Spinal cord injury Angina

Poorly controlled pain following major surgery incites several pathophysiological responses that increase postoperative morbidity, increase the incidence of prolonged rehabilitative pain, and may lead to persistent pain. The development of chronic pain is a tragedy that must be prevented. By assessing the severity and character of the pain stimulus, optimum pain control may be provided at each phase of the recovery process. Analgesic regimens in the acute phase, including opioid infusions, IV- and epidural-PCA, and continuous regional blockade, not only provide effective pain relief and high patient satisfaction, but also lead to improved function, decreased recovery time, and shortened hospitalization. Diagnosis and effective treatment of chronic pain are still in their infancy, but the development of new, creative techniques and the simple application of good medicine are making headway in the care of these disabled patients.

is effective for a variety of painful neuropathic conditions. Spinal cord stimulation, a form of neuroaugmentation, appears to be in favor by pain specialists nowadays, as opposed to older, destructive procedures such as neurolysis. Table 7 illustrates the types of pain amenable to spinal cord stimulation. Intrathecal Pump

Radiofrequency ablation and cryodenervation are methods of applying thermal destruction to a nerve for the treatment of a variety of painful conditions involving peripheral nerves and sympathetic fibers. Both these techniques have had varying degrees of success in the treatment of intercostal neuralgia, painful neuroma, biomechanical spinal pain, peripheral neuropathies, and a variety of cranial and facial pain syndromes. It is essential that the provider of these techniques be fully aware of the regional anatomy required for any specific procedure. These techniques are generally applied to sensory nerve fibers, and not mixed fibers carrying motor and sensory information, as lesioning of the latter can produce profound motor weakness of an extremity or important muscular structure.

The provision of neurolytic blockade in patients with chronic nonmalignant pain is controversial. The reappearance of causalgic pain is a feature common to most ablative procedures (85,86). This can be minimized by limiting the selection of patients to those with a short-life expectancy. Here, the patient is unlikely to outlive the duration of pain relief. A potential for damage to nontargeted tissue is of concern with any destructive procedure, though it is less likely to occur when peripheral neurolysis, as opposed to central or deep sympathetic neurolytic blockade, is undertaken. This is particularly true when localization is facilitated by diagnostic electrical stimulation, radiographic guidance, or administration of test doses of local anesthetic (87,88). Peripheral neurolysis has specific but important indications in the management of intractable cancer pain syndromes. It is generally thought inappropriate and inadequate for benign intractable nonmalignant pain.

Chemical Neurolysis.

REFERENCES 1. Woolf CJ, Max MB. Mechanism-based pain diagnosis: issues for analgesic drug development. Anesthesiology 2001; 95:241–249. 2. Woolf CJ, Thompson SW. The induction and maintenance of central sensitization is depedent on N-methyl D-aspartic acid receptor activation; implications for the treatment of post-injury pain hypersensitivity states. Pain 1991; 44:293–299. 3. Carr DB. JAMA 1998; 279:1114–1115. 4. Apfelbaum JL, Chen C, Mehta S, et al. Postoperative pain experiences: results from a national survey suggest postoperative pain continues to be undermanaged. Anesth Analg 2003; 97:534–540. 5. Dolin SJ, Cashman JN, Bland JM. Effectiveness of acute postoperative pain management: I: evidence from published data. Br J Anaesth 2002; 89:409–423. 6. Melzack R, Abbott FV, Zackon W, et al. Pain on a surgical ward: a survey of the duration and intensity of pain and the effectiveness of medication. Pain 1987; 29: W67–W72. 7. Belville JW, Forrest WH, Miller E, et al. Influence of age on pain relief from analgesics. J Am Med Assoc 1971; 217:1835–1841. 8. Gagliese L, Jackson M, Ritvo P, et al. Age is not an impediment to effective use of patient controlled analgesia by surgical patients. Anesthesiology 2000; 93:601–610. 9. Macintyre PE, Jarvis DA. Age is the best predictor of postoperative morphine requirements. Pain 64:357–364. 10. Merskey H, Bogduk N, eds. Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms. 2nd ed. Seattle: IASP Press, 1994. 11. Isacson D, Bingefors K. Epidemiology of analgesic use: a gender perspective. Eur J Anesth Suppl 2002; 26:5–15. 12. Merskey H, Bogduk N, eds. Classification of Chronic Pain. 2nd ed. Seattle: IASP Press, 1994. 13. Sullivan MJ, Thorn B, Haythornthwaite JA, et al. Theoretical perspectives on the relation between catastrophizing and pain. Clin J Pain 2001; 17:52–64.

480

Lubin et al.

14. Butler RW, Damarin FL, Beaulieu C, et al. Assessing cognitive strategies for acute postsurgical pain. Psychol Assess 1989; 1:41–45. 15. Portenoy RK. Appropriate use of opioids for persistent non-cancer pain. Lancet 2004; 364(9436):739–740. 16. Portenoy RK. Current pharmacotherapy of chronic pain. J Pain Symptom Manage 2000; 19(suppl 1): S16–S20. 17. Portenoy RK. The effect of drug regulation on the management of cancer pain. NYS J Med 1991; 91(suppl 11): 13S–18S. 18. Portenoy RK. Management of common opioid side effects during long term therapy of cancer pain. Ann Acad Med (Singapore) 1994; 23(2):160–170. 19. McMahon SB, Koltzenburg M. Opioids: clinical use. Wall and Melzack’s Textbook of Pain. 5th ed. London: Elsevier, 2006:443–457. 20. Lubenow TR, McCarthy RJ, Ivankovich AD. Management of acute postoperative pain. In: Barash PG, Cullen BF, Stoelting RK, eds. Clinical Anesthesia. Philadelphia: JB Lippincott, 1992:1547–1577. 21. Price DD, Dubner R. Neurons that subserve the sensory discriminative aspects of pain. Pain 1977; 3:307. 22. Willis WD. The pain system: the neural basis of nociceptive transmission in the mammalian nervous system. In: Gildenberg PL, ed. Pain and Headache. Vol. 8. Basel: S. Karger, 1985:1547–1577. 23. Price DD, Hayes RL, Ruda M, Dubner R. Spatial and temporal transformations of input to spinothalamic tract neurons and their relation to somatic sensations. J Neurophysiol 1980; 41:933. 24. Salt TE, Hill RG. Transmitter candidates of somatosensory primary afferent fibres. Neuroscience 1983; 10:1083–1103. 25. Woolf CJ, Chong MS. Preemptive analgesia-treating postoperative pain by preventing the establishment of central sensitization. Anesth Analg 1993; 77:362–379. 26. Mannion RJ, Woolf C. Pain mechanisms and management: a central perspective. Clin J Pain 2000; 16(suppl 3): S144–S156. 27. Sang CN. Clinically available glutamate receptor antagonists in neuropathic pain states. In: Sirinathsinghji DJS, Hill RG, eds. NMDA Antagonists as Potential Analgesic Drugs. Basel: Birkhauser, 2002:165–180. 28. McMahon SB, Bennett DLH, Bevan S. Inflammatory mediators and modulators of pain. In: Wall and Melzack’s Textbook of Pain. 5th ed. London: Elsevier, 2006:49–71. 29. Berde CB. Pediatric postoperative pain management. Pediatr Clin North Am 1989; 36:921–940. 30. Brown DL, Carpenter RL. Perioperative analgesia: a review of risks and benefits. J Cardiothorac Anesth 1990; 4:368–383. 31. Ali J, Weisel RD, Layug AB, et al. Consequences of postoperative alterations in respiratory mechanics. Am J Surg 1974; 128:376–382. 32. Sinatra RS: Unpublished Observations, 1994–1996, Yale University Acute Pain Service. 33. Grass JA, Sakima NT, Valley M, et al. Assessment of ketorolac as an adjuvant to fentanyl patient-controlled epidural analgesia after radical retropubic prostatectomy. Anesthesiology 1993; 78:642–648. 34. Lubin E. Unpublished Observations, 2000–2001, Yale University Acute and Chronic Pain Service.

35. Abboud TK, Dror A, Mosaad P, et al. Minidose intrathecal morphine for the relief of post-cesarean section pain: safety and efficacy, and ventilatory responses to carbon dioxide. Anesth Analg 1988; 67:137–141. 36. Anand KJS, Hickey PR. Halothane-morphine compared with high dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery. N Engl J Med 1992; 326:1–9. 37. Beattie WS, Buckley DN, Forrest JB. Epidural morphine reduces the risk of postoperative myocardial ischemia in patients with cardiac risk factors. Can J Anaesth 1993; 40:523–541. 38. Breslow MJ, Jordan DA, Christopherson R, et al. Epidural morphine decreases postoperative hypertension by attenuating sympathetic nervous system hyperactivity. JAMA 1989; 261:3577–3581. 39. Brodsky JB, Chaplan SR, Brose WG, Mark JBD. Continuous epidural hydromorphone for postthoracotomy pain relief. Ann Thorac Surg 1990; 50:888–893. 40. Bromage PR, Camporesi EM, Durant PAC, et al. Nonrespiratory side effects of epidural morphine. Anesth Analg 1982; 61:490–495. 41. Christopherson R, Beattie C, Meinert CL, et al. Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery. Anesthesiology 1993; 79:1–12. 42. de Leon-Casasola OA, Parker B, Lema MJ, et al. Postoperative epidural-bupivacaine-morphine therapy: experience with 4,227 surgical cancer patients. Anesthesiology 1994; 81:368–375. 43. Chaplan SR, Duncan SR, Brodsky JB, Brose WG. Morphine and hydromorphone epidural analgesia: a prospective, randomized comparison. Anesthesiology 1992; 77:1090–1094. 44. Ferrante FM, Covino BG. Patient-controlled analgesia: a historical perspective. In: Ferrante FM, Ostheimer GW, Covino BG, eds. Patient-Controlled Analgesia. Boston: Blackwell Scientific, 1990. 45. El-Baz NMI, Faber LP, Jensik RJ. Continuous epidural infusion of morphine for treatment of pain after thoracic surgery: a new technique. Anesth Analg 1984; 63:757–764. 46. Geller E, Chrubasik J, Graf R, et al. A randomized double-blind comparison of epidural sufentanil versus intravenous sufentanil or epidural fentanyl analgesia after major abdominal surgery. Anesth Analg 1993; 76:1243–1250. 47. Glass PSA, Estok P, Ginsberg B, et al. Use of patientcontrolled analgesia to compare the efficacy of epidural to intravenous fentanyl administration. Anesth Analg 1992; 74:345–351. 48. Grass JA, Zuckerman RL, Sakima NT, Harris AP. Patient controlled analgesia after cesarean deliveryepidural sufentanil versus intravenous morphine. Reg Anesth 1994; 19:90–97. 49. Gwirtz KH, Young JV, Walker SG, et al. Intrathecal opioid analgesia for acute postoperative pain: experience with 4,134 surgical patients. Anesthesiology 1995; 83:A780. 50. Parker RK, White PF. Epidural patient-controlled analgesia: an alternative to intravenous patientcontrolled analgesia for pain relief after cesarean delivery. Anesth Analg 1992; 75:245–251. 51. Sinatra RS, Sevarino FB, Paige D, et al. Patientcontrolled analgesia with sufentanil: a comparison of

Chapter 37: Postoperative Pain Management

52. 53. 54.

55. 56.

57. 58. 59. 60. 61. 62. 63. 64. 65.

66.

67. 68. 69.

70.

epidural versus intravenous administration. J Clin Anesth 1996; 8:123–129. Woolf CJ, Chong MS. Preemptive analgesia-treating postoperative pain by preventing the establishment of central sensitization. Anesth Analg 1993; 77:362–379. Cousins MJ. Acute pain and the injury response: immediate and prolonged effects. Reg Anesth 1989; 16:162–176. Kehlet H. Modification of responses to surgery by neural blockade: clinical implications. In: Cousins MJ, Bridenbaugh PO, eds. Neural Blockade in Clinical Anesthesia and Management of Pain. Philadelphia: Lippincott, 1987. Liu S, Carpenter RL, Mackey DC, et al. Effects of perioperative analgesic technique on rate of recovery after colon surgery. Anesthesiology 1995; 83:757–765. Stevens DS, Dunn WT. Acute pain management for the trauma patient. In: Sinatra RS, Hord AH, Ginsberg B, Preble LM, eds. Acute Pain Mechanisms and Management. St Louis: Mosby Yearbook, 1992. Modig J, Malmberg P, Karlstrom G. Effect of epidural versus general anesthesia on calf blood flow. Acta Anesthesiol Scand 1980; 24:305–311. Scott DA, Beilby DS, McClymont C. Postoperative analgesia using epidural infusions of fentanyl with bupivacaine. Anesthesiology 1995; 83:727–737. Yeager MP, Glass DG, Neff RK. Epidural anesthesia and analgesia in high-risk surgical patients. Anesthesiology 1987; 66:729–736. Dhuner KG. Nerve injury following operations: survey of cases and occurring during a 6-year period. Anesthesiology 1950; 11:202–207. Kroll DA, Caplan RA, Posner K, Ward RJ, Cheney FW. Nerve injury associated with anesthesia. Anesthesiology 1990; 73:202–207. Cousins M, Mather L. Intrathecal and epidural administration of opioids. Anesthesiology 1984; 1:276–310. Seyfer AE, Grammar NY, Bogumill GP, Provost JM, Chaudry U. Upper extremity neuropathies after cardiac surgery. J Hand Surg 1985; 10:16–19. Casscells CD, Lindsey RW, Ebersole J, Li B. Ulnar neuropathy after median sternotomy. Clin Orth Rel Res 1993; 291:259–265. Vahl CF, Carl I, Muller-Vahl H, Struck E. Brachial plexus injury after cardiac surgery: the role of internal mammary artery preparation: a prospective study of 1,000 consecutive patients. J Thoracic Cardiovasc Surg 1991; 102:724–729. Roy RC, Stafford MA, Charlton JE. Nerve injury and musculoskeletal complaints after cardiac surgery: influence the internal mammary artery dissection and left arm position. Anesth Analg 1998; 67:277–279. Martin JT. The ventral decubitus (prone) positions. Positioning in Anesthesia and Surgery. 3rd ed. Philadelphia: Saunders, 1997:155–195. Kroll DA, Caplan RA, Posner K, Ward RJ, Cheney FW. Nerve injury associated with anesthesia. Anesthesiology 1990; 73:202–207. Warner MA, Warner ME, Martin JT. Ulnar neuropathy: incidence, outcome, and risk factors in sedated or anesthetized patients. Anesthesiology 1994; 81: 1332–1340. Cunningham J, Temple WJ, Mitchell P, Nixon JA, Preshaw RM, Hagen NA. Cooperative hernia study. Pain in the postoperative patient. Ann Surg 1996; 224:598–602.

481

71. Lichtenstein IL, Shulman AG, Amid PK, Montlor MM. Cause and prevention of post herniorrhaphy neuralgia: a proposed protocol for treatment. Am J Surg 1998; 155:786–790. 72. Kark AE, Kurzer M, Waters KJ. Tension-free mesh hernia repair: review of 1098 cases effect local anesthesia in a day unit. Ann R Coll Surg Engl 1995; 77: 299–304. 73. Heise CP, Starling JR. Mesh inguinodynia: a new clinical syndrome after inguinal herniorrhaphy? J Am Coll Surg 1998; 187:514–518. 74. Callesen T, Bech K, Neilsen R, et al. Pain after groin hernia repair. Br J Surg 1998; 85:1412–1414. 75. Barth RJ Jr, Burchard KW, Tosteson A, et al. Short term outcome after mesh or shouldice herniorrhaphy: a randomized, prospective study. Surgery 1998; 123:121–126. 76. Buckhart FL, Daly JW. Sciatic and peroneal nerve injury: a complication of vaginal operations. Obstet Gynecol 1966; 28:99–102. 77. Rose HA, Hood RW, Otis JC, Ranawat CS, Insall JN. Peroneal nerve palsy following total knee arthroplasty: a review of the Hospital for Special Surgery experience. J Bone Joint Surg Am 1982; 64:347–351. 78. James SE, Wade PJ. Lateral popliteal nerve palsy as a complication of the use of a continuous passive motion knee machine-a case report. Injury 1987; 18:72–73. 79. Weber ER, Daube JR, Coventry MB. Peripheral neuropathies associated with total hip arthroplasty. J Bone Joint Surg Am 1976; 58:66–69. 80. Warner MA, Martin JT, Schroeder DR. Offord KP, Chute CG. Lower extremity motor neuropathy associated with surgery performed on patients a lithotomy position. Anesthesiology 1994; 81:6–12. 81. Dornette WH. Compression neuropathies: medical aspects and legal implications. Int Anesthesiol Clin 1986 (Winter); 24:201–229. 82. Rosenblum J, Schwartz GA, Bendler E. Femoral neuropathy-a neurological complication of hysterectomy. JAMA 1966; 195:409–414. 83. X-femoral neuropathy and type II diabetics. 84. Warner MA, Warner DO, Matsumoto JY, Schroeder DR, Offord KP, Maxson PM. Ulnar neuropathy in surgical patients [abstract]. Anesthesiology 1996; 85(suppl):A921. 85. Patt R. Neurosurgical interventions for chronic pain problems. Anesth Clin N Am 1987; 5:609. 86. Ramamurthy S, Walsh NE, Schoenfeld LS, et al. Evaluation of neurolytic blocks using phenol and cryogenic block in the management of chronic pain. J Pain Symptom Manage 1989; 4(2):72–75. 87. Pender JW, Pugh DG. Diagnostic and therapeutic nerve blocks: necessity for roentgenograms. JAMA 1951; 146:798. 88. Raj PP, Rosenblatt R, Montgomery S. Uses of the nerve stimulator for peripheral blocks. Reg Anesth 1980; 5:14. 89. Shealy CN, Mortimer JT, Reswick J. Electrical inhibition of pain by stimulation of the dorsal column: preliminary clinical reports. Anesth Analg 1967; 46:489–491. 90. Melzack R, Wall P. Pain mechanisms: a new theory. Science 1965; 150:971–978. 91. Graves DA et al. Ann Intern Med 1983; 99:360–366. 92. Bach S, Noreng MF, Tjellden NU. Phantom limb pain in amputees during the first 12 months following limb amputation after preoperative lumbar epidural blockade. Pain 1988; 33:297–301.

38 Complications After Craniotomy Andrew Jea and Nizam Razack Department of Neurological Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A.

The care of a patient with head injury can be particularly intimidating to the uninitiated physician. The severe nature of an extensive neurologic insult requires that the patient receive care in the intensive care unit. This chapter will describe the most common complications that occur after a patient undergoes a craniotomy: mass lesions, central nervous system infections, cerebral infarctions, and neurogenic metabolic imbalances. The chapter will also include an algorithm for managing complications after craniotomy.

MASS LESIONS Hematoma Before patients undergo intracranial procedures, their clotting ability should be carefully evaluated. Platelet counts should exceed 100,000 and the platelets should be functional. Essential screening tests include prothrombin time, partial thromboplastin, and international normalized ratio. It is essential that dissemination intravascular coagulation, if present, be detected during or prior to surgery (1,2). A postoperative hematoma may be extra-axial, occurring in the epidural or subdural space (Fig. 1), or intra-axial, occurring within the parenchyma of the brain (3,4). The clinical presentation of these lesions is not specific, and both should be included in the differential diagnosis for postoperative patients who become increasingly lethargic and exhibit focal signs such as hemiparesis, aphasia, cranial nerve palsy (5), or seizure. Changes in vital signs, such as Cushing’s triad (hypertension, bradycardia, and abnormal respiratory pattern), may reflect increasing intracranial pressure (ICP) (6,7). The routine technical use of tack-up sutures to surrounding bone structures reduces the occurrence of epidural hematoma. Waxing the bone edges is also fundamental. Follow-up computed tomography (CT) scans after surgery should not show depression of the dura. In ideal circumstances, the goal of closure is to reduce the dead space in the epidural compartment.

Achieving hemostasis before closing the dura is essential. Occasionally a subdural drain is necessary when oozing cannot be controlled despite all efforts. Some subdural hematomas resolve spontaneously after surgery, but others may cause significant mass effect and require reexploration and drainage. Intraparenchymal hematomas develop at the point of maximal dissection or retraction (8). These hematomas are common after biopsy or resection of tumors (Fig. 2). Increasing the patient’s blood pressure intraoperatively may help identify potential sources of hemorrhage. Hematomas are usually treated by reexploration and evacuation of the mass lesion. Their development is usually recognized by bedside evaluation. Determining the anatomic location of the hematoma—epidural,

Figure 1 Development of right temporoparietal epidural hematoma after evacuation of left subdural hematoma and decompressive craniectomy.

484

Jea and Razack

Figure 2 (A) Computed tomography scan of brain immediately after stereotactic biopsy of right parietal high-grade glioma. (B) Blossoming of intraparenchymal hematoma two hours after stereotactic biopsy of right parietal high-grade glioma.

subdural, or intracerebral—is unimportant during the initial examination (8–11). A CT scan should be obtained if a developing hematoma is suspected and the patient’s condition is stable enough to permit a rapid work-up (12). However, if the patient’s condition is rapidly deteriorating, the more prudent course may be immediate exploration after localization of the lesion by a brief but precise neurologic examination. Regardless of whether the patient has to undergo CT scanning or emergent reexploration, some steps can be taken immediately to increase the patient’s tolerance of the suspected mass lesion (13). Mannitol is an effective osmotic agent and will rapidly decrease ICP (14–20). It is given as an initial intravenous bolus of 1 to 1.5 g/kg body weight with subsequent doses of 0.5 g/kg body weight every four hours (9). Measuring serum osmolarity is important for avoiding a hyperosmolar state with the potential for severe metabolic derangements and acute renal failure. Because of this danger, mannitol should be infused through a central line to allow rapid resuscitation should the effects of mannitol become supratherapeutic (21). Mannitol should be discontinued if the serum osmolarity is greater than 320 mOsm because the effect of mannitol, as an osmotic agent, plateaus when the serum osmolarity exceeds this level (22). Patients whose neurological condition is deteriorating and whose level of consciousness is decreasing require an endotracheal tube not only for airway protection, but also for allowing hyperventilation (9,16,23–26). Monitoring arterial blood gases is essential, and the ventilator setting should be adjusted

to keep the patient hyperventilated with a partial pressure of carbon dioxide level between 30 and 35 torr (15,24,27,28). If an extraventricular drain is in place, it may be opened to allow cerebrospinal fluid (CSF) to drain at a constant height above the ear (5 to 10 cm should suffice). This procedure will result in decompression but not in overdrainage (29–41). Although postoperatively the patient’s condition may deteriorate as the result of a hematoma, such deterioration may also be caused by brain edema or infarct, neither of which requires or responds to reexploration. For this reason, CT scanning is valuable when the patient’s condition is stable enough to allow it. The patient whose condition deteriorates many days or weeks after craniotomy may have a chronic subdural hematoma (9,42). This lesion appears on CT scan as a lucent, subdural mass lesion and can be drained by burr holes with gentle aspiration, provided that the collection is not associated with enclosing membranes, which may be detected on a CT scan with contrast (43). A chronic subdural hematoma with membranes can be evacuated only by full craniotomy. If the subdural hematoma appears isodense on CT scan, the dense, acute clot is at an intermediate stage of liquefaction before acquiring the final lucent, liquid appearance of the chronic subdural hematoma (29). The presence of a chronic subdural hematoma is suggested by clinical changes and by the appearance on a CT scan of a mass effect, i.e., a midline shift with ventricular collapse and sulcus effacement. Such a hematoma can usually be evacuated by burr holes (44).

Chapter 38: Complications After Craniotomy

Brain Edema When brain edema (45) occurs during the postoperative period, its cause may be excessive retraction or intraoperative trauma (46). On CT scan, brain edema appears as an area of decreased density associated with brain shift. Brain edema is commonly associated with intracerebral hemorrhage and contusion. Edema associated with cerebral infarction generally indicates a severe stroke and may lead to herniation. All of these causes may be seen during the postoperative period. Edema often accompanies neoplastic lesions and is more commonly associated with metastatic tumors. The treatment of brain edema depends on the cause of the lesion. Lesions caused by neoplasia or inflammation respond to treatment with steroids. The role of steroids in treating edema caused by trauma, infarction, or anoxia is unproven (47–50). In fact, using steroids in these settings may actually increase morbidity and mortality by increasing risk of septic complications or gastrointestinal hemorrhage. Brain edema that occurs after surgery for trauma, infarction, or hemorrhage represents increased tissue water. It exerts a mass effect and will usually be accompanied by an increase in ICP. This vasogenic edema is best treated with an intravenous infusion of mannitol. The typical dosage is an initial bolus of 1 g/kg followed by a maintenance dose of 0.5 g/kg every four hours, not to exceed a serum osmolarity of 320 mOsm (51).

Pneumocephalus Pneumocephalus is simply the accumulation of air in the intracranial spaces. It commonly occurs after craniotomy if the air is not completely evacuated before the bone flap is replaced. It may also occur after a traumatic basilar skull fracture when air is introduced into the subarachnoid space by communication with the exterior environment, usually through the ethmoid, sphenoid, or frontal sinuses. Pneumocephalus may cause a patient to become lethargic and confused (52). A CT scan may show the accumulation of air beneath a bone flap or in communication with one of the sinuses (53). Most cases of pneumocephalus are treated with 100% oxygen by a nonrebreather mask, followed by keeping the patient in a completely flat position. Tension pneumocephalus marked by an enlarging pocket of air causing mass effect (midline shift, sulcal effacement, or both) demands more aggressive and invasive intervention. Emergency surgery is necessary to resolve the mass effect. Pneumocephalus indicates communication between the exterior environment and the intracranial cavity. This condition can be a precursor of CSF leakage. CSF may drain through the ethmoid or sphenoid sinus complex, causing rhinorrhea, or through the mastoid air cells, causing otorrhea. Although pneumocephalus indicates a tear in the dura, a CSF leak indicates a relatively large dural tear allowing a

485

stream of CSF to flow. CSF may also leak from the scalp suture line; the flow of CSF to the external environment should stop after a watertight closure has been established by oversewing the suture line. When rhinorrhea or otorrhea occurs postoperatively, it should be treated conservatively with a lumbar drain. The drain should be opened and the height should be adjusted to allow a flow of 10 to 15 cc/hr for three to five days. Before the lumbar drain is removed, it should be clamped for 24 hours to determine whether any further drainage from the nose or ear will occur. If a seal is not accomplished after 10 to 14 days of conservative treatment, surgical intervention is necessary. The use of antibiotics to treat pneumocephalus alone or pneumocephalus with subsequent CSF leak is controversial. Treatment with antibiotics should not be initiated unless signs and symptoms of CSF infection develop. Prophylactic treatment with antibiotics should be initiated only if the patient has sinusitis.

Hydrocephalus Hydrocephalus or a loculated (‘‘trapped’’) ventricle may cause symptoms resembling those caused by focal, expanding mass lesions. A loculated ventricle occurs when the drainage pathway from one lateral ventricle into the third ventricle is blocked. In the postoperative period, this blockage typically results from unilateral intraventricular hemorrhage, which causes a blood clot at the foramen of Monro, or from a midline shift, which causes a block at the foramen of Monro by obliteration. The patient will become lethargic and will demonstrate progressive signs of hemiparesis consistent with that caused by a hemispheric mass lesion. Hydrocephalus can be diagnosed by CT scan; diagnosis must be followed by permanent drainage of the loculus. The treatment of choice is emergent ventriculostomy and placement of a shunt. There are two primary types of hydrocephalus: communicating and noncommunicating. Communicating hydrocephalus blocks the reabsorption of CSF downstream from the foramen of Luschka to its point of reabsorption through the arachnoid villi into the major venous sinuses. The point of obstruction may be at the tentorial incisure as the result of scarring from meningitis or damage to the arachnoid granulations. The most common cause of communicating hydrocephalus in the postoperative period is the blockage of absorption pathways by subarachnoid blood. Communicating hydrocephalus causes the patient’s condition to deteriorate slowly, generally over a period of days. No focal signs are usually present; the patient may exhibit gait ataxia, difficulty with memory, or urinary incontinence. A CT scan shows universal dilation of all ventricles. Lumbar puncture may demonstrate a high opening pressure. The CSF may be xanthochromic with a high protein level, indicating the viscous nature of the fluid and the potential

486

Jea and Razack

problem in filtering it through the arachnoid villi. Serial lumbar punctures may be performed as a temporizing measure to diagnose and treat communicating hydrocephalus. If the patient’s neurological condition improves after lumbar puncture, definitive treatment by shunting may be required. Any lesion that causes an obstruction at the narrow fourth-ventricular inflow or outflow track can create noncommunicating or obstructive hydrocephalus. Such lesions include cerebellar edema or infarct after surgery to the posterior fossa or an intraventricular blood clot in the fourth ventricle. It is often difficult to determine the pathophysiologic cause of hydrocephalus; however, resolution is the primary focus of treatment. Noncommunicating hydrocephalus usually causes rapid deterioration of the patient’s condition. The patient may at first be agitated but then enters a comatose state. Sedating a patient who is agitated after craniotomy without first ruling out hydrocephalus as the cause of this alteration in mental status is a common fatal mistake because sedation masks the increase in ICP and the eventual herniation. Therefore, patients at risk of blockage of CSF flow, such as those who have recently undergone surgery to the posterior fossa or those with intraventricular hemorrhage, require more careful, anticipatory observation for the signs of deterioration caused by acute hydrocephalus. A ventriculostomy kit should be placed at the bedside of the patient so that immediate decompression can be provided if necessary. In contrast to patients who have a communicating hydrocephalus, patients with an obstructive hydrocephalus can never be safely treated with lumbar puncture because the pressure gradient created by this procedure places the patient at risk of tonsillar herniation and sudden death. The patient may be temporarily stabilized with a ventriculostomy to provide decompression by draining CSF out of the intracranial cavity. Permanent shunt placement is the definitive treatment for obstructive hydrocephalus. Obstructive hydrocephalus is commonly associated with lesions of the posterior fossa and is a dreaded complication of surgical procedures to this area of the brain. After surgery to the posterior fossa, obstructive hydrocephalus must be considered as dangerous a complication as postoperative hematoma.

INFECTION Meningitis Meningitis is an infection of the leptomeninges (the pia and the arachnoid) and thus of the subarachnoid space (54). This space is continuous from the hemispheric convexities to the lumbosacral subarachnoid space. Infection of the subarachnoid space can be diagnosed by sampling the CSF through a lumbar puncture. Infection localized to the subdural space (subdural empyema) may leave the ventricular and lumbar fluids sterile with little more than a parameningeal

reaction or reactive pleocytosis. The same is true of cerebritis and brain abscess, unless there is erosion into the ventricular system or the subarachnoid space. Meningitis typically causes high fever, meningismus, positive Kernig’s and Brudzinski’s signs, headache, a depressed level of consciousness, seizures, syndrome of inappropriate antidiuretic hormone (SIADH) secretion, and, in severe advanced cases, diabetes insipidus (DI). Meningitis may occur as late as four weeks after surgery because of violation of mastoid air cells in the face of a CSF leak. After craniotomy, the patient’s preoperative depressed level of consciousness may persist, rendering the patient unable to complain of headaches; the patient may also be predisposed to seizures or meningeal irritation as the result of blood in the subarachnoid spaces after the surgical procedure. Unfortunately, after craniotomy the patient may exhibit all of the clinical signs of an aseptic meningitis, including fever; therefore, the diagnosis may depend entirely upon examination of CSF and careful observation. The manifestations of postoperative meningitis are often much more subtle than those of the typical pneumococcal or meningococcal variety. If signs of meningeal irritation should occur in isolation or in association with any other changes, neurologic or metabolic, examination of the CSF is mandatory before any antibiotics are administered. Because cell count, glucose concentration, and protein concentration are abnormal after craniotomy, an absolute diagnosis must await the result of CSF culture or the demonstration of bacteria on gram stain. Empiric treatment with broad-spectrum intravenous antibiotics should be started while the results of culture are awaited. Therapy directed at gram-positive cocci and gram-negative organisms must be instituted. The antibiotic regimen should then be tailored once the final culture results and sensitivities have been obtained.

Ventriculitis The clinical picture of ventriculitis differs little from that of meningitis, although the presentation is usually much more subtle. Meningeal symptoms may be minimal and fever variable, whereas alteration in mental status and neurologic function predominate. Diagnosis requires careful observation; the only diagnostic test is microscopic and bacteriologic examination of the ventricular fluid. Both meningitis and ventriculitis tend to occur in the postoperative period more than three days after violation and contamination of the subarachnoid or ventricular space. The usual postoperative effects of operative trauma and brain edema begin to resolve during this period. Any reversal in this pattern of healing should alert the clinician to infection in one of these spaces. Both meningitis and ventriculitis may be associated with elevated ICP, and infection should be considered in a patient with increasing ICP, especially if the elevation has no clear or reasonable cause and if no mass effect is visible on CT scan.

Chapter 38: Complications After Craniotomy

Again, selecting a treatment regimen depends on the results of CSF cultures and sensitivities. In the meantime, the intrathecal administration of antibiotics may be considered so as to provide broad-spectrum coverage. The antibiotic regimen can be tailored once the results of final cultures and sensitivities have been obtained.

487

Inflammatory effects are predominant and treatment with drainage and antibiotics are the gold standard. The drugs and doses are the same as those recommended for meningitis, which may also be present. Drainage may be accomplished by reoperation or burr holes, and many surgeons recommend placing subdural catheters for irrigation of this space with antibiotic solutions such as concentrated bacitracin.

Abscess Brain abscess (55)—or its immediate precursor, cerebritis—is relatively rare in the postoperative period. The development of meningeal signs or infected CSF in the face of focal deficits suggests that this process must be ruled out. The absence of focal deficit does not rule out the presence of abscess. If an abscess does not communicate with the ventricular or subarachnoid space, meningeal signs will usually be absent. In 95% of cases of cerebral abscesses, the CSF may be completely normal and the patient can be afebrile. As is the case with meningitis or ventriculitis, steroids may suppress or delay neurologic change in the developing abscess; therefore, abscess must be considered when a patient’s condition worsens after discontinuation of steroids. The treatment (55) of brain abscess is the same as that of any other abscess: incision and drainage. This procedure is also diagnostic. Needle aspiration combined with the administration of high-dose antibiotics will clear approximately 80% to 85% of abscesses. The remainder will require craniotomy for complete cure. If infection also involves the craniotomy flap, then reoperation, bone flap removal, and drainage of the abscess should be carried out for definitive therapy. The increased use of intraventricular antibiotics in the past decade has provided an effective means of treating certain forms of infection, especially meningitis and ventriculitis caused by highly resistant gram-positive and gram-negative organisms.

Subdural Empyema Subdural empyema is a specific form of abscess (56,57). This entity is also marked by neurologic deterioration, with the development of focal signs of hemiparesis, seizures, or both. The seizures associated with subdural empyema tend to begin focally, then become generalized, and then quickly progress to status epilepticus. These neurologic findings are related to mass effect from edema. Unlike subdural hematomas, subdural empyemas are associated with edema that is out of proportion to the volume of fluid in the subdural space. Subdural empyema is rare after craniotomy but may follow burr hole drainage of a chronic subdural hematoma. Most subdural empyemas are associated with chronic sinusitis in adults or with middle ear infections in young children and infants. Diagnosis by CT scan may be difficult and a high index of suspicion is required. However, a parafalcine subdural collection, which can be seen on CT scan, is pathognomonic of subdural abscess.

INFARCTIONS Arterial Infarcts Infarctions resulting from occlusion of the arterial vascular supply are most commonly associated with atherosclerotic disease. Arterial infarct is a rare complication after craniotomy but may occur if there has been substantial intraoperative manipulation of cerebral vessels (7,16,58–60). Intraoperative coagulation or ligation of bleeding vessels in patients without good collateral circulation leads to postoperative infarction. An intraoperative angiogram may be obtained to predict the deficits that the patient is expected to have in the postoperative period as the result of the arterial territory loss. A CT scan performed in the immediate postoperative period may not show areas of infarct; however, a second CT scan 24 to 48 hours later will show areas of hypodensity representing infarct (61). Clinically, the patient will usually exhibit focal neurological deficits. If a large area or bilateral areas of the brain are involved, the patient may experience a global decrease in level of consciousness and more extensive neurologic deficits. If the stroke is detected early, an attempt to save the penumbra should be made by improving blood flow through collateral vessels to keep the patient euvolemic, hypertensive, and anticoagulated. Any involvement of the arteries serving the cerebellum may lead to a cerebellar infarct; thus, immediate neurosurgical attention is necessary. Cerebellar infarction places the patient at high risk of obstructive hydrocephalus as the cerebellar swelling continues to deform and occlude the fourth ventricle. In the conscious patient, there is an orderly progression of clinical signs and symptoms. Symptoms and signs related to cerebellar dysfunction, such as dizziness, vertigo, nausea, vomiting, truncal ataxia, nystagmus, and dysarthria, appear first. Next, the patient may suffer from the onset of hydrocephalus with symptoms of headaches, agitation, and finally obtundation. The patient suffers a palsy of the sixth cranial nerve that cannot be overcome by doll’s eye maneuver, followed by a peripheral deficit of the seventh cranial nerve, resulting in hemifacial weakness. The development of cranial dysfunction necessitates neurosurgical intervention for decompression of the posterior fossa.

Venous Infarcts Venous infarcts are generally seen after craniotomy, especially if the venous sinuses are involved in the

488

Jea and Razack

surgical field. Repair of dural sinus lacerations or prolonged compression of a sinus by an extrinsic force places the patient at risk of venous sinus thrombosis and infarction postoperatively (62). In the conscious patient, thrombosis results in symptoms that include headache, nausea, vomiting, seizures, and symptoms resembling those caused by pseudotumor cerebri. Cerebral venous thrombosis and/or dural sinus thrombosis can lead to venous infarction (63). This infarction may be hemorrhagic and often involves the subcortical white matter and traverses the typical arteriovascular boundaries seen on CT scan. In the absence of hemorrhage, the patient should be kept euvolemic to hypervolemic to prevent further exacerbation of thrombosis in the dehydrated state. The component of hemorrhage or significant mass effect resulting from edema becomes a neurosurgical emergency. Evacuation of the clot may be necessary, as may decompressive craniectomy to reduce the increase of ICP in a patient whose condition is deteriorating.

METABOLIC IMBALANCES Hyponatremia Sodium imbalance is the most common metabolic disturbance experienced by a neurosurgical patient. Electrolyte levels should be checked daily after craniotomy until the levels are stable. Sodium concentrations outside the normal physiological range lead to a decreased level of consciousness, disorientation, seizures, and global encephalopathy. After craniotomy, low sodium concentrations may be attributable to cerebral salt wasting (CSW) or SIADH. CSW is a condition in which an unknown mechanism, currently believed to be a signal for overproduction of atrial natriuretic protein (ANP), causes a natriuresis or overexcretion of sodium into the urine. This condition causes the body to respond to hypovolemia by increasing the secretion of antidiuretic hormone (ADH). Water is reabsorbed from the effects of ADH, whereas sodium continues to be excreted under the influence of ANP, thus resulting in a hyponatremic state. In contrast, SIADH causes water to be reabsorbed despite a euvolemic or hypervolemic state, thus also resulting in hyponatremia. It is important to differentiate between the two mechanisms of hyponatremia because they require different treatments. CSW causes the urine electrolytes to show an inappropriately high content of sodium. The patient may have clinical signs and symptoms of dehydration, along with a high serum osmolality. Treatment of this condition involves a combination of fluid restriction and replacement of intravascular volume with a colloid solution high in sodium content, most commonly albumin. Replacing the intravascular volume with colloid breaks the vicious cycle of ADH secretion.

SIADH (7,9) causes a hypervolemic state and may produce the clinical signs and symptoms of low serum osmolality. Urine electrolyte levels will show a high sodium concentration, but not as high as that seen with CSW. The treatment is strict restriction of parenterally and enterally administered fluids. The use of hypertonic saline should be considered for cases in which hyponatremia continues to progress despite the initiation of fluid restriction, the infusion of colloid, or both. The crystalloid solution should be administered at a rate of 20 to 30 cc/hr, and sodium levels should be checked frequently. A too rapid correction of sodium levels may lead to the devastating complication of central pontine myelinolysis. Sodium concentrations should be corrected to 126 mEq/L with hypertonic saline. Hyponatremia after craniotomy is usually temporary. It is important to support the patient until his own physiologic mechanisms of dealing with sodium balance return to full function.

Hypernatremia After craniotomy, hypernatremia frequently occurs with dysfunction of the hypothalamic or pituitary axis resulting from inadvertent manipulation of these areas during surgery. This dysfunction results in DI (64), which causes too much water to be lost in the urine and produces a high serum sodium concentration as the result of inadequate production of ADH. Postoperatively, DI may be diagnosed by using three criteria: (i) urine output of more than 250 cc/hr for two consecutive hours; (ii) an increasing serum sodium concentration higher than 145 mEq/L; and (iii) specific gravity of the urine below 1.005. Continued DI should be managed by administering a hypotonic saline solution delivered at a rate higher than that used for maintenance intravenous fluids. The urine output should be replaced cubic centimeter per cubic centimeter hourly up to a maximum of 250 cc. A conscious patient should be allowed to drink water in an amount equaling that of the urine output. Aqueous pitressin should be used only if the DI is refractory to intravenous fluid therapy or if the patient gets tired of drinking large volumes of water and is unable to keep up with the urine output. A one-time subcutaneous injection of five units of aqueous pitressin should be administered with continued close monitoring of sodium concentrations and urine output. If the DI continues, another injection of aqueous pitressin may be administered. The goal is a slow, gradual correction of the serum sodium concentration. Correcting hypernatremia too rapidly may result in lethal brain edema. After craniotomy, patients are expected to experience only a temporary problem with hypernatremia unless the pituitary stalk or posterior lobe of the pituitary has been completely destroyed. Once the stunned pituitary recovers function, the hypernatremia is expected to resolve.

Chapter 38: Complications After Craniotomy

REFERENCES 1. Clark JA, Finelli RE, Netsky MG. Disseminated intravascular coagulation following cranial trauma. Case report. J Neurosurg 1980; 52:266–269. 2. Pitts LH. Comment on ‘‘The incidence and significance of hemostatic abnormalities in patients with head injuries.’’ Neurosurgery 1989; 24:832. 3. Ribas GC, Jane JA. Traumatic contusions and intracerebral hematomas. J Neurotrauma 1992; 9:S265–S278. 4. Miller JD, Murray LS, Teasdale GM. Development of a traumatic intracerebral hematoma after a ‘‘minor’’ head injury. Neurosurgery 1990; 27:669–673. 5. Marshall LF, Barba D, Toold BM, Bowers SA. The oval pupil: clinical significance and relationship to intracranial hypertension. J Neurosurg 1983; 58:566–568. 6. Savitz SI. Cushing’s contributions to neuroscience, part 1: through the looking glass. Neuroscientist 2000; 6:411–414. 7. Savitz SI. Cushing’s contributions to neuroscience, part 2: Cushing and several dwarfs. Neuroscientist 2001; 7:469–473. 8. Lobato RD, Rivas JJ, Gomez PA, et al. Head-injured patients who talk and deteriorate into coma. Analysis of 211 cases studied with computerized tomography. J Neurosurg 1991; 75:256–261. 9. Miller SM. Management of central nervous system injuries. In: Turndorf H, Capan LM, Miller SM, eds. Trauma: Anesthesia and Intensive Care. Philadelphia: JB Lippincott Co., 1991:321–355. 10. Rockswold GL, Pheley PJ. Patients who talk and deteriorate. Ann Emerg Med 1993; 22:1004–1007. 11. Andrews BT. Management of delayed posttraumatic intracerebral hemorrhage. Contemp Neurosurg 1988; 10:1. 12. Wilberger JE Jr, Rothfus WE, Tabas J, Goldberg AL, Deeb ZL. Acute tissue tear hemorrhages of the brain: computed tomography and clinicopathological correlations. Neurosurgery 1990; 27:208–213. 13. Woster PS, LeBlanc KL. Management of elevated intracranial pressure. Clin Pharm 1990; 9:762–772. 14. Muizelaar JP, Lutz HA, Becker DP. Effect of mannitol on ICP and CBF and correlation with pressure autoregulation in severely head-injured patients. J Neurosurg 1984; 61:700–706. 15. Rosner MJ, Daughton S. Cerebral perfusion pressure management in head injury. J Trauma 1990; 30:933–940. 16. Rosner MJ, Coley I. Cerebral perfusion pressure: a hemodynamic mechanism of mannitol and the postmannitol hemogram. Neurosurgery 1987; 21: 147–156. 17. Bouma GJ, Muizelaar JP, Bandoh K, Marmarou A. Blood pressure and intracranial pressure-volume dynamic in severe head injury: relationship with cerebral blood flow. J Neurosurg 1992; 77:15–19. 18. Cruz J, Minoja G, Okuchi K, Facco E. Successful use of the new high-dose mannitol treatment in patients with Glasgow Coma Scale scores of 3 and bilateral abnormal papillary widening: a randomized trial. J Neurosurg 2004; 100:376–383. 19. Muizelaar JP, Wei EP, Kantos HA, Becker DP. Mannitol causes compensatory cerebral vasoconstriction and vasodilation in response to blood viscosity changes. J Neurosurg 1983; 59:822–828. 20. Schrot RJ, Muizelaar JP. Mannitol in acute traumatic brain injury. Lancet 2002; 359:1633–1634.

489

21. Burke AM, Quest DO, Chien S, Cerri C. The effects of mannitol on blood viscosity. J Neurosurg 1981; 55:550–553. 22. Marshall LF, Smith RW, Rauscher LA, Shapiro HM. Mannitol dose requirements in brain-injured patients. J Neurosurg 1978; 48:169–172. 23. Diringer MN, Videen TO, Yundt K, et al. Regional cerebrovascular and metabolic effects of hyperventilation after severe traumatic brain injury. J Neurosurg 2002; 96:103–108. 24. Fessler RD, Diaz FG. The management of cerebral perfusion pressure and intracranial pressure after severe head injury. Ann Emerg Med 1993; 22:998–1003. 25. Oertel M, Kelly DF, Lee JH, et al. Efficacy of hyperventilation, blood pressure elevation, and metabolic suppression therapy in controlling intracranial pressure after head injury. J Neurosurg 2002; 97:1045–1053. 26. Lee JH, Kelly DF, Oertel M, et al. Carbon dioxide reactivity, pressure autoregulation, and metabolic suppression reactivity after head injury: a transcranial Doppler study. J Neursurg 2001; 95:222–232. 27. Artru AA. Reduction of cerebrospinal fluid pressure by hypocapnia: changes in cerebral blood volume, cerebrospinal fluid volume, and brain tissue water and electrolytes. J Cereb Blood Flow Metab 1987; 7:471–479. 28. Newell DW, Weber JP, Watson R, Aaslid R, Winn HR. Effect of transient moderate hyperventilation on dynamic cerebral autoregulation after severe head injury. Neurosurgery 1996; 39:35–43. 29. Stern WE. Intracranial fluid dynamics: the relationship of intracranial pressure to the Monro-Kellie doctrine and the reliability of pressure assessment. J R Coll Surg Edinb 1963; 168:18–36. 30. Kosteljanetz M. Intracranial pressure: cerebrospinal fluid dynamics and pressure-volume relations. Acta Neurol Scand Suppl 1987; 111:1–23. 31. Maset AL, Marmarou A, Ward JD, et al. Pressurevolume index in head injury. J Neurosurg 1987; 67:832–840. 32. Lundberg N. Continuous recording and control of ventricular fluid pressure in neurosurgical practice. Acta Psychiatr Scand 1960; 36(suppl 149):1–193. 33. Neff S, Subramaniam RP. Monro-Kellie doctrine. J Neurosurg 1996; 85:1195. 34. The Brain Trauma Foundation. The American Association of Neurological Surgeons. The joint section on neurotrauma and critical care. Recommendations for intracranial pressure monitoring technology. J Neurotrauma 2000; 17:497–506. 35. Narayan RK, Kishore DR, Becker DP, et al. Intracranial pressure: to monitor or not to monitor? A review of our experience with severe head injury. J Neurosurg 1982; 56:650–659. 36. Yano M, Kobayashi S, Otsuka T. Useful ICP monitoring with subarachnoid catheter method in severe head injuries. J Trauma 1988; 28:476–480. 37. The Brain Trauma Foundation. The American Association of Neurological Surgeons. The joint section on neurotrauma and critical care. Indications for intracranial pressure monitoring. J Neurotrauma 2000; 17:479–491. 38. Zentner J, Duffner F, Behrens E. Percutaneous needle trephination for external CSF drainage: experience with 226 punctures. Neurosurg Rev 1995; 18:31–34. 39. Caruselli G, Recchioni MA, Occhipinti C, Bernardini M, Caruselli M. The role of CSF ventricular drainage in

490

40.

41. 42. 43. 44.

45.

46. 47.

48.

49.

50. 51.

Jea and Razack

controlling intracranial hypertension in patients with brain lesions. Comparison of three methods. Preliminary results. J Neurosurg Sci 1992; 36:219–225. Gudeman SK, Young HF, Miller JD, Ward JD, Becker DP. Indications for operative treatment and operative technique in closed head injury. In: Becker DP, Miller JD, Povlishock J, Selhorst J, Young H, eds. Textbook of Head Injury. Philadelphia: WB Saunders, 1989:138–181. Marmarou A, Maset AL, Ward JD, et al. Contribution of CSF and vascular factors to elevation of ICP in severely head-injured patients. J Neurosurg 1987; 66:883–890. Yamashima T, Friede RL. Why do bridging veins rupture into the virtual subdural space? J Neurol Neurosurg Psychiatr 1984; 47:121–127. Markwalder TM. Chronic subdural hematomas: a review. J Neurosurg 1981; 54:637–645. Lee JY, Ebel H, Ernestus RI, Klug N. Various surgical treatments of chronic subdural hematoma and outcome in 172 patients: is membranectomy necessary? Surg Neurol 2004; 61:523–527. Durward QJ, Del Maestro RF, Amacher A, Farrar JK. The influence of systemic arterial pressure and intracranial pressure on the development of cerebral vasogenic edema. J Neurosurg 1983; 59:803–809. Yoshino E, Yamaki T, Higuchi T, Horikawa Y, Hirakawa K. Acute brain edema in fatal head injury: analysis by dynamic CT scanning. J Neurosurg 1985; 63:830–839. Gudeman SK, Miller JD, Becker DP. Failure of high dose steroid therapy to influence intracranial pressure in patients with severe head injury. J Neurosurg 1979; 51:301–306. Dearden NM, Gibson JS, MacDowall DG, Gibson RM, Cameron MM. Effect of high dose dexamethasone on outcome from severe head injury. J Neurosurg 1986; 64:81–88. Braakman R, Schouten HJ, Blaauw-van Dishoeck M, Minderhoud JM. Megadose steroids in severe head injury. Results of a prospective double-blind trial. J Neurosurg 1983; 58:326–330. Giannotta SL, Weiss H, Apuzzo ML, Martin E. High dose glucocorticoids in the management of severe head injury. Neurosurgery 1984; 15:497–501. Miller JD, Leech P. Effects of mannitol and steroid therapy on intracranial volume-pressure relationships in patients. J Neurosurg 1975; 42:274–281.

52. Markham JW. The clinical features of pneumocephalus based upon a survey of 284 cases with report of 11 additional cases. Acta Neurochir (Wien) 1967; 16:1–78. 53. Osborn AG, Daines JH, Wing SD, Anderson RE. Intracranial air on computerized tomography. J Neurosurg 1978; 48:355–359. 54. Baltas I, Tsoulfa S, Sakellariou P, Vogas V, Fylaktakis M, Kondodimou A. Posttraumatic meningitis: bacteriology, hydrocephalus, and outcome. Neurosurgery 1994; 35:422–427. 55. Stephanov S. Surgical treatment of brain abscess. Neurosurgery 1988; 22:724–730. 56. Kubik CS, Adams RD. Subdural empyema. Brain 1943; 66:18–42. 57. Dill SR, Cobbs CG, McDonald CK. Subdural empyema: analysis of 32 cases and review. Clin Infect Dis 1995; 20:372–386. 58. Martin NA, Doberstein C, Zane C, Caron MJ, Thomas K, Becker DP. Posttraumatic cerebral arterial spasm: transcranial Doppler ultrasound, cerebral blood flow, and angiographic findings. J Neurosurg 1992; 77: 575–583. 59. Hoh BL, Topcuoglu MA, Singhal AB, et al. Effect of clipping, craniotomy, or intravascular coiling on cerebral vasospasm and patient outcome after aneurysmal subarachnoid hemorrhage. Neurosurgery 2004; 55: 779–786. 60. Suwanwela C, Suwanwela N. Intracranial arterial narrowing and spasm in acute head injury. J Neurosurg 1972; 36:314–323. 61. Krause GS, White BC, Aust SD, Nayini NR, Kumar K. Brain cell death after ischemia and reperfusion: a proposed biochemical sequence. Crit Care Med 1988; 16:714–726. 62. Ferrera PC, Pauze DR, Chan L. Sagittal sinus thrombosis after closed head injury. Am J Emerg Med 1998; 16:382–385. 63. Stefini R, Latronico N, Cornali C, Rasulo F, Bollati A. Emergent decompressive craniectomy in patients with fixed dilated pupils due to cerebral venous and dural sinus thrombosis: report of three cases. Neurosurgery 1999; 45:626–629. 64. Griffin JM, Hartley JH Jr, Crow RW, Schatten WE. Diabetes insipidus caused by craniofacial trauma. J Trauma 1976; 16:979–984.

39 Spinal Cord Trauma Michael Y. Wang, Iftikharul Haq, and Barth A. Green Department of Neurological Surgery & The Miami Project to Cure Paralysis, University of Miami School of Medicine, Miami, Florida, U.S.A.

EPIDEMIOLOGY Accounts of spinal cord injury date back more than four millennia to the Edwin Smith Papyrus (1,2). The record of a case managed by Imhotep, physician to Pharoh Zoser III, describes incontinence, paralysis, and loss of sensation. Imhotep’s recommendation that cervical spine injuries were ‘‘an ailment not to be treated’’ persisted until the last half-century. Until the advent of modern nursing care and antibiotics, even young patients were quick to succumb to pneumonia, sepsis, and thromboembolism after spinal cord injuries. Indeed, as recently as 1924, the British Medical Council stated, ‘‘the paraplegic may live a few years in a state of more or less ill-health’’ (3). Sir Ludwig Guttman established the Spinal Injuries Centre of Stoke Mandeville Hospital in 1944. This center was established in part in response to the devastating casualties of World War II; it focused on providing aggressive medical care for paraplegic patients. Physical therapy, occupational therapy, and nursing services targeted at returning these patients to independent living were successful in prolonging the life expectancy and improving the quality of life of these patients (3). Since that time, the proliferation of spinal cord injury centers, particularly in the Veterans Administration Medical Centers in America, has substantially improved the outlook for paraplegic and quadriplegic patients. The patient with spinal cord injury now has a life expectancy approximating that of uninjured adults. The past half-century has seen remarkable advances in spinal instrumentation technology. Restoring spinal stability through internal fixation has obviated the need for prolonged immobilization, thereby reducing the risk of medical complications; the increasing safety of general anesthesia and improvements in microsurgical technique raise the question of whether emergent surgery to relieve compressed neural structures may be beneficial even after the loss of neurologic function; and the advent of effective pharmacologic interventions to limit secondary injury offers the hope of finding a cure for this ailment.

Traumatic spinal cord injuries are a serious public health problem in North America. Each year an estimated 11,000 new cases occur, but the tragedy of spinal cord injury lies in its devastating effect on predominantly young, healthy adults between 15 and 35 years of age (4). Because of improvements in prehospital care and post-trauma medical and surgical management, survival even after severe injuries is commonplace. As a consequence, it is estimated that more than 200,000 patients with spinal cord injury are currently alive in the United States at an annual financial cost of roughly $4 billion (3). Motor vehicle accidents are the most common mechanism of injury (55%), followed by occupationrelated trauma (22%), sports injuries (18%), and assault 5% (4). Most injuries are due to blunt trauma resulting in fracture, dislocation, or subluxation of the vertebrae, although penetrating injuries from gunshots and stabbings do occur. Injuries are most common at the transitional regions of the spine at the junction where a more mobile segment meets a less mobile one (i.e., craniovertebral, cervicothoracic, and thoracolumbar regions). The cervical spine is the most commonly affected region, followed by the thoracic and lumbar regions (5). When all spinal levels are considered, a fracture or dislocation of the vertebral column carries a 14% chance of neural injury. However, the spinal level of involvement will influence the likelihood of neurological impairment: 40% in the cervical region, 10% in the thoracic region, 35% in the thoracolumbar area, and 3% in the lumbar region.

BIOMECHANICS OF INJURY The neural and musculoskeletal components of the human spine are intimately associated. Thus, any discussion of blunt traumatic spinal cord injury requires an understanding of the vertebral column. Motion

492

Wang et al.

occurs between the 25 distinct vertebrae, and although these motion segments are stereotyped, variabilities exist in each spinal region. Rotatory motion in the axial plane occurs primarily at the occipitocervical and thoracic regions. Flexion and extension in the sagittal plane occur at the cervical and lumbar levels. The orientation and configuration of the facet joints determine to a large extent the degree of mobility at each motion segment. The thoracic spine is also made less mobile because of its articulations with the rib cage. Concepts of stability in the vertebral column are complex and frequently confusing. The vertebral column serves to transmit loads, permit motion, and protect the spinal cord. Instability of the spinal column may then be defined as its failure to perform any of these functions under physiologic levels of mechanical loading. This failure may occur either acutely or in a progressive, delayed manner. In cases of traumatic spinal cord injury, the vertebral column acutely fails to shield the neural elements from external forces as a result of being stressed beyond its mechanical tolerances. Numerous classification schemes have been devised to predict whether the spine is unstable. The most common of these is the three-column theory introduced by Denis (6). Although this theory was originally based on studies of thoracolumbar fractures, its principles have also been applied to other regions of the spine (Fig. 1). This classification system divides the spine into anterior, middle, and posterior columns. The anterior column consists of the anterior half of the vertebral body, the anterior half of the intervertebral disk, and the anterior longitudinal ligament. The middle column consists of the posterior half of vertebral body, the posterior half of the intervertebral disk, and the posterior longitudinal ligament. The posterior column consists of the posterior arch, the facet joint

complex, the interspinous ligament, the supraspinous ligament, and the ligament flavum. The diagnosis of instability is made if two or more of the columns are compromised. External forces placed upon the spine include axial compression, distraction, flexion, extension, and translation. Axial compression in the cervical spine results in disruptions of the ring of C1 and in burst fractures of the remaining vertebrae. Axial compression in the thoracolumbar spine results in burst fractures. When compressive forces are applied anterior to the spinal column and result in a component of flexion, anterior compression fractures result. Severe flexion is the most common mechanism of injury in the cervical spine. This motion can cause odontoid fractures, teardrop fractures of the vertebral bodies, dislocations of the vertebral bodies, and jumped facets. In the thoracolumbar spine, severe flexion results in compression of the anterior vertebral body. If the fulcrum of force is anterior to the vertebral column, as occurs when a seat-belted passenger is involved in a motor vehicle accident, a flexion-distraction injury of the thoracolumbar junction may result.

PATHOPHYSIOLOGY The pathological outcome of trauma to the spinal cord is related to a primary mechanical injury at the epicenter of the damage. Direct crush, stretch, and shear injuries to neurons and axons within the spinal cord lead to immediate cell death. However, delayed cascades of cellular and molecular events known as secondary spinal cord injury occur in the hours and days after the traumatic event and lead to further cell death. The release of excitotoxic amino acids such as glutamate disturbs ionic homeostasis in neural tissues. The resulting increases in intracellular calcium ions, cellular energy failure, and accumulation of free radicals lead to local cell death in a delayed fashion (7,8). Because patients with spinal cord injury frequently suffer polytrauma, they are susceptible to derangements of systemic homeostasis. Cardiovascular and pulmonary compromise may affect perfusion and oxygen delivery to the spinal cord, thereby exacerbating the damage. Vasoactive substances released by injured cells and endothelin released by damaged capillaries may also disrupt the microcirculation of the spinal cord. Ischemia may thus cause neurologic deficits to extend rostrally beyond the initially injured area (9,10). Because cell death due to secondary injury and ischemia occurs after the patient has reached a medical treatment facility, it is hoped that early pharmacologic intervention and maintenance of adequate tissue perfusion can salvage these neurons.

CLINICAL FEATURES Figure 1 Three-column theory of spinal stability as proposed by F. Denis.

The neurological examination is of paramount importance in localizing the probable site of injury.

Chapter 39: Spinal Cord Trauma

Particular attention should be paid to the motor, sensory, reflex, and rectal examinations. On the basis of the degree of functional impairment, the American Spinal Injury Association (ASIA) has proposed an easily used scoring system (Table 1). The ASIA score, in conjunction with the lowest normal segmental level, defines the neurologic injury in simple terms. In this classification scheme, grade ‘‘A’’ denotes a complete injury, and grades ‘‘B’’ through ‘‘D’’ denote incomplete injuries (11). Complete recovery of function after an ASIA ‘‘A’’ injury is exceedingly rare; however, improvement of one or two ASIA grades is seen in more than 10% of patients. Recovery is most likely after grade ‘‘D’’ injuries (12). Neural compression typically results from acute displacement of bone fragments, ligaments, and herniated discs. Delayed spinal cord compression may also result from a hematoma within the spinal canal or from movement of bone or prolapsed disc in a spine that is not properly immobilized. The characteristic clinical picture is that of a patient without neurologic deficits or with an incomplete injury who then experiences complete paralysis, particularly after intubation or transportation. Deterioration can also occur in the chronic setting weeks to months after injury. Post-traumatic syringomyelia and progressive bony deformity are the most frequent causes. Overall, loss of neurologic function occurs in roughly 3% of patients who are admitted for spinal cord trauma (12). Specific neurological syndromes have been described for particular partial cord injuries. The anterior cord syndrome is characterized by complete paralysis and hypoalgesia (anterior and anterolateral column function) below the level of injury with preservation of the senses of position, vibration, and light touch (posterior column function). This syndrome occurs most commonly after ischemia in the territory supplied by the anterior spinal artery, which supplies the corticospinal and spinothalamic tracts. The central cord syndrome is characterized by more pronounced motor dysfunction in the distal upper extremities, accompanied by various degrees of sensory loss and bladder dysfunction. This injury characteristically occurs after a hyperextension injury in elderly patients and can occur even in the absence of any clear radiographic evidence of disruption of the bones or ligaments. Most patients recover the ability to walk, with partial restoration of upper extremity strength. The posterior cord syndrome is uncommon; the senses Table 1 American Spinal Injury Association Grading Scale for Spinal Cord Injury Clinical grade A B C D E

Results of neurological examination No motor or sensory function preserved Sensory but no motor function preserved Nonuseful motor function preserved (< antigravity strength) Motor function preserved but weak Normal motor and sensory function

493

of position and vibration are impaired because of injury to the dorsal columns. The Brown-Sequard syndrome or hemisection cord syndrome causes ipsilateral paresis and loss of proprioception below the level of the lesion and contralateral loss of pain and temperature sensations. This syndrome can be the result of penetrating injuries or tumor compression and is usually not seen in a pure form. The conus medullaris syndrome occurs with injuries at the thoracolumbar junction. This syndrome has components of both spinal cord injury and nerve root injury because of the dense population of lower nerve roots emerging from the caudal end of the spinal cord. Symmetrical lower extremity motor impairment and anesthesia with bowel and bladder dysfunction are typical. Recovery from this syndrome is unlikely. In contrast, partial recovery after cauda equina syndrome is possible with early decompression. Cauda equina injuries occur at spinal levels below the termination of the cord at L1 or L2. Cord concussions exhibit fleeting neurologic symptoms followed by rapid resolution. These injuries, also called ‘‘stingers,’’ occur most commonly among athletes with low-velocity hyperflexion or extension injuries of the cervical spine. Complete recovery is the rule; however, patients should be evaluated meticulously for occult spinal instability and intraspinal hematomas.

EVALUATION OF SUSPECTED INJURY The current medicolegal environment in the United States is intolerant of missed spinal injuries. Indeed, failure to detect spinal instability can cause delayed loss of neurologic function. In the most extreme case, a patient who is not paralyzed by the traumatic event may become quadriplegic after inappropriate mobilization by the medical team. Thus, it is not surprising that tremendous resources and efforts are directed at detecting spinal injuries. The diagnosis of a spinal column injury is based on the findings of clinical and radiological examinations. In a conscious, nonintoxicated patient, the absence of pain along the spinal axis is useful in ruling out injury. In these patients, diagnosis of a low-velocity injury may require no radiography, and diagnosis of a high-velocity injury may require only limited plain films. It is essential that radiographic evidence of spinal column injury be correlated with the findings of the clinical examination because 10% of patients will have injuries at multiple spinal segments. Radiography, computed tomography (CT), and magnetic resonance imaging (MRI) are needed when patients are not able to cooperate fully with the neurologic examination. Radiographs are useful not only for detecting but also for classifying injuries. The fracture type and the degree of cord compression are particularly important aspects of the injury that will determine the

494

Wang et al.

Figure 2 Cervical subluxation with unilateral jumped facet and acute disk herniation causing quadriplegia after a rollover motor vehicle accident. (A) Lateral radiograph showing subluxation at C. (B) Computed tomography scan showing ‘‘reverse hamburger sign’’ of jumped facets on one side. (C) Magnetic resonance image showing a herniated disk compressing the cervical spinal cord. (D) Postoperative radiograph after diskectomy for decompression and fusion with plating.

management strategy. For the cervical spine, plain lateral radiographs must include the C7–T1 junction because 31% of injuries occur between C6 and T1 (Fig. 2). When patients are large and bulky, downward traction on the shoulders, a swimmer’s view, or CT scanning may be needed to properly visualize the cervicothoracic junction. Lateral radiographs allow evaluation of vertebral alignment, canal diameter (normal is >12 mm), angulation of the intervertebral space (normal is 5 cm from ureterovesical junction Transection with loss of ureteral segment

Large segment and/or unhealthy organs

Urethra External layer only Full thickness of the wall Source: From Ref. 23.

Chapter 46:

Complications in Gynecologic Surgery

557

Ureter

After the bladder, the genitourinary organ next most commonly injured during pelvic surgery is the ureter. The most recently reported incidence of ureteral injuries during abdominal hysterectomies is 0.2% (Table 2) (16,24). Ureteral injury is more common during radical hysterectomy; the reported average incidence is 1.2% (16,17,19,20,25,26). The highest risk of urinary injury is associated with procedures performed for ovarian remnant syndrome; injury occurs during as many as 25% of such cases (27). The three segments of the ureter that are most vulnerable to damage because of their relation to the reproductive tract are the segment of ureter adjacent to the infundibular pelvic ligament, the segment at the level of the uterine artery, and the segment adjacent to the attachment of the cardinal ligament and the vaginal cuff (Fig. 2). The ureter may be inadvertently clamped or transected when it is displaced by tumor, when it takes an anomalous course, or when it is involved with adhesions from pelvic inflammatory disease. The segment most commonly damaged is the segment located where the ureter crosses the uterine artery (28). The pelvic surgeon must identify the normal ureteral anatomy at the initiation of the procedure. Before hysterectomies and oophorectomies proceed, the retroperitoneum is entered and dissected between the round ligament and the infundibulopelvic ligament until the course of the ureter has been identified on the medial aspect of the posterior leaf of the broad ligament. When a needle, suture, or clamp is placed on or near the ureter, it should be removed expeditiously and the patient should be given intravenous indigo carmine with 10 mL of furosemide so that the integrity of the ureter can be established. If the ureter is intact and the external sheath is well vascularized and not blanched, no further treatment will be required. If the external sheath is torn, a 4–0 or 5–0 interrupted delayed absorbable suture is placed with care to avoid the muscularis. A full-thickness tear or a break in the external sheath that is larger than 1 cm should be repaired as described above; an 8-Fr ureteral stent is guided into place through an intentional cystotomy. This stent may be removed 14 to 21 days postoperatively. A percutaneous suction catheter is positioned adjacent to the repair and is removed when the amount of drainage has decreased to less than 30 to 50 mL per day. The use of this catheter will assist with the diagnosis and therapy of a urinary leak. The ureter can be transected when it is incorporated with the uterine artery or the vaginal cuff pedicle at the time of hysterectomy or upper vaginectomy. When the ureter is transected without segmental loss and within 5 cm of the bladder, ureteroneocystostomy can be performed (Fig. 3) (28–30). The remaining proximal ureteral stump is ligated with a permanent suture, and the distal portion of the ureter is mobilized so that tension on the anastomotic site can be prevented. An intentional cystostomy is created and

Figure 3 A bladder flap with ureteroneocystostomy. (A) The site of the bladder flap and the ureteroneocystostomy are identified. (B) The ureteroneocystostomy is shown with the ureteral catheter in place and the flap being closed with 3–0 or 4–0 absorbable suture. Source: From Ref. 31.

the distal end of the ureter is tunneled under the submucosa. The ureter is spatulated 0.5 to 1 cm from the end and is secured to the mucosa with 4–0 and 5–0 interrupted absorbable or delayed absorbable suture. On the serosal aspect of the bladder, two 4–0 interrupted delayed absorbable sutures are placed through the seromuscular surface of both the newly reimplanted ureter and the bladder so that the anastomotic tension can be decreased. If the anastomotic site between the ureter and the bladder is under undue tension, a bladder flap can be created, or a psoas hitch can be accomplished by suturing the seromuscular layer of the bladder to the psoas muscle (Fig. 3). A 7- or 8-Fr ureteral stent is passed into the bladder and is removed by cystoscopy after 14 to 21 days.

558

Salom and Penalver

Alternatively, the stent can be passed through the urethra and removed externally. A suction drain is placed in the retroperitoneum, adjacent to the bladder and the ureter, so that any leaks can be detected. If the output from this retroperitoneal drain is high, the fluid should be tested for measurement of blood urea nitrogen and creatinine concentrations. The transurethral or suprapubic catheter can be removed after 7 to 10 days if no leak if present. After the stents have been removed, intravenous pyelography (IVP) is performed; this procedure ensures that there is no extravasation of contrast. If a ureteral injury occurs at the level of the infundibulopelvic ligament, the ureter should be repaired end-to-end by ureteroureterostomy (Fig. 4). The devitalized ends are excised and the end is spatulated for approximately 0.5 cm. A double J stent is passed from the renal pelvis to the bladder. A 4–0 to 5–0 delayed absorbable interrupted suture is placed circumferentially through the seromuscular surface, with care to avoid the mucosa. Finally, a suction drain is placed adjacent to the anastomosis. The stent is removed via cystoscopy, manually, or percutaneously after 14 to 21 days. Other techniques for repairing the ureter when there is segmental loss or when the injury is not adjacent to the ureterovesical junction are bladder flap augmentation, ureteroileocystostomy, and transureterostomy (Table 4).

Gastrointestinal Injury The incidence of iatrogenic injuries to the gastrointestinal tract is low, but such injuries are possible because of the complexity of the surgical procedure and the disease process. The incidence of injury to the bowel during abdominal hysterectomy for a benign condition ranges from 0.3% to 0.5% (16,28–30). Most intestinal injuries occur upon entry into the peritoneal cavity when the bowel is adherent to the anterior abdominal wall, during lysis of adhesions, and during dissection of the rectovaginal space. We recommend preoperative bowel preparation for all patients who have undergone previous surgery or who have pelvic inflammatory disease, endometriosis, suspected malignancy, or gastrointestinal complaints. Such complications can be avoided by extending the skin incision above the previous incision and entering the peritoneum above that point. When dense abdominal adhesions are encountered, the agglutinated serosal interface must be identified. Once the surgeon has outlined the plane of agglutination, the adhesions can be easily separated by sharp and blunt dissection. We advocate sharp dissection when dense inflammatory adhesions have developed as a result of endometriosis, radiation therapy, or malignancy. In such cases, blunt dissection should be avoided because the approximated serosal surfaces lose their plane of demarcation. Small-Bowel Injuries

Figure 4 Ureteral end-to-end anastomosis (ureteroureterostomy). (A) A transected ureter with the spatulated ends. The anastomosis is performed with full-thickness interrupted stitches with a ureteral catheter in place. (B) Completed anastomosis and a suction drain close to the anastomotic site. Source: From Ref. 31.

The management of injuries to the gastrointestinal tract depends on the segment of bowel involved and the type and size of the tear. When a tear is superficial, involving only the serosa and muscularis, a 3–0 interrupted delayed absorbable or permanent suture can be used to close the defect. If the intestinal lumen is perforated, fecal contamination must be minimized by placing a moist laparotomy pack under the leak and occluding the intestinal lumen with proximal and distal linen–shod intestinal clamps. A small full-thickness enterotomy can be repaired in a two-layer technique; the suture line must be maintained perpendicular to the axis of the lumen. The first layer is closed with a 3–0 interrupted absorbable or delayed absorbable suture, which runs the full thickness of the intestinal wall (32). A second 3–0 interrupted permanent imbricating suture is used to incorporate the seromuscular layers. Resection and primary anastomosis are indicated when the repair would constrict the intestinal lumen to less than 2 cm, when the vascular supply to a segment of bowel is sacrificed, or when there are multiple adjacent tears. We prefer to use an automatic stapling device to perform the resection and repair in a side-to-side functional end-to-end technique whenever feasible. Large-Bowel Injuries

Although injury to the colon is less common than injury to the small bowel, colon injuries may be

Chapter 46:

associated with more serious postoperative morbidity because of the increased bacterial colonization found in the large bowel. Dissecting adhesions between the rectum and the vagina, which have resulted from stage IV endometriosis, can be tedious and can result in damage to the rectosigmoid colon. Superficial tears can be repaired as previously outlined. Lacerations that result in full-thickness injuries can be repaired by the two-layer closure technique described above for the repair of small-bowel injuries. Simple injuries near the narrow sigmoid colon should be repaired with a single-thickness closure so that narrowing and postoperative obstruction can be prevented. When lacerations affect more than 40% of the circumference of the large bowel, when there are multiple adjacent injuries, or when the vascular supply to the corresponding segment of bowel is compromised, we recommend resection and primary anastomosis. An automatic stapling device is used for performing a side-to-side functional end-to-end anastomosis; alternatively, a two-layer hand-sewn closure can be used. In the past, small lacerations to the colon were traditionally repaired with a protective colostomy so that wound dehiscence and peritonitis could be prevented. Several prospective and retrospective studies have found that primary repair without a protective colostomy is safe in the absence of fecal contamination (22,33–40). However, creating a protective colostomy is advisable whenever the anastomosis involves an irradiated segment of bowel or when ascites, an infectious process, or fecal contamination with spillage into more than one abdominal quadrant is present (34).

Abdominal Retropubic Procedures for Urinary Stress Incontinence: Burch and Marshal–Marchetti–Krantz Procedures First described in the 1950s, the Marshal–Merchetti– Krantz (MMK) procedure and the Burch procedure are the retropubic vesicourethral suspension techniques most commonly used for treating genuine urinary stress incontinence. Both techniques entail dissection and identification of the retropubic space of Retzius, the Cooper ligament, the pubic symphysis, and the periurethral and paravaginal fascia. A supportive in situ sling is created by suturing the periurethral fascia and the perivaginal fascia under the urethra to either the Cooper ligament or the periosteum of the pubic symphysis. The surgeon must be familiar with the retropubic space so that injury to the engorged venous plexus, urethra, bladder, or ureter can be avoided. The space of Retzius can be developed by using blunt or sharp dissection, but sharp dissection should be used only for patients who have undergone previous surgery and have dense adhesions. The bulb of a transurethral catheter is used to identify the vesicourethral junction. The index and middle finger of the surgeon’s nondominant hand tent the lateral aspects of the vagina to allow placement of two

Complications in Gynecologic Surgery

559

bilateral sutures 2 cm lateral to the midline at the level of the mid-urethra and the vesicourethral junction. The incidence of bladder and urethral injury associated with the MMK procedure reportedly ranges from 0.3% to 0.7% (41). Identification of the bladder margin can be facilitated by filling the bladder with 100 mL of indigo carmine. The incidence of ureteral obstruction ranges from 0.1% to 1.6% (41,42). For this reason, whenever obstruction or damage to the ureter is suspected, we recommend the use of cystoscopy to confirm the efflux of the dye from the ureteral opening. Transurethral or suprapubic catheters draining the bladder should be removed three to seven days postoperatively so that urinary retention can be minimized. Hemorrhage, which occasionally ensues from damage to the venous plexus during dissection of the retropubic space, can be controlled with constant pressure, electrocautery, figure eight sutures, or by simply tying down the sutures previously applied during the Burch or MMK procedure.

VAGINAL SURGERY Patient Positioning and Nerve Injury Improper alignment of a patient in the dorsal lithotomy position during vaginal surgery can result in nerve damage. The femoral, sciatic, and peroneal nerves can be injured by compression, although such injury occurs less frequently during vaginal surgery than during abdominal surgery. Excessive hip flexion, abduction, and external rotation damage the sciatic nerve during 0.4% of cases and the femoral nerve during 4.4% of cases (9,10). Both nerves can be damaged by stretch injuries when hip flexion and leg extension are extended; such extension compresses the femoral nerve under the inguinal ligament and traps the sciatic nerve within the greater sciatic notch. The peroneal nerve can be damaged if the knee is allowed to rest on the side of the stirrups because this positioning causes direct compression of the nerve against the tibia as the nerve courses into the lateral compartment. To avoid such complications, we use lithotomy boots that provide support and limit external rotation of the leg. These boots restrict external rotation, once the long axis of the lower leg has been properly aligned to the contralateral clavicular notch. The thigh should be flexed at an 80 angle to the pelvis, and the knee should be flexed at a 90 to 100 angle to the posterior thigh.

Dilation and Curettage Dilation and curettage (D&C) is a simple and rapid procedure performed routinely across the United States each year for missed, inevitable, and therapeutic abortions and for irregular vaginal bleeding. The procedure is relatively simple but can result in many serious complications such as uterine perforation, hemorrhage, incompetent cervix, intrauterine synechiae, and cervical stenosis. Although D&C is perceived to be a simple procedure, the surgeon should always be vigilant so

560

Salom and Penalver

that the reported rate of 1 complication per 1000 cases and the reported mortality rate of 1 patient per 10,000 cases can be avoided (38). The patient should undergo the preoperative preparation and receive antibiotics appropriate for limiting ascending infections that could result in salpingitis and decreased fertility. The size of the uterus should be gauged by a bimanual examination with the patient under anesthesia; the uterus should then be sounded so that the length of the endometrial cavity can be determined. A discrepancy between the depth of the uterus as measured by the uterine sound and the bimanual examination increases risk of uterine perforation during D&C. The anterior portion of the cervix is grasped with a single-toothed tenaculum, and gentle traction is placed so that the uterine cavity can be aligned in a straight configuration. Such alignment prevents the sound, the dilator, or the curette from perforating the posterior wall of an anteverted uterus or the anterior wall of a retroverted uterus (Fig. 5). The reported incidence of uterine perforation during D&C ranges from 0.6% to 1.3%. The risk of perforation is as high as 2.6% for menopausal patients and as high as 5.1% for patients who experience postpartum bleeding (5.1%) (43). If perforation is suspected to have occurred, the surgeon can observe the patient for signs of perforation or consider diagnostic laparoscopy if there is active bleeding. In cases of anterior perforation, cystoscopy will confirm whether the bladder has been entered. Because the likelihood of perforation is highest when the uterus is gravid or infected, sounding the uterus in these cases is contraindicated. Aggressive dilation may also lead to cervical incompetence and cervical stenosis. Dilation should be no greater than the amount necessary for allowing entry of the smallest curette required for the procedure. Hemorrhage is a rare complication of therapeutic curettage, but its likelihood is higher with a gravid uterus. If bleeding ensues, the surgeon must confirm that the product of conception has been completely evacuated from the cavity or verify that a perforation has occurred laterally at the level of the uterine arteries. If the injury occurs on the lateral wall of the uterine cavity, laparoscopy is necessary for confirming the perforation and controlling bleeding from the uterine artery. When bleeding is believed to have occurred as the result of an atonic uterus rather than a perforation, methergine (methyl ergonovine) should be administered by intramuscular injection. Alternatively, an intravenous infusion of 40 U oxytocin diluted in 1 L of normal saline can be administered. Recent pilot studies have shown that rectal administration of 600 to 1000 mg misoprostol is highly effective in decreasing blood loss by causing myometrial contractions. The terms ‘‘Asherman syndrome’’ and ‘‘synechia’’ refer to intrauterine adhesions that can involve most of the endometrium and lead to amenorrhea and infertility. Aggressive curettage of a gravid or infected

Figure 5 Perforation during dilation and curettage. A curette can perforate either the anterior or the posterior wall of the uterus and consequently injure the bladder, bowel, or rectum, depending on the orientation of the uterus. The uterus is perforated along the posterior wall when in the anteverted position (A) and along the anterior wall when in the retroverted position (B).

Chapter 46:

endometrium has been associated with the development of this syndrome. If oligomenorrhea or amenorrhea develops after curettage, then synechiae should be confirmed hysteroscopically. Treatment consists of adhesionolysis, insertion of an intrauterine device, and estrogen therapy. Intrauterine pregnancies that occur following this procedure require careful monitoring due to the increased risk of uterine perforation.

Conization of the Cervix Cervical dysplasia or carcinoma is diagnosed and treated by removing a cone-shaped biopsy from the uterine cervix with a loop electrocautery excisional procedure, a laser procedure, or a cold-knife conization. The complications associated with this procedure include hemorrhage, ascending infection, cervical stenosis, and cervical incompetence. Patients with high-grade dysplastic cells or carcinoma are predisposed to surgical hemorrhage as a result of angiogenesis and neovascularization. Obstetrical patients are also at a higher risk of intraoperative complications. For this reason, the surgeon must be aware of and able to perform several preventative and therapeutic options. After the patient has been appropriately positioned and prepared, the cervix should be bathed with iodine solution for one minute. The iodine in the solution reacts with glycogen; because normal cells contain more glycogen than dysplastic cells, the normal cells become more darkly pigmented. This coloration clearly demarcates the lighter or hypopigmented dysplastic tissue from the darker normal tissue and allows the surgeon to excise the diseased tissue while avoiding the healthy tissue and the lateral extent of the cervix that contains the vascular supply. Before cold-knife conization is performed, the branches of the cervical artery can be ligated by placing #-0 absorbable or delayed absorbable figure-ofeight sutures into the cervix at 3 o’clock and 9 o’clock approximately 3 cm from the external os. Some gynecologists prefer to inject the cervix with vasopressin or saline circumferentially to provide a medical tourniquet. If bleeding is not controlled by electrocautery after the cone-shaped biopsy material has been excised, a piece of surgicel can be placed into the endocervix and held in place by approximating the lateral stitches that were previously placed. If the bleeding is more substantial, the anterior and posterior transected edges of the cervix can be sutured together with an interrupted #-0 absorbable or delayed absorbable suture. During the early postoperative period, the patient may experience vaginal bleeding as a result of a loose suture and may require further treatment as outlined above. If the patient complains of a foul-smelling discharge, examination through a speculum may reveal a cervical infection that will require bacterial culture and antibiotic therapy. One long-term complication is cervical stenosis, which may lead to infertility or dysmenorrhea. If the cone margins involve the internal

Complications in Gynecologic Surgery

561

os of the cervix, the patient is at a higher risk of premature labor and cervical incompetence. In this scenario, placement of a cerclage may be required before the patient’s next pregnancy.

Total Vaginal Hysterectomy The preoperative assessment of the patient is essential for limiting the incidence of perioperative complications during vaginal hysterectomy. The technical difficulty of vaginal surgery is inherently greater than that of abdominal surgery because of the limited operative field. For this reason, the pubic arch, vaginal introitus, intertuberous diameter, and amount of uterine descensus and mobility must be evaluated preoperatively so that the surgeon can determine the accessibility of the planes of dissection and the vascular pedicles. Contraindications to the vaginal approach to hysterectomy are a history of active pelvic inflammatory disease, endometriosis, large immobile pelvic masses, suspected pelvic malignancy (not including endometrial cancer), chronic pelvic pain that requires abdominal visualization, and a contracted pelvis. Vaginal hysterectomy is associated with all of the complications associated with abdominal hysterectomy such as infection, hemorrhage, and urinary and intestinal complications. However, Meltomaa et al. found that the incidence of bleeding complications requiring transfusions was twice as high during vaginal hysterectomy (2.9%) as during abdominal hysterectomy (1.1%) (28). The pelvic surgeon should not hesitate to convert a vaginal procedure to an abdominal approach when the source of bleeding cannot be identified vaginally. Published reports indicate that the incidence of urinary and intestinal complications is similar for vaginal and abdominal hysterectomy. However because abdominal hysterectomy is used to treat more complicated pathology, this similarity may be misleading, and vaginal hysterectomy may, in fact, be associated with a higher risk of complications. The risk of intraoperative bleeding can be reduced by injecting diluted epinephrine (1:200,000) circumferentially around the cervix. The vasoconstriction caused by injecting 0.25% marcaine with epinephrine (1:200,000) will limit blood loss, increase visibility, and decrease postoperative pain. The hydrodissection produced by the injection may simplify localization and dissection of the surgical planes. Some surgeons believe that injecting marcaine with epinephrine into the pubovesical space facilitates the creation of the bladder flap and thereby decreases the risk of bladder injury. Bladder injury occurs in association with vaginal hysterectomy when the bladder is dissected from the anterior portion of the cervix; the reported incidence ranges from 0% to 1.6% (14,16). We recommend sharp dissection for creating the surgical plane in the vesicouterine space by maintaining constant forceful downward traction on the uterus. Using finger dissection or blunt dissection with surgical gauze for separating the

562

Salom and Penalver

bladder may result in a shearing tear into the bladder lumen, especially when the patient has undergone previous surgery. Injury to the bladder during vaginal hysterectomy occurs in the area of the supratrigonal area of the bladder base. When such an injury is suspected because of the proximity of the trigone, spillage of indigo carmine and ureteral patency can be confirmed by cystoscopy. A simple incidental cystotomy should be repaired with a two-layer closure, and the bladder should be drained with a transurethral or suprapubic catheter for 7 to 10 days (Table 4). Ureteral injury occurs less frequently than bladder injury during vaginal hysterectomy, and the incidence is 0.2% (14). During vaginal hysterectomy, downward traction on the uterus can bring the ureter into close proximity with the uterine artery. The surgical clamp must be applied at a right angle to the arterial insertion and flush to the uterus. Ureteral compromise may also occur during uterosacral ligament suspension for support of the vaginal cuff when uterine prolapse is substantial. This complication occurs when the ureter is in close proximity to the uterosacral ligament as it nears the insertion to the bladder. Before sutures are placed, the uterosacral ligament must be identified and the ureter should be palpated 2 to 4 cm away. If ureteral compromise is suspected, cystoscopy should be performed to verify that indigo carmine is expressed through the ureterovesical orifice. A crushing clamp injury to the ureter can be simply released if the integrity of the lumen is not disrupted and no ischemic damage is noted. When the ureter is incorporated within a surgical pedicle and transected, a ureteroneocystostomy is used to reimplant the ureter, as described above. An unrecognized ureteral obstruction that results from incorporation into a surgical pedicle during vaginal hysterectomy and uterosacral ligament suspension can lead to postoperative unilateral flank pain, a mild increase in the creatinine concentration, and eventual failure of the ipsilateral kidney. A nuclear medicine flow and function study or IVP can be used to diagnose ureteral obstruction. With complete obstruction, we prefer to insert percutaneous nephrostomy tubes and subsequently to perform ureterolysis. If a partial obstruction is detected, a transurethral nephroureteral stent can be inserted and then removed after four to six weeks with a follow-up IVP. If obstruction still persists at that time, we recommend ureterolysis or ureteral reimplantation (Table 4). Unrecognized tears or ischemic injuries to the ureter can lead to fistula formation 3 to 12 days after surgery (22).

Anterior Colporrhaphy Anterior colporrhaphy is commonly used for reconstruction of the anterior wall of the vagina after prolapse of the bladder (cystocele), the urethra (urethrocele), or both into the vagina. Previously, anterior colporrhaphy was used to treat urinary stress incontinence, but it is currently used only in conjunction with other

procedures because the success rate of this procedure alone is only 30% to 40%. Patients with large cystoceles, with advanced vaginal vault prolapse, and who have undergone previous surgery are at higher risk of injury and bleeding. Injecting epinephrine (1:200,000) or 0.25% marcaine with epinephrine (1:200,000) into the submucosa will facilitate the identification of dissection planes and thus decrease blood loss. Furthermore, sharp dissection and proper identification of the avascular vesicovaginal and pubocervical fascia will also decrease blood loss and prevent iatrogenic injuries to the bladder and urethra. When a laceration to the seromuscular layer of the urethra is identified and the mucosa is intact, the laceration should be closed with an interrupted 3–0 or 4–0 absorbable or delayed absorbable suture with a protective imbricating second layer. A fullthickness laceration into the urethral lumen can be closed in two layers. The mucosa is closed with interrupted 3–0 to 4–0 absorbable suture through the mucosa and muscularis (22). The second layer closes the seromuscular tissue in an interrupted fashion and may include the surrounding periurethral tissue of the vesicovaginal fascia. Using transurethral bladder drainage for 7 to 10 days assists in the healing process (Table 4). Some early postoperative complications include infection and fistula formation. In a few cases, fibrosis can result in narrowing of the caliber of the urethral lumen; such narrowing in turn causes urinary retention, which may require prolonged transurethral drainage and urethral dilation. Injuries to the mid-urethra, if not repaired appropriately, can lead to urinary incontinence and fistula formation. After anterior colporrhaphy, transurethral bladder drainage should be continued for three to four days postoperatively; such drainage avoids acute urinary retention, which is fairly common. Before the transurethral catheter is removed, a postvoid measurement of residual urine can be performed for assessing the risk of retention. Such a measurement is performed by filling the bladder with 300 to 400 mL of normal saline, removing the transurethral catheter, and encouraging the patient to void spontaneously. The voided amount is subtracted from the original amount instilled; if less than one-third of the original amount remains in the bladder, replacing the catheter is not necessary.

Posterior Colporrhaphy and Sacrospinous Colpopexy The treatments of choice for the reconstruction of posterior vaginal compartment defects such as rectoceles and enteroceles are posterior colporrhaphy and sacrospinous colpopexy. These procedures should be performed during vaginal hysterectomy or vaginal vault prolapse repair. Patients who have undergone previous posterior repairs and patients with vaginal vault prolapse and enteroceles are at a greater risk of small bowel and rectal injury. Hoffman et al. (39) reported that the rate of rectal injury at the time of

Chapter 46:

Complications in Gynecologic Surgery

563

Figure 6 Vaginal vault prolapse, cystocele, rectocele, and enterocele. (A) Severe vaginal vault prolapse with a third-degree cystocele, rectocele, and enterocele. (B) The anterior vaginal mucosa is incised and the fascia is separated from the underlying peritoneum. (C) When the peritoneum is opened, herniation of the small bowel through the vagina is revealed. (D) The peritoneum is excised and closed. The defect is managed with a sacrospinous ligament fixation and an anterior and posterior colporrhaphy. Finally, the skin is closed with 3–0 absorbable suture.

posterior colpoperineorrhaphy is 0.07%. The incidence of rectal injury at the time of sacrospinous fixation has been reported to range from 0.4% to 4% (39,40). During vaginal hysterectomy, rectal injuries occur during entry into the posterior cul-de-sac, which may be obliterated by fibrosis resulting from endometriosis, inflammatory bowel disease, or diverticulitis. Epinephrine or 0.25% marcaine with epinephrine (1:200,000) is injected as described earlier for anterior colporrhaphy, and the avascular plane between the vaginal mucosa and the rectovaginal fascia is developed. Rectal injuries occur during this portion of the procedure when fibrotic tissue from previous surgery or displacement of the normal anatomy because of severe vault prolapse distorts the surgical planes (Fig. 6). If the mucosa is intact, a single layer of interrupted 2–0 or 3–0 absorbable or delayed absorbable suture is used to imbricate the defect. When a full-thickness injury is found, a two-layer closure is performed. The overlying tissue is then plicated over the area so that the vaginal reconstruction can be completed without the added risk of breakdown and fistula formation. Unrecognized injuries lead to rectovaginal fistula formation.

Infrequent but significant hemorrhagic complications have been reported in association with sacrospinous colpopexy. At the University of Miami, we have found that when a Deschamps needle driver is used, vulvar hematoma occurs in 3% of cases and ischiorectal hematoma occurs in 0.1% of cases during the immediate postoperative period (Fig. 7) (40). If the patient remains hemodynamically normal, most of these complications can be treated conservatively by placing vaginal packing to tamponade the pararectal and ischiorectal spaces while serial hematocrit determinations are performed every 4 hours for 8 to 12 hours. If the vulvar hematoma continues to expand, or if the hematocrit continues to decrease because of bleeding into the retroperitoneum, surgical exploration is necessary. Control of bleeding in this area is difficult because the pudendal artery is encased in bone as it travels through the obturator foramen. One approach is incising the vulvar hematoma and placing tight vaginal packing. If bleeding continues, an exploratory laparotomy can be performed. The hypogastric arteries are ligated bilaterally and a large number of laparotomy packs are

564

Salom and Penalver

guides the positioning of the suburethral prolene mesh. Cystoscopy is performed for verifying proper placement of the trocar. If a perforation is noted, the trocar is simply removed and replaced. In our experience, no sequelae have been associated with conservative management of bladder perforations. After the procedure, the patient is observed for signs of hemorrhage, which can originate from the retropubic venous plexus or the external iliac artery as it enters the inguinal canal. If bleeding is substantial, exploratory laparotomy may be necessary. Another rare complication is urinary retention, which can be managed conservatively with transurethral bladder drainage. If the retention persists, surgical transection of the prolene mesh may be necessary.

HYSTEROSCOPY Hysteroscopy as a Diagnostic and Therapeutic Method Figure 7 Ischiorectal hematoma.

placed into the pelvis; these packs are removed once the hemoglobin and hematocrit levels have been stabilized. Another therapeutic alternative for achieving hemostasis is catheterization and embolization of the pudendal artery.

Sling Procedures for Urinary Incontinence The vaginal approaches to the treatment of urinary stress incontinence are the Pereyra, Stamey, Raz, and Gittes vaginal needle suspensions; the Kelly plication; and the tension-free vaginal taping (TVT) technique. Except for the TVT, which has a cure rate of 80%, these transvaginal urethral suspension techniques have fallen out of favor because of their poor five-year success rates. The TVT is performed by making bilateral incisions on the anterior wall of the vagina under the urethra and sharply dissecting the periurethral space to just under the pubic rami. Intraoperative bleeding and postoperative pain can be minimized by injecting marcaine with epinephrine into the periurethral and suprapubic areas. A sharp trocar is inserted into the periurethral space and passed just under the pubic rami, perforating the anterior fascia and exiting suprapubically. Although this procedure is relatively simple, it can result in a number of perioperative complications, including bladder perforation, urethral perforation, and hemorrhage. Trocar insertion is the crucial portion of the operation and may lead to numerous complications. Precautions for avoiding urethral or bladder perforation include inserting a transurethral catheter with a stylet to deviate the urethra to the side opposite to that of trocar placement. The trocar is passed under the pubic rami and directed 2 to 3 cm lateral to the midline on the anterior abdominal wall; the trocar

Hysteroscopy has allowed advances in the treatment of intrauterine pathology and may be used to treat endometrial polyps, submucosal myomas, dysfunctional uterine bleeding with endometrial ablation, and synechiae with adhesiolysis. The conventional method of treatment, D&C, was a blind attempt at diagnosing and treating intrauterine abnormalities that would usually result in a hysterectomy. Direct visualization of the endometrial cavity has facilitated advances in diagnosis and therapeutic methods for endometrial abnormalities. Fertility is maintained because extirpation of the uterus is avoided. Although hysteroscopy is a relatively simple procedure, the surgeon should be well versed in correcting the intraoperative and postoperative complications that may be associated with it.

Complications During Insertion and Uterine Perforation Laceration to the cervix can occur when traction is applied to the anterior portion of the cervix with a tenaculum while the dilator or the hysteroscope is inserted. Bleeding from such tears will usually stop when pressure is applied, but occasionally a fine 3–0 or 4–0 absorbable suture will be required. Bleeding can also be caused by abrasion to the endocervix; such bleeding can be stopped with electrocautery. The most troublesome complication is hemorrhage resulting from uterine perforation; the incidence of this complication is 0.1% during diagnostic procedures and 1% to 3% during operative hysteroscopy (43–45). Risk factors for such complications include distortion of the uterus, endometrial adhesions, uterine anomalies, hypoplastic uterus, and endometrial cancer. The hysteroscope should be inserted under direct vision with traction on the cervix; insertion should always follow the ‘‘dark spot’’ that represents the lumen. If a perforation occurs in the area of the fundus but no

Chapter 46:

active bleeding ensues, treatment involves monitoring the patient’s vital signs, watching for abdominal discomfort, and performing postoperative determinations of hematocrit. If the perforation occurs in the anterior or lateral portion of the uterus, or if active bleeding ensues, laparoscopy or cystoscopy is mandated.

Uterine Distention The uterine cavity should be distended at a maximum pressure of 60 to 75 mmHg with Hyskon, dextrose solution, sorbitol, glycine, normal saline, or Ringer lactate solution (43). Higher pressures can cause rupture of the uterus and fallopian tubes or excessive absorption of the distention medium. Hyskon has a high viscosity but is no longer routinely used because it has been reported to cause coagulopathy and adult respiratory distress syndrome. Low-viscosity fluids are also associated with complications, depending on the specific properties of each solution. Nonisotonic solutions such as glycine, sorbitol, and dextrose in water may be associated with life-threatening hyponatremia if more than 1000 mL of fluid is absorbed. Isotonic and hypertonic solutions, such as normal saline and Ringer lactate, are associated with volume overload. Both conditions may be treated with diuretics and electrolyte replacement. Fluid balance must be monitored throughout the procedure, and the procedure will occasionally have to be aborted so that such complications can be avoided.

LAPAROSCOPY Introduction Laparoscopy has been a great advance in the treatment of gynecologic diseases because it reduces postoperative pain, morbidity, time off work, and overall cost. Procedures currently performed via this method include laparoscopically assisted vaginal hysterectomy, myomectomy, ovarian cystectomy, tubal ligation, pelvic reconstruction, urinary continence procedures, staging for pelvic malignancy, and lymphadenectomy. The complications associated with laparoscopy in terms of injury to blood vessels, bladder, ureter, and bowel are similar to those associated with open abdominal and vaginal surgery. However, certain injuries are unique to laparoscopy, and their incidence depends on the expertise of the surgeon. Management strategies are similar to those used during open procedures. The surgeon must always be cognizant of the increased technical difficulty involved in the laparoscopic management of iatrogenic injuries and should not hesitate to convert the procedure to laparotomy when the situation demands.

Complications in Gynecologic Surgery

565

injuries is unknown, but the following steps can be taken to prevent such complications. After the skin incision has been made, the anterior abdominal wall can be elevated so that the distance between the fascia and the major vessels can be increased. The Veress needle or trocar should be inserted with the angle of entry directed toward the pelvis or the uterine fundus. If the intra-abdominal pressure produced by the Veress needle exceeds 15 mmHg and the patient’s blood pressure decreases dramatically, or if blood returns from the needle, the surgeon should withdraw the needle and perform a midline incision to explore the abdomen. The inferior epigastric vessels may be lacerated when the lateral lower quadrant ports are inserted, especially when patients are at high risk because of obesity or previous abdominal surgery. Traumatic entry can be avoided by identifying and transilluminating the inferior epigastric vessel with the superior port and placing the trocars at least 4 cm lateral to the midline. If a laceration ensues, the vessel can be cauterized with the contralateral port, the area can be tamponaded with a Foley catheter bulb, or the epigastric vessel can be ligated by placing a conventional suture or a J-needle suture either transabdominally or intra-abdominally (46).

Urinary Tract Injuries The incidence of urinary tract injury during laparoscopy is reported to be 0.5% (47,48). Before the trocar or needle is placed, the bladder is decompressed by the insertion of a transurethral Foley catheter. Transillumination can be used in patients at high risk of adhesions to view and avoid these areas. In general, a Veress needle approach should not be used in these patients and a open technique is employed. However, should a Veress needle puncture occur, it can usually be treated simply by removing the needle. In contrast, the repair of a laceration or trocar injury must be accomplished by a traditional closure. Failure to diagnose an intraoperative bladder injury may result in oliguria, anuria, hematuria, suprapubic pain, and ileus during the patient’s postoperative course. The diagnosis of bladder injury is suggested by infusing 300 mL of normal saline into the bladder and aspirating less fluid. A retrograde cystourethrogram will demonstrate the defect. Ureteral injuries can occur during laparoscopically assisted vaginal hysterectomy, supracervical hysterectomy, or ovarian cyst resection for endometriosis. Complications involving the ureter are treated as described earlier for the abdominal approach. Ureteroureterostomy and ureteroneocystostomy have been performed laparoscopically, but these procedures are technically challenging and should be attempted only by surgeons who are comfortable with performing them.

Vascular Injuries Most injuries to major vessels occur during initial entry with a needle or trocar during the creation of pneumoperitoneum. The true incidence of vascular

Intestinal Injuries The incidence of needle or trocar injury to the small or large bowel ranges from 0.1% to 0.6% (47,49,50).

566

Salom and Penalver

Mechanical injuries to the gastrointestinal tract occur more frequently during adhesiolysis than during trocar or needle entry. The risk of such injuries is greatest for patients who have undergone previous surgery and those with a history of pelvic inflammatory disease, ruptured appendicitis, or pelvic malignancies. Such complications can be prevented by placing the initial port in the left upper quadrant or by beginning the procedure with the open technique. If a Veress needle injury occurs, simply removing the needle is usually sufficient. Electrocautery can cause injuries that require treatment. Any blanched area should always be repaired and imbricated with a 3–0 absorbable or delayed absorbable suture. Superficial and full-thickness lacerations can be repaired laparoscopically, as described above (51). When laparoscopic injuries are undiagnosed, they can lead to postoperative fever, nausea, vomiting, abdominal pain, peritonitis, and sepsis. If such a complication is suspected, exploratory laparotomy is indicated.

RADICAL PELVIC SURGERY FOR GYNECOLOGIC MALIGNANCY Radical Abdominal Hysterectomy, Cytoreductive Surgery for Ovarian Cancer, Pelvic and Para-Aortic Lymph Node Dissection, Pelvic Exenteration, and Urinary Diversion The incidence of complications increases inherently when a gynecologic malignancy is treated because of the aggressive nature of the disease process and the radical procedures used to achieve a cure. Many patients with such malignancies are at high risk of complications because they have undergone multiple previous laparotomies or radiation therapy or because their pelvic anatomy has been distorted by the malignant process (52,53). The gynecologic oncologist routinely performs a wide array of both curative and palliative surgeries for treating these destructive tumors. The traditional treatment for vulvar cancer has been radical vulvectomy. This approach is associated with a high incidence of wound breakdown and infection as a consequence of the en bloc butterfly resection of the entire vulva and inguinal area. Consequently, this procedure has been replaced with localized resection and separate-incision unilateral inguinal lymph node dissection; this newer surgical procedure produces comparable outcomes but is associated with a lower morbidity rate. Late complications of inguinal lymphadenectomy include chronic lymphedema and venous stasis. Radical hysterectomy for the treatment of cervical cancer is associated with higher rates of pelvic hemorrhage, injuries to the ureters, and intestinal injuries than is simple hysterectomy (18,53). Late complications of radical hysterectomy include vaginal prolapse and bladder dysfunction caused by nerve injury during the radical dissection. Resection of retroperitoneal tumors and pelvic exenteration with resection of

the bladder, rectosigmoid, vagina, and perineum may be associated with pelvic hemorrhage, infections, respiratory distress, and presacral bleeding.

Pelvic Hemorrhage One of the most ominous and life-threatening complications encountered during surgery for gynecologic malignancies is intractable pelvic bleeding. Because of the generous amount of blood flow to the pelvic organs, the large vessels in the area, and the complex venous plexus in the retroperitoneal space, a simple vascular injury during resection of a tumor or endometriosis can cause copious bleeding. The extent and location of pelvic bleeding depend on the site of the tumor and the surgical procedure being performed. For example, presacral bleeding can occur during pelvic exenteration for recurrent cervical cancer or during the removal of a retroperitoneal tumor such as a schwannoma. During a cytoreductive or debulking procedure for ovarian cancer, the resection of a tumor that extends into the retroperitoneal space may precipitate a venous hemorrhage that quickly pools and obscures the visual field. Performing para-aortic and pelvic lymphadenectomy for treating cervical cancer can result in catastrophic bleeding from the vena cava or the obturator fossa—the most difficult areas in which to obtain hemostasis. Even pelvic surgery for benign conditions such as sacrocolpopexy can be associated with presacral hemorrhage. The surgeon must be aware of the potential complications and, even more importantly, must be prepared to execute the proper life-saving procedures. In most cases, pelvic bleeding can be controlled by routine maneuvers such as applying firm pressure or hemoclips, cauterization, and suture ligation. When such maneuvers are not sufficient for controlling hemorrhage, the surgeon must be able to perform certain intraoperative alternatives that will allow hemodynamic stabilization. During such an event, the surgeon must always inform the anesthesiologist about the circumstances surrounding the blood loss, so that fluid resuscitation, blood transfusions, and the administration of blood products can be planned. In the following sections, we detail some of the techniques performed at our institution for treating catastrophic bleeding, including bilateral ligation of the hypogastric artery, mass suture, presacral thumbtack placement, pelvic packing, and arterial embolization.

Bilateral Ligation of the Hypogastric Artery Bilateral ligation of the hypogastric artery minimizes the hemorrhaging and generalized oozing that occur after complicated radical hysterectomy, radical vulvectomy, ovarian cancer debulking, complicated gynecologic conditions, and even obstetrical operations. Ligation of the internal hypogastric artery controls hemorrhagic events by reducing the pelvic blood pressure by 24% and the blood flow by 50%;

Chapter 46:

this reduction gives the normal coagulation system an opportunity to work (54). Minimal sequelae have been reported after bilateral ligation of the internal hypogastric artery, and full-term deliveries have even been achieved. However, this procedure is associated with known hazards. Intraoperative complications have been reported to occur in as many as 15.7% of cases; these complications include ureteral injuries and injuries to the internal iliac vein (55). Postoperative complications include fistula formation, gluteal cramping, and ischemia as the result of inadequate perfusion. The abdominal approach for the ligation of the hypogastric artery is accomplished by entering

Complications in Gynecologic Surgery

567

the retroperitoneal space through an incision in the posterior parietal peritoneum over the common iliac artery. After dissecting and reflecting the peritoneum, the surgeon identifies the hypogastric artery at its bifurcation by clearing away the surrounding connective areolar tissue. The ureter should always be identified and moved away from the iliac vessels. The tips of a long right-angle clamp are passed under the internal iliac artery (the hypogastric artery) on the lateral aspect of the vessel 2 cm distal to the bifurcation of the common iliac artery, so that ligation of the posterior branch of the internal iliac artery can be avoided (Fig. 8). After the long right-angle clamp is placed under the vessel to be ligated, it is tilted and pushed from lateral to medial so that the tips are visible on the medial side of the vessel. Next, a 0-silk suture is used to tie the vessel, and a second silk suture is placed 0.5 to 1 cm caudal to the first. The same procedure is performed on the contralateral side. The pulses of the external and femoral arteries should be palpated before the suture is secured so that the patency of the external iliac vessel can be verified and the surgeon can be confident that this vessel has not been ligated in error. If ligation of the internal iliac artery fails to control the bleeding, the ovarian artery can also be ligated before the surgeon commits to performing a hysterectomy (56).

Mass Suture

Figure 8 Bilateral hypogastric artery ligation. (A) A right-angle clamp is placed under the hypogastric artery 2 cm from the bifurcation of the common iliac artery. The surgeon standing on the contralateral side passes the instrument from lateral to medial to avoid injury to the hypogastric vein. (B) Two #-0 silk sutures are placed proximally and distally from the 2-cm mark and secured.

The extensive venous plexus that drains the pelvic floor can be damaged during lymph node dissection or resection of a retroperitoneal tumor. Because the source of bleeding is rarely a single vessel, it is usually futile to attempt to locate a single bleeding site and place a hemoclip or suture. Clamping and ligating the bleeding venous plexus is difficult and may require placement of a mass suture. Before proceeding, the surgeon must identify the adjacent pelvic anatomy, including the ureter, bladder, rectum, major vessels, and nerves, so that any injury to these structures can be avoided. At this point, a #-0 absorbable or delayed absorbable suture on a CT-1 needle is placed as a large figure-of-eight that incorporates the soft tissue around the bleeding site; this suture is tied firmly. If space is limited, a CT-2 needle can be chosen. Multiple mass sutures are often necessary for controlling the bleeding. If bleeding continues after the mass sutures have been placed, administering pressure with warm disposable lap sponges should be attempted for a few minutes. In addition, the surgeon can consider the use of thrombostatic agents. Finally, packing the pelvis can be the final attempt at stabilizing the patient so that coagulopathy can be corrected and further resuscitation can be performed.

Thumbtack for Presacral Bleeding Pelvic exenteration, abdominal sacral colpopexy, and resection of retroperitoneal malignancies can cause

568

Salom and Penalver

CASE STUDY A 39-year-old woman with an intrauterine pregnancy at 36 weeks gestation came to the hospital for an elective cesarean delivery. She was found to have multiple large uterine fibroids and a complete placenta previa, a condition in which the placenta overlies the birth canal and predisposes the patient to third-trimester bleeding. During her pregnancy, the patient had been admitted to the hospital twice for vaginal spotting and degenerating fibroids. Before the operative delivery was scheduled, fetal lung maturity was confirmed by amniocentesis. A healthy baby girl was delivered by cesarean section. Immediately after delivery, the patient experienced a substantial uterine hemorrhage from the abnormally attached placental tissue. At this time, an emergency hysterectomy was begun; packed red blood cells and fresh frozen plasma were administered to correct clinically evident disseminated intravascular coagulopathy and a 3-L blood loss. The patient was transferred to the surgical intensive care unit, where her condition stabilized. The patient was discharged from the hospital on postoperative day 10. Three weeks after discharge, the patient complained of urinary incontinence that was continuous but increased with the Valsalva maneuver. An examination through a speculum showed a clear watery discharge with no visible mucosal defects. The patient underwent specialized urodynamic studies as part of the work-up for urinary incontinence; the results were negative. When urinary incontinence continued, voiding cystourethrography was performed; the results were normal. IVP showed extravasation of contrast into the vagina but no clear source of the leak (Fig. 9). Computed tomography of the pelvis confirmed a persistent ureterovaginal fistula causing the symptoms of urinary incontinence (Fig. 10). A nephroureteral stent was placed for bypassing the fistula tract and allowing spontaneous healing. After 12 weeks of conservative management with no resolution, we performed ureteroneocystostomy. The patient remains free of symptoms.

laceration of the presacral vessels (57). Once these presacral vessels have been damaged, they retract and recoil into the bony surfaces of the sacrum; this retraction renders the conventional methods of controlling hemorrhage ineffective. Occasionally, applying pressure or temporarily packing the pelvis can achieve hemostasis, but once the packing is removed, bleeding may resume and blood can rapidly pool in the pelvis. Attempts at applying mass sutures onto the bony structures of the sacrum are usually futile and risk further vascular injury. When uncontrollable bleeding occurs in this area, a stainless steel thumbtack on the tips of a long Kelly or Kocher clamp can be applied to the source of the hemorrhage. Timmons et al. (58) used a 12-inch stainless steel rod with a recessed magnet at the end for securing and delivering the thumbtack.

experience of the surgeon. We place warm laparotomy packs, starting from the deepest portion of the pelvis, until enough tension has been obtained to maintain constant compression on the bleeding site.

Pelvic Packing Another method of achieving hemostasis and stabilization for patients with extensive and intractable bleeding from raw surfaces, venous plexuses, and inaccessible areas is the prolonged use of abdominal and pelvic laparotomy packing. Various techniques are used for packing the pelvis, depending on the

Figure 9 Intravenous pyelography demonstrates extravasation of the intravenous contrast material into the vagina.

Chapter 46:

Complications in Gynecologic Surgery

569

so that intra-abdominal bleeding can be monitored and unnecessary wound closure can be avoided. Determining whether hemostasis has been achieved by placing a Jackson–Pratt drain in the area; however, this drain may not allow accurate monitoring of ongoing hemorrhage. The packing should be removed after 24 to 48 hours, provided the patient’s hemodynamic condition is stable and the bleeding has stopped. To remove the packing, we reopen the abdomen to allow for proper visualization of the pelvic cavity and the bowel, thereby avoiding shearing tears and perforation.

SUMMARY The complexity of the female pelvic anatomy, the small field of vision, and the distortion created by disease processes all contribute to the challenges faced by the pelvic surgeon. Innovations such as laparoscopy have advanced the quality of medical care provided in this field. However, careful preoperative preparation, surgical expertise, and familiarity with the prevention, recognition, and therapy of potential complications in this area are needed to achieve the best outcomes.

REFERENCES

Figure 10 Computed tomography of the pelvis. (A) The contrast material within the fistula is seen tracking toward the vagina posterior to the ureter. (B) The fistula is seen entering and filling the vagina.

The number of laparotomy packs used depends on the size of the patient and the depth of the pelvis. Next, we close the fascia, but intra-abdominal pressures must be closely monitored for compartment syndrome. The skin incision is left open and packed

1. Mowat J, Bonnar J. Abdominal wound dehiscence after cesarean delivery. Br Med J 1971; 2:256–257. 2. Ellis H, Coleridge-Smith PD, Joyce AD. Abdominal incisions—vertical or transverse? Postgrad Med J 1984; 60:407–410. 3. Stone HH, Hoefling SJ, Strom PR, Dunlop WE, Fabian TC. Abdominal incisions: transverse vs vertical placement and continuous vs interrupted closure. South Med J 1983; 76:1106–1108. 4. Guillon PJ, Hall TJ, Donaldson DR, Broughton AC, Brennan TG. Vertical abdominal incisions—a choice? Br J Surg 1980; 67:395–399. 5. Becker JM, Stucchi AF. Intra-abdominal adhesion prevention: are we getting any closer? Ann Surg 2004; 240:202–204. 6. Cherney LS. A modified transverse incision for low abdominal operations. Surg Gynecol Obstet 1941; 72:92–95. 7. Kvist-Poulsen H, Borel J. Iatrogenic femoral neuropathy subsequent to abdominal hysterectomy: incidence and prevention. Obstet Gynecol 1982; 60:516–520. 8. Goldman JA, Feldberg D, Dicker D, Samuel N, Dekel A. Femoral neuropathy subsequent to abdominal hysterectomy. A comparative study. Eur J Obstet Gynecol Reprod Biol 1985; 20:385–392. 9. Burkhart F, Daly JW. Sciatic and peroneal nerve injury: a complication of vaginal operations. Obstet Gynecol 1966; 28:99–102. 10. McQuarrie HG, Harris JW, Ellsworth HS, Stone RA, Anderson AE III. Sciatic neuropathy complicating vaginal hysterectomy. Am J Obset Gynecol 1972; 113:223–232. 11. Morey SS. ACOG issues report on management of operative injuries of the urinary tract. Am Fam Physician 1998; 57:870.

570

Salom and Penalver

12. Benson RC, Hinman F Jr. Urinary tract injuries in obstetrics and gynecology. Am J Obstet Gynecol 1955; 70: 467–485. 13. Schmidt JD. Management of urinary tract injuries. In: Buchsbaum HJ, Schmidt JD, eds. Gynecologic and Obstetric Urology. 3rd ed. Philadelphia: WB Saunders, 1993:155–162. 14. Harkki-Siren P, Sjoberg J, Tiitinen A. Urinary tract injuries after hysterectomy. Obstet Gynecol 1998; 92: 113–118. 15. Johnson N, Barlow D, Lethaby A, Tavender E, Curr L, Garry R. Methods of hysterectomy: systematic review and meta-analysis of randomized controlled trials. BMJ 2005; 330:1478–1486. 16. Dicker RC, Greenspan JR, Strauss LT, et al. Complications of abdominal and vaginal hysterectomy among women of reproductive age in the United States. The Collaborative Review of Sterilization. Am J Obstet Gynecol 1982; 144:841–848. 17. Takamizawa S, Minakami H, Usui R, et al. Risk of complications and uterine malignancies in women undergoing hysterectomy for presumed benign leiomyomas. Gynecol Obstet Invest 1999; 48:193–196. 18. Averette HE, Nguyen HN, Donato DM, et al. Radical hysterectomy for invasive cervical cancer. A 25-year prospective experience with the Miami technique. Cancer 1993; 71:1422–1437. 19. Christensen A, Foglmann R. Cervical carcinoma stage I and II treated by primary radical hysterectomy and pelvic lymphadenectomy, 320 cases by the method of Meigs-Taussig and 350 by the method of Okabayaschi. Acta Obstet Gynecol Scand Suppl 1976; 58:1–44. 20. Rampone JF, Klem V, Kolstad P. Combined treatment of stage Ib carcinoma of the cervix. Obstet Gynecol 1973; 41:163–167. 21. Mikuta JJ, Giuntoli RL, Rubin EL, Mangan CE. The ‘‘problem’’ radical hysterectomy. Am J Obstet Gynecol 1977; 128:119–127. 22. Holley RL, Kilgore LC. Urologic complications. In: Orr JW Jr., Shingleton HM, eds. Complications in Gynecologic Surgery: Prevention, Recognition and Management. Philadelphia: Lippincott, 1994:131–166. 23. Angioli R, Penalver M. Urinary tract injuries. In: Hurt WG, ed. Urogynecologic Surgery. 2nd ed. Philadelphia: Lippincott-Raven, 2000:177–186. 24. Dowling RA, Corriere JN Jr., Sandler CM. Iatrogenic ureteral injury. J Urol 1986; 135:912–915. 25. Dorairajan G, Rani PR, Habeebullah S, Dorairajan LN. Urologic injuries during hysterectomies: a 6-year review. J Obstet Gynecol Res 2004; 30:430–435. 26. Kuno K, Menzin A, Kauder HH, Sison C, Gal D. Prophylactic ureteral catheterization in gynecologic surgery. Urology 1998; 52:1004–1008. 27. Lafferty HW, Angioli R, Rudolph J, Penalver MA. Ovarian remnant syndrome: experience at Jackson Memorial Hospital, University of Miami, 1985 through 1993. Am J Obstet Gynecol 1996; 174:641–645. 28. Meltomaa SS, Makinen JI, Taalikka MO, Helenius HY. One-year cohort of abdominal, vaginal, and laparoscopic hysterectomies: complications and subjective outcomes. J Am Coll Surg 1999; 189:389–396. 29. Mattlingly RF, Borkowf HI. Acute operative injury to the lower urinary tract. Clin Obstet Gynaecol 1978; 5: 123–149.

30. Fry DE, Milholen L, Harbrecht PJ. Iatrogenic ureteral injury. Options in management. Arch Surg 1983; 118: 454–457. 31. Angioli R, Penalver M. Ureteral injury at the time of radical pelvic surgery. Op Techniques Gynecol Surg 1998; 3:132–140. 32. Burke TW, Levenback C. Gastrointestinal tract. In: Orr JW Jr., Shingleton HM, eds. Complications in Gynecologic Surgery: Prevention, Recognition and Management. Philadelphia: Lippincott, 1994:103–130. 33. Sasaki LS, Allaben RD, Golwala R, Mittal VK. Primary repair of colon injuries: a prospective randomized study. J Trauma 1995; 39:895–901. 34. Gonzalez RP, Merlotti GJ, Holevar MR. Colostomy in penetrating colon injury: is it necessary? J Trauma 1996; 41:271–275. 35. Frame SB, Ridgeway CA, Rice JC, McSwain NE Jr., Kerstein MD. Penetrating injuries to the colon: analysis by anatomic region of injury. South Med J 1989; 82:1099–1102. 36. George SM Jr., Fabian TC, Mangiante EC. Colon trauma: further support for primary repair. Am J Surg 1988; 156:16–20. 37. Murray JJ, Schoetz DJ Jr., Coller JA, Roberts PL, Veidenheimer MC. Intraoperative colonic lavage and primary anastomosis in nonelective colon resection. Dis Colon Rectum 1991; 34:527–531. 38. Houston MC, Ratcliff DG, Hays JT, Gluck FW. Preoperative medical consultation and evaluation of surgical risk. South Med J 1987; 80:1386–1397. 39. Hoffman MS, Lynch C, Lockhart J, Knapp R. Injury of the rectum during vaginal surgery. Am J Obset Gynecol 1999; 181:274–277. 40. Salom EM, Penalver MA. Pelvic extenteration and reconstruction. Cancer J 2003; 9:415–424. 41. Mainprize TC, Drutz HP. The Marshall–Marchetti– Krantz procedure: a critical review. Obstet Gynecol Surv 1988; 43:724–729. 42. Maulik TG. Kinked ureter with unilateral obstructive uropathy complicating Burch colposuspension. J Urol 1983; 130:135. 43. Shirk GJ. Uterine perforation: endoscopic complications and treatment. In: Corfman RS, Diamond MP, DeCherney AH, eds. Complications of Laparoscopy and Hysteroscopy. Boston: Blackwell, 1997:221–225. 44. Lindelmann HJ, Mohr J. CO2-hysteroscopy: diagnosis and treatment. Am J Obstet Gynecol 1976; 124:129–133. 45. Hulka JF, Peterson HB, Phillips JM, Surrey MW. Operative hysteroscopy. American Association of Gynecologic Laparoscopists 1991 membership survey. J Reprod Med 1993; 38:572–573. 46. John DA. Perforation of the inferior epigastric vessels. In: Corfman RS, Diamond MP, DeCherney AH, eds. Complications of Laparoscopy and Hysteroscopy. Boston: Blackwell, 1997:30–35. 47. Loffer FD, Pent D. Indications, contraindications and complications of laparoscopy. Obstet Gynecol Surv 1975; 30:407–427. 48. Schanbacher PD, Rossi LJ Jr., Salem MR, Joseph NJ. Detection of urinary bladder perforation during laparoscopy by distention of the collection bag with carbon dioxide. Anesthesiology 1994; 80:680–681. 49. Krebs HB. Intestinal injury in gynecologic surgery: a ten-year experience. Am J Obstet Gynecol 1986; 155: 509–514.

Chapter 46:

50. Peterson HB, Hulka JF, Phillips JM. American Association of Gynecologic Laparoscopists’ 1988 membership survey on operative laparoscopy. J Reprod Med 1990; 35:587–589. 51. Nezhat CR, Nezhat FR, Nezhat C, Luciano AA, Siegler AM, Metzger DA. Complications. In: Siegler A, Nezhat FR, Nezhat C, Siedman DS, Luciano AA, Nezhat CR, eds. Operative Gynecologic Laparoscopy: Principles and Techniques. 2nd ed. New York: McGraw-Hill, 2000:287–311. 52. Morley GW, Seski JC. Radical pelvic surgery versus radiation therapy for stage I carcinoma of the cervix (exclusive of microinvasion). Am J Obstet Gynecol 1976; 126:785–798. 53. Benedetti-Panici P, Scambia G, Baiocchi G, Maneschi F, Greggi S, Mancuso S. Radical hysterectomy: a randomized study comparing two techniques for resection of the cardinal ligament. Gynecol Oncol 1993; 50:226–231.

Complications in Gynecologic Surgery

571

54. Burchell RC. Physiology of internal iliac artery ligation. J Obstet Gynaecol Br Commonw 1968; 75: 642–651. 55. Chattopadhyay SK, Deb Roy B, Edrees YB. Surgical control of obstetric hemorrhage: hypogastric artery ligation or hysterectomy? Int J Gynaecol Obstet 1990; 32:345–351. 56. Fehrman H. Surgical management of life-threatening obstetric and gynecologic hemorrhage. Acta Obstet Gynecol Scand 1988; 67:125–128. 57. Sutton GP, Addison WA, Livengood CH III, Hammond CB. Life-threatening hemorrhage complicating sacral colpopexy. Am J Obstet Gynecol 1981; 140: 836–837. 58. Timmons MC, Kohler MF, Addison WA. Thumbtack use for control of presacral bleeding, with description of an instrument for thumbtack application. Obstet Gynecol 1991; 78:313–315.

47 Complications of Bladder and Prostate Surgery Adam J. Bell, Josh M. Randall, and Raymond J. Leveillee Division of Endourology and Laparoscopy, Department of Urology, University of Miami Miller School of Medicine, Miami, Florida, U.S.A.

Reports of surgical procedures involving the bladder date back to 200 B.C., when Ammonius wrote about perineal lithotomy for bladder stone removal. Since that time, a wide range of additional indications for bladder surgery have developed. Malignancy is the most common reason for surgical intervention involving the bladder. The most common urinary tract malignancy derived from the urothelial lining of the urinary system is bladder cancer in which transitional cell carcinoma accounts for approximately 90% of these tumors. In the United States, bladder cancer is the fourth leading cause of cancer death for men, after lung, prostate, and colon cancer. The remaining bladder tumors are accounted for by squamous cell carcinoma (3–7%), adenocarcinoma (2–3%), and metastatic lesions (1–2%) (1). This chapter presents surgical treatment options for bladder cancer, as well as surgical intervention for other disease processes, such as neurogenic bladder, urinary incontinence, genitourinary anomalies, infections, intractable hematuria, ureteral and renal pathology, and trauma. To treat these conditions, surgeons may use a wide range of operative (and nonoperative) techniques, tools, and talents, but they must also be able to deal with a multitude of potential surgical complications. This chapter discusses the diagnosis, treatment, and prevention of complications associated with bladder surgery.

BLADDER CANCER Transurethral Resection of Bladder Tumor Endoscopic management of genitourinary pathology has become a cornerstone of urologic surgery (2). Cystoscopy allows for complete evaluation of the urethra, prostate, and bladder, and it is used as both a diagnostic and a therapeutic procedure. For example, cystourethroscopy is performed as part of the initial workup for hematuria; if a bladder tumor is discovered, it may be resected transurethrally. Approximately 55% to 60% of cases of transitional cell carcinoma (TCC) are diagnosed when they are still considered superficial disease, i.e., tumor confined to the mucosa (stage Ta or Cis) or the submucosa (stage T1) (1). Initial management requires performing transurethral resection of

the bladder tumor (TURBT) so that a good amount of tissue can be obtained in order to make a diagnosis of cancer as well as stage the patient appropriately. This technique also adequately provides hemostasis and assesses the uninvolved urothelium and bladder tissue (2). The rate of recurrence of these superficial bladder tumors is approximately 75%, but only 10% to 15% will progress to muscle-invasive disease (1). The technique of transurethral resection has been well described and essentially involves performing initial endoscopic inspection of the bladder and noting the area(s) of suspected pathology, resecting the bladder tumor, removing the specimen, and obtaining hemostasis (2). Although infrequent, complications may occur in association with transurethral resection; these complications can be characterized as major or minor and as early or late. One potential early complication is postoperative bleeding. Although some degree of hematuria can be expected after transurethral resection, failure to achieve sufficient hemostasis after the procedure can result in postoperative gross hematuria, clot retention, and acute anemia. Initial treatment involves placing a three-way Foley catheter, irrigating the bladder by hand, removing the blood clot, initiating continuous bladder irrigation with close observation, and administering blood transfusions when indicated. A second surgical procedure may be required if bleeding does not subside. This complication can be prevented by meticulous electrosurgical hemostasis at the conclusion of the resection. Free perforation through the bladder wall during resection of the bladder tumor is a worrisome and serious complication. The risk of perforation can be reduced by performing precise, controlled ‘‘swipes’’ of the tumor with the resectoscope loop, noting the depth of resection each time and limiting the bladder inflow (the bladder wall thins as it becomes progressively fuller). When perforation does occur, it may be either extraperitoneal or intraperitoneal. Most small extraperitoneal bladder perforations can be managed expectantly by placing a Foley catheter for 7 to 10 days so that complete healing can occur; a gravity cystogram is performed before the catheter

574

Bell et al.

is removed so that healing can be demonstrated (2). Substantial extravasation of perivesical fluid and suprapubic fullness may occur with large defects and can be treated by placing a drain through a small suprapubic incision. Small intraperitoneal bladder perforations can also be managed conservatively, but if the patient experiences any signs or symptoms of acute peritonitis, open surgical exploration and closure are indicated; this is because intraperitoneal bladder perforation indicates that the bowel is at risk of injury and delayed perforation as the result of thermal necrosis (2). The tumor may recur locally or distantly after bladder perforation, but the risk is relatively low. In one report, metastatic disease developed after intraperitoneal bladder perforation in only 1 of 16 patients with extravesical tumor recurrence (3).

Vesicoureteral reflux (VUR) or stenosis may occur after tumor resection over or near the ureteral orifice. This risk can be minimized by limited use of electrical coagulation in this area. As is the case with any transurethral manipulation, iatrogenic injury to the urethra and bladder neck can result in either acute or chronic urethral strictures. Proper lubrication and direct visual insertion of instruments is crucial. A situation unique to TURBT is the obturator reflex (Fig. 1). This reflex most commonly occurs during resection of a lateral wall tumor: electrical stimulation of the obturator nerve results in a sudden abduction of the leg. Bladder perforation and neurovascular injury have resulted from the obturator reflex. One strategy for prevention is neuromuscular paralysis with general anesthesia.

Figure 1 Pelvic anatomy. Source: From Netter, Frank H. Atlas of Human Anatomy plate 385.

Chapter 47:

Lasers have been used for some time as an alternative treatment for superficial bladder tumors, usually those less than 2 cm in diameter. The most frequently used lasers [neodymium þ yttrium–aluminum–garnet (Nd:YAG) and holmium:YAG] offer several advantages: less bleeding, subjectively less pain, lower incidence of bladder spasms and dysuria, absence of obturator nerve stimulation, reduced need for Foley catheter, and substantially lower risk of bladder perforation (4). Specific laser training and safety issues must be addressed, and the lack of pathological tissue for review must be recognized. Because of the ‘‘forward scatter’’ effects of the Nd:YAG laser, the risk of bowel perforation exists, even when the bladder is apparently intact (5). If bowel perforation occurs, the signs and symptoms of peritonitis, such as free air under the diaphragm and pain, usually begin within the first 24 hours postoperatively. Ultimately, patients require laparotomy with resection of the damaged bowel segment. The risk of bowel injury can be reduced by avoiding excessive application of the laser energy in any given area and avoiding overdistention of the bladder, which thins the bladder wall. The risk of postoperative urinary tract infection (UTI) or urosepsis is reduced by ensuring negative preoperative urine cultures and perioperative antibiotic therapy. Rapid recognition of postoperative urosepsis before hospital discharge will allow treatment with appropriate empiric intravenous antibiotic therapy, intravenous fluid hydration, and definitive therapy based on culture and sensitivity results.

Radical Cystoprostatectomy Invasive bladder cancer refers to tumor penetration through the lamina propria and into the muscularis propria (muscle layer of the bladder), with or without extension into the perivesical soft tissue (1). Most clinical urologists agree that bilateral pelvic lymph node dissection together with en bloc single-stage radical cystectomy and the creation of a continent or cutaneous urinary diversion is the treatment of choice for muscle-invasive bladder cancer (6). Exenterative pelvic surgery is rarely required. Alternative treatments exist for selected patients with locally advanced bladder cancer, such as various combinations of deep TURBT (7), partial cystectomy (8), radiation therapy (1), and chemotherapy (1). Although the survival rates associated with these alternative procedures may be lower than those associated with radical cystectomy, such treatments may be an alternative for high-risk patients or those who, for other reasons, are poor surgical candidates; these treatments may also be palliative (6,9,10). During the past few decades, improvements in anesthesia, antibacterial agents, and surgical techniques have dramatically decreased the morbidity and 30-day mortality rates associated with radical cystectomy. The morbidity rates have declined from approximately 35% to less than 10%, and the 30-day mortality rates have declined from

Complications of Bladder and Prostate Surgery

575

approximately 20% to less than 2% (11). Several reports have classified the surgical complications associated with radical cystectomy as acute or chronic and as minor or major (10–16). Most muscle-invasive bladder cancers are diagnosed after endoscopic tumor resection. Imperative to the success of surgery for this type of cancer is the preoperative education of the patient about the type of urinary diversion that will be created. After a detailed review of all pathologic specimens and indepth patient counseling, a metastatic evaluation is performed. This evaluation includes assessment of the upper urinary tract, urethral sampling, chest radiographs, renal and hepatic function examinations, and determination of cardiopulmonary risk status (12). Other important considerations are the patient’s history of radiation exposure, other abdominal surgery, conditions such as inflammatory bowel disease and diverticulitis, hand–eye coordination, and manual dexterity. The stoma site is marked preoperatively in conjunction with an enterostomal nurse, and preoperative bowel preparation and antibiotic coverage are undertaken. Briefly, the following general steps are involved in radical cystoprostatectomy: (i) laparotomy and exclusion of grossly metastatic intraperitoneal disease; (ii) bilateral pelvic–iliac lymphadenectomy; (iii) ligation and division of the anterolateral and dorsolateral vesical pedicles with subsequent ligation of Santorini’s plexus; (iv) ureteral mobilization, tagging, and transection at the level of the iliac vessels; (v) incision of the peritoneum of the rectovesical pouch and blunt mobilization of the bladder posteriorly between Denonvillier’s fascia and the rectum; (vi) ligation and transection of the dorsomedial pedicles; (vii) transection of Santorini’s plexus and the urethra just distal to the prostate; and (viii) retrograde mobilization of the prostate with preservation of the neurovascular bundles, followed by excision of the gland (12). Routine removal of the appendix is associated with low morbidity rates and avoids future appendicitis but precludes the use of the appendix in any future reconstructive surgery that may be required (17). A urethrectomy is indicated at the time of cystectomy if prostatic urethral biopsies indicate TCC or if gross carcinoma is detected at the prostatic urethral margin during excision; however, urethrectomy precludes construction of a neobladder (6,12,18,19,. A delayed urethrectomy may be performed after initial cystectomy and urinary diversion if urethral recurrence is discovered during routine follow-up. In addition, there is a 1% to 9% risk of upper-tract TCC after cystectomy; therefore, routine follow-up is encouraged (20). Either a cutaneous (e.g., ileal conduit) or a continent (e.g., neobladder) urinary diversion is performed. Additional maneuvers may include placing temporary stents across the ureteral anastomosis, inserting a closed suction drain, and performing appropriate urinary diversion drainage.

576

Bell et al.

Urinary Diversion The popularity of the various methods of urinary diversion has varied throughout the last century. An ideal reservoir should preserve upper urinary tract function by avoiding reflux and obstruction; allow high capacity with low pressure, thereby providing adequate compliance; and provide a continence mechanism that will allow achievable continence yet ease of emptying (Table 1) (21). The oldest form of urinary diversion is the ureterosigmoidostomy; in this procedure, the ureters are anastomosed directly to the sigmoid colon (22). Before the antibiotic era, free reflux of urine contaminated with stool resulted in substantial morbidity and mortality rates related to repeated bouts of pyelonephritis and upper-tract deterioration (21). In response to this complication, antireflux procedures such as the Mainz II pouch were developed (23). Ureterosigmoidostomy is further complicated by a hyperchloremic metabolic acidosis, which is amplified by renal failure, and by an increased rate of colonic neoplasms (5–40%), especially among younger patients who require diversion for benign conditions and who have a long life expectancy (21). Although many of these young patients have already undergone rediversion, the risk of malignancy at the anastomotic site remains and lifelong surveillance is required (24–27). Another form of urinary diversion is continent or incontinent cutaneous diversion. The ileal conduit popularized by Bricker anastomoses the ureters to an isolated segment of mid to distal ileum and then brings them through the abdominal wall as a stoma (28). An external appliance is required for collecting freely draining urine. A continent cutaneous diversion, in contrast, uses an isolated detubularized segment of bowel, usually ileum with or without cecum, to store adequate amounts of urine (typically

600–700 cc) at low pressures (< 25 cmH2O). This storage pouch (detubularized segment of bowel) prevents reflux; it is drained through a catheterizable continent stoma brought out to the skin. Many variations of this procedure use different segments of bowel to construct different stoma mechanisms, such as the Kock pouch (29), the Indiana pouch (30), the Mainz pouch (21,31), and the University of California Los Angeles (UCLA) right colonic pouch (21). Examples of such reservoirs are shown in Figure 2. The overall rate of complications associated with ‘‘pouch’’ reservoirs is estimated at 10% to 35%; the rates of reoperation range from 10% to 20% (21,31–34). Corrective surgery may be indicated for urinary leakage through the efferent limb, difficulty in catheterizing the stoma, parastomal hernia, or afferent limb problems, such as stenosis or reflux leading to hydronephrosis and compromised renal function (34). The surgeon contemplating reoperation must have a thorough knowledge of the type of diversion and the segment of bowel used, so that damage to the diversion itself and inadvertent enterotomies can be avoided. Proper construction of the intussuscepted nipple valve and fixation of the distal efferent segment to the abdominal wall are crucial to preventing some of these complications (35). Removing a portion of the mesenteric fat before bringing the ileum through the stoma site may reduce the risk of parastomal hernia (32,34). Other common complications are incontinence (most commonly nocturnal), stones in the pouch,

Table 1 Long-term complications in 198 of 675 patients treated with cystectomy and urinary diversion Complication

No. (%)

Small bowel obstruction Ureteroenteric stricture Renal calculi Acute pyelonephritis Parastomal hernia Stomal stenosis Incisional bernia Fistula Colobic obstruction Radiation proctitis Hepatitis Chronic renal failure Pelvic abscess Abdomincal wall abscess Stomal prolapse Other

50 47 26 21 19 19 15 9 8 6 3 2 2 2 2 6

(7.4) (7.0) (3.9) (3.1) (2.8) (2.8) (2.2) (1.8) (1.2) (0.9) (0.4) (0.3) (0.3) (0.3) (0.3) (0.8)

Note: Total number of patients with one or more long-term complications 198 (29.3%). Source: From Ref. 21.

Figure 2 Indiana pouch. Source: From Ref. 10.

Chapter 47:

metabolic acidosis, and chronic bacteruria leading to ‘‘pouchitis.’’ The risk of pouch stone formation can be reduced by eliminating staples from the exposed end of the nipple valves and replacing nonabsorbable sutures with absorbable sutures (34). Treatment of bladder stones is discussed below; however, options such as endoscopic lithotripsy (36), extracorporeal shock wave lithotripsy (37), percutaneous lithotripsy (38,39), and open stone removal (40) have been advocated for removing calculi from continent urinary diversions. Cases of malignancy within a pouch reservoir have also been documented (41,42). Complications similar to those listed above are also possible after ileal conduit diversion (43). However, the complication rate associated with this procedure has steadily declined with improvements in preoperative nutritional assessment, surgical technique, antibiotic prophylaxis, absorbable suture material, and intensive care. Meticulous care during manipulation of the bowel and the ureter is necessary for preserving maximum blood supply. In addition, watertight ureterointestinal anastomoses, internal drainage with silicone stents, external drainage with Jackson Pratt or Penrose drains, and stoma drainage with red rubber catheters help to decrease the risk of early postoperative leakage. Ureterointestinal leakage should be suspected when the patient is well hydrated, but stomal output decreases with a concurrent increase in drain output. The initial workup should include a measurement of the creatinine concentration in the drain fluid and either a loopogram or an intravenous pyelogram. Maintenance of the drains already in place is essential, but a certain percentage of patients may require additional urinary diversion, such as nephrostomy tubes or percutaneous drainage of a urinary collection. Definitive surgical management with open revision will often be necessary if conservative measures fail. Early urinary leakage may lead to later ureterointestinal anastomotic stricture. Management techniques described for such a problem include endoscopic incision and stenting with open revision (44). Prolonged ileus, wound infection, sepsis, myocardial infarction, pulmonary embolism, and small-bowel obstruction are potential problems after any open abdominal procedure, and typical medical and surgical management is initiated. In cases of prolonged ileus or small-bowel obstruction, consultation with a general surgeon is often helpful. When concurrent bowel and urinary complications are encountered, such as a bowel leak and a urinary leak, treatment of the bowel condition should take precedence over treatment of the urinary condition because bowel complications are more likely to result in pelvic abscess, sepsis, and even death. Finally, adenocarcinoma has been reported to occur in the created ileal conduit diversion (45), but the rates are much lower than those associated with adenocarcinoma after ureterosigmoidostomy. A final type of urinary diversion is an orthotopic bladder substitution or a neobladder. Usually, a

Complications of Bladder and Prostate Surgery

577

Figure 3 Studer neobladder. Source: From Ref. 32.

segment of distal ileum is isolated and detubularized for constructing a low-pressure, high-volume reservoir that is anastomosed to the native urethra. Continence is maintained via the external sphincter under voluntary control. Initially described by Camey et al. (46), orthotopic diversion may be performed by using any of the various techniques described by Elmajian et al. (47,48), Hautman et al. and Flohr et al. (49,50), Ghoneim et al. (51), Studer et al. (52), and others (53–55). Examples of orthotopic urinary diversion are shown in Figures 3 and 4. The Kock pouch has also been modified for emptying via the urethra.

Figure 4 Ileal (Hautmann) neobladder. Source: From Ref. 32.

578

Bell et al.

Some of the complications associated with orthotopic bladder substitution are similar to those associated with cutaneous diversion, such as urinary leakage, infection with urosepsis, pouch stones, ileus, obstruction, and incisional hernia. In one report, urinary continence was achieved by 85% of Kock neobladder substitutes; continence was defined as dryness or the use of no more than one pad during the day and night. This high rate of continence may be related to minimal anterior urethral dissection at the time of prostatic dissection (47). According to Schiff and Lytton (56), continence rates are mainly related to the pressure generated by the reservoir, the outflow resistance of the outlet, and the use of bowel detubularization to reduce uninhibited involuntary bowel contractions. A secure, tension-free anastomosis between the urethra and the neobladder will reduce the likelihood of enterourethral fistula; however, if this complication does occur, it can usually be treated conservatively with drainage through a Foley catheter. Rupture of the neobladder must be included in the differential diagnosis of any patient who experiences acute abdomen and peritonitis during the postoperative period (57). Routine irrigation of the neobladder with saline is important because such irrigation will reduce the accumulation of intestinal mucus. The use of a neobladder is generally associated with a lower rate of reoperation during the late postoperative period. Erectile dysfunction may occur after radical cystoprostatectomy, regardless of the type of diversion used (58). Preoperative assessment of erectile function is crucial for managing postoperative difficulties (59). The introduction of the nerve-sparing cystoprostatectomy by Walsh and Mostwin (60) has led to improved postoperative potency rates (60). Treatments for patients for whom erectile dysfunction persists include Viagra1, vacuum erection devices, injectable intracavernosal agents, transurethral agents, and, ultimately, penile prosthesis (61). Metabolic complications of urinary diversion are always of concern to the urologist and are often an important cause of morbidity for the patient (62–64). Such consequences are mainly related to the type and length of bowel used, the type of diversion created, the patient’s overall general health, and a history of abdominal surgery or irradiation (64). The pathophysiology of these metabolic derangements has been studied in depth and may be beyond the scope of this chapter, but a brief review of the therapeutic options and metabolic consequences associated with each bowel segment is provided below. Parietal cells are stimulated to produce gastric acid (HCl) under the influences of the vagus nerve, histamine, and gastrin. Normally, the stomach exhibits a negative feedback loop through which increased production of acid decreases the antral production of gastrin. If this portion of the stomach is incorporated into the urinary reconstruction, a hypokalemic, hypochloremic metabolic alkalosis may occur and may lead

to ulcerations in the pouch (62). This systemic alkalosis may be worsened by severe renal failure, which impairs the kidney’s ability to secrete bicarbonate. Ulcerations may be part of the reason for the hematuria–dysuria syndrome (HDS) associated with gastrocystoplasty (see below). Treatment involves histamine receptor blockers, such as cimetidine or ranitidine, and proton pump inhibitors, such as omeprazole or lansoprazole (62). Prolonged exposure of the jejunum to urine results in severe and possibly fatal electrolyte abnormalities; thus, the use of the jejunum for urinary diversion has generally been abandoned. A hyponatremic, hypochloremic, hyperkalemic metabolic acidosis results from the enhanced secretion of sodium and chloride and the increased absorption of potassium and hydrogen (62). Patients soon become dehydrated because water follows sodium down its equilibrium gradient, and this dehydration results in increased production of aldosterone. Increased reabsorption of sodium by the kidney results in urine low in sodium and high in potassium. When the jejunal mucosal surface is exposed to urine with these concentrations, the cycle is perpetuated as more sodium is lost and more potassium is absorbed. In the severe form of this condition, known as the jejunal conduit syndrome, patients may experience nausea, vomiting, anorexia, dehydration, lethargy, muscle weakness, and fever (63). Correction of the acidosis with administration of bicarbonate, infusion of saline, and drainage of the reservoir via a catheter are the initial treatment steps, and prolonged oral supplementation of sodium chloride is usually required (63). The incorporation of ileum into urinary diversions is associated with much less severe metabolic abnormalities than is the incorporation of jejunum. A hypokalemic, hyperchloremic metabolic acidosis occurs; the extent of this acidosis may depend on the amount of ileum resected, the method of diversion used, and certain patient characteristics (64). A severe potential side effect of chronic acidosis is bone demineralization, which appears among children as rickets and among adults as osteomalacia. Osteoid replaces bone mineral because calcium and carbonate act as hydrogen ion buffers. Acidosis also impairs the activation of vitamin D and stimulates the activity of osteoclasts. These complications may be more evident among growing children and postmenopausal women than among other patients. The administration of sodium or potassium citrate is usually effective in correcting the acidosis, but supplemental chlorpromazine or nicotinic acid may be necessary for preventing chloride reabsorption because these two agents inhibit cyclic adenosine monophosphate, which impedes chloride reabsorption (62). Possible side effects of these medications include tardive dyskinesia (chlorpromazine) and exacerbation of liver dysfunction and peptic ulcer disease (nicotinic acid) (62). When large sections of the terminal ileum are resected, malabsorption of vitamin B12, bile acid,

Chapter 47:

fatty acid, and fat-soluble vitamins may occur (64). Monthly injections of vitamin B12 are recommended once a deficiency has been detected. Excessive concentrations of bile salts and fatty acids may lead to soponification with calcium and resultant steatorrhea and selective oxalate reabsorption; in such cases, treatment begins with cholestyramine (62). The use of large segments of colon in urinary diversion generally results in few major metabolic disturbances. Maintenance of the ileocecal valve is important for preventing the rapid transit of feces through the small bowel and colon and the reflux of large amounts of bacteria into the small intestine (64). Hypomagnesemia occasionally occurs because of malabsorption and renal tubular loss as the result of acidosis, and neuromuscular symptoms such as muscle fasciculation, tremor, tetany, and seizures may become evident (64). Treatment involves magnesium replacement. An increased ammonia load (because of increased absorption) in a patient with impaired hepatic function may result in hyperammonemia and hepatic coma (62). This rare syndrome is most commonly associated with ureterosigmoidostomy because urease-producing bacteria in the colon split urea to produce ammonia. Treatment involves draining the reservoir, administering lactulose, limiting protein intake, and administering antibiotics (62). Finally, adjustments in dosage may prevent the toxic effects of certain drugs, such as methotrexate, phenytoin, and theophylline, which are secreted unchanged in the urine and absorbed by the intestinal tract (62).

Partial Cystectomy A final method of bladder tumor resection is partial cystectomy (8). Several indications for partial cystectomy are a solitary, primary, muscle-invasive tumor in an area of the bladder that will allow for adequate surgical margins (usually 2 cm); inability to achieve complete transurethral resection because of location or size of the tumor; tumor in a bladder diverticulum; or the inability to perform urinary diversion, either because the patient refuses it or because the patient is not a candidate for such a procedure (65). The ideal indications for partial cystectomy include tumor along the posterior wall or tumors in the dome of the bladder; the dome is a common site for urachal adenocarcinomas. In addition, certain metastatic lesions to the bladder, such as colon cancer, may be amendable to partial cystectomy (66). Several important contraindications to partial cystectomy are multiple tumors, carcinoma in situ or cellular atypia, trigonal or prostatic invasion, the inability to obtain adequate surgical margins, the inability to maintain functional bladder capacity, and the presence of extravesical tumor extension (65). If ureteral reimplantation is necessary after resection, strict adherence to surgical margin requirements is essential (66). The general steps involved in partial cystectomy are transurethral resection or biopsy with pathological

Complications of Bladder and Prostate Surgery

579

review; bimanual examination; peritoneal exploration to rule out tumor spread with subsequent closure of the peritoneum before the procedure continues; dissection of pelvic lymph nodes; mobilization of the bladder and excision of the tumor with wide, 2-cm margins; frozen-section analysis of full-thickness bladder wall and adjacent perivesical fat so that surgical margin status can be assessed; bladder closure in two layers with absorbable sutures, with placement of a closed-suction drain near the surgical site; and bladder drainage with a Foley catheter (65). Placing a suprapubic tube should be avoided because the development of a vesicocutaneous fistula may lead to tumor recurrence in this area; the recurrence rate for TCC after partial cystectomy ranges from 0% to 18% (65). Neoadjuvant and adjuvant chemotherapy or radiation after bladder-sparing surgery has been advocated for reducing the risk of wound recurrence (66). Most complications associated with partial cystectomy are considered minor and include wound hematoma or infection, urinary leak, transient hydronephrosis, and diminished bladder capacity. The overall complication rates range from 11% to 29% (65). Before the Foley catheter is removed, a cystogram should be performed so that the completeness of healing can be assessed. Indications for partial cystectomy other than malignancy include severe vesical endometriosis, refractory interstitial cystitis, and traumatic repair.

BLADDER AUGMENTATION Another important procedure associated with potential urologic complications is bladder augmentation. Decades of work, mostly with pediatric patients, have provided various alternatives for increasing bladder capacity and compliance, improving bladder emptying, and strengthening outlet resistance. Initially described in the 1890s, bladder augmentation was performed to correct small, contracted bladders that resulted from tuberculosis cystitis (67). This condition remained the primary indication for bladder augmentation throughout the first half of the 20th century, until Lapides et al. (68) introduced clean intermittent catheterization and McGuire et al. (69) described the concept of bladder pressure and compliance. A wide range of subsequent indications for bladder augmentation soon developed, such as congenital anomalies (myelodysplasia, tethered spinal cord, posterior urethral valves, prune belly syndrome, bladder extrophy, and cloacal extrophy), spinal cord injury, multiple sclerosis, refractory interstitial and radiation cystitis, previous radical pelvic surgery, and traumatic injury (70). Patients, especially children, may exhibit severe frequency, urgency, incontinence, upper and lower UTIs, VUR, various degrees of renal scarring or impaired renal growth, and, most importantly, a lowvolume, poorly compliant, high-pressure bladder, as demonstrated by urodynamic evaluation (71). Usually, a conservative trial of behavioral training, intermittent self-catheterization, and anticholinergic medications is

580

Bell et al.

attempted, but when these measures fail to alleviate symptoms, surgical intervention is often required (71). Preoperative evaluation may include a voiding diary, urinalysis and urine culture, cytology (for adult patients), determination of serum blood count and chemistries for ruling out renal failure, liver function tests for ruling out hepatic insufficiency, video urodynamics, cystoscopy, and an evaluation of urinary tract anatomy and function. Examinations may include intravenous pyelogram, voiding cystourethrogram, and radionuclide scans (67). The selection of tissue for the augmentation may depend on several factors. A severely dilated ureter, as seen in some cases of VUR or posterior urethral valves, may be used after ipsilateral nephrectomy for a poorly functioning or nonfunctioning kidney (72–74). Although such a ureter is rarely available, ureterocystoplasty results in no production of mucus and no electrolyte abnormalities because enteric segments are kept free of exposure to urine (72). Using the stomach for gastrocystoplasty has been advocated for patients with renal dysfunction and acidosis because of the net secretion of hydrogen and chloride ions and the lack of absorption of ammonium (Fig. 5) (75–77). Some patients, however, are at risk of hypokalemic, hypochloremic metabolic alkalosis after gastrocystoplasty (74,78,79). The production of mucus and the occurrence of UTIs and stones are limited after gastrocystoplasty (76). As many as onethird of patients who undergo this type of augmentation experience repeated bouts of coffee-brown or red urine, suprapubic pain, cutaneous irritation, and burning with urination. This unique condition is referred to as the HDS (78–80). It is probably related to the lowered urinary pH that results from excessive secretion of HCl or from hypergastrinemia (76,79,80).

Most patients experience mild to moderate symptoms that are treatable with oral hydration and H-2 blockers or proton pump inhibitors. A small percentage of patients with intractable symptoms will require surgical revision for removing the gastric segment and replacing it with another intestinal segment (80). As with continent urinary diversion, augmentation cystoplasty may use either small or large intestine. Described by Bramble (81), the ‘‘clam’’ enterocystoplasty, which uses a terminal ileal patch, provides relief of retractable bladder instability, increased compliance, and functional bladder capacity; it is associated with minimal early and late complications (81). Another option is ileocystoplasty, provided that no short-gut syndrome or other ileal pathology exists (67,70,82). Problems such as the production of mucus and electrolyte abnormalities (hypokalemia, hyperchloremia, and metabolic acidosis) are similar to those associated with ileal neobladders and ileal conduits. Treatment, especially for children, is important so that the development of osteomalacia and possible growth delay can be prevented (82). Daily bladder irrigation is essential during the early postoperative period because excessive production of mucus can obstruct urinary drainage, increase the risk of UTI, and act as a nidus for stone formation (83–85). Rates of stone formation as high as 52% have been reported after augmentation cystoplasty (84). The management of bladder stones is discussed below. Sigmoid colonic segments are also used for bladder augmentation because for some patients, especially children with myelodysplasia, leaving the ileocecal valve intact is imperative so that diarrhea can be prevented and fecal continence can be maintained (67). An alternative technique, known as seromuscular colocystoplasty lined with urothelium, involves

Figure 5 Gastrocystoplasty. Source: From Ref. 75.

Chapter 47:

removing the detrusor while sparing the urothelium; this is followed by an onlay of detubularized bowel void of its mucosa (74). The urodynamic results and continence rates associated with this procedure are similar to those associated with standard augmentation; however, this procedure avoids morbidity related to excessive production of mucus, metabolic derangement, and spontaneous perforation (86,87). Regardless of the tissue segment used, most patients will require life-long intermittent selfcatheterization because the bladder remains hypocontractile. During follow-up, routine ultrasound measurement of postvoid residual urine will help gauge how well the bladder is emptying. When urethral catheterization is difficult, uncomfortable, or impossible, a catheterizable stoma is constructed, usually from the appendix, and is brought from the augmented bladder to the skin (88,89). The Mitrofanoff flap valve stoma (90) has provided durable results with respect to continence rates and stomal stenosis or stricture. Perforation of the augmented bladder is always a feared complication; it occurs in as many as 3% to 6% of cases (67,71). Suspected causes of perforation are ischemia of the bowel segment wall as the result of overdistension with urine, trauma resulting from self-catheterization, chronic UTI, and a competent bladder outlet (91,92). Patients may exhibit acute abdomen or florid urosepsis or may simply complain of nausea with vague abdominal pain and distension; therefore, the physician must maintain a high degree of suspicion with regard to perforation. Although this complication usually occurs during the early postoperative period, an increasing number of cases of delayed perforation have been reported (92). If the situation permits, administering intravesical contrast and performing a cystogram will aid in the diagnosis of a leak or rupture. Intravenous hydration, the administration of broad-spectrum antibiotics, and urgent operative exploration and repair are generally the rule, although small leaks may be treated with urethral or suprapubic catheter drainage (92). Perforation sites may include the bowel segment, the anastomosis, and the native bladder. Preventing bladder overdistension by using strictly timed bladder emptying is encouraged. Postoperative incontinence may be related to transposition of too short an intestinal segment and persistent low bladder capacity, a poorly constructed stoma, repeated bouts of UTI, or an incompetent bladder neck and external urethral sphincter (56). Therapeutic options for patients with poor outlet resistance are concomitant bladder neck sling or suspension, periurethral injection of a bulking agent, artificial urinary sphincter, bladder neck tubularization such as the Kropp procedure or Pippi Salle procedure, or bladder neck closure (93). After reviewing the available literature, Kryger et al. (93) suggested that placing an artificial urinary sphincter may be the procedure of choice for treating neurogenic sphincteric

Complications of Bladder and Prostate Surgery

581

incontinence with the goal of preserving spontaneous voiding (93). Some patients may be affected by nocturnal incontinence resulting from a full bladder (94). A strict sterile technique is crucial for preventing infection of an artificial urinary sphincter; such infection would require surgical removal. Small-bowel obstruction related to intraabdominal adhesions usually requires operative intervention. Although the true incidence of tumor occurrence is not known, several cases of malignancy within augmented bladders created with various intestinal segments have been reported (95–99). Lifelong tumor surveillance with annual cystoscopy is recommended and should begin 10 years after surgery, especially for younger patients with long life expectancies (96).

BLADDER STONES Reports of and references to bladder calculi are numerous throughout medical history. Stones have been found in Egyptian mummies, were mentioned in the Hippocratic oath, and were removed from the bladder by the earliest surgeons via perineal lithotomy around the first century, via suprapubic lithotomy in the 1500s, and via transurethral lithotomy in the 1800s (100). Various minimally invasive procedures are used today for removing bladder stones. Open surgery is reserved for the largest stones and is performed during concurrent open prostatectomy. In the past, bladder stones resulted primarily from malnutrition, dietary inadequacies, and UTI (100). Throughout much of the developed world, the incidence of bladder stones related to these causes has decreased dramatically; however, the incidence of bladder stones remains high in certain areas of Southeast Asia, Northern Africa, and the Middle East, mainly among children (100). For adults, the development of bladder stones is mainly related to urinary stasis, as seen with outlet obstruction and neurogenic bladder (101). Leading causes of bladder stones among men are benign prostatic hypertrophy, urethral stricture disease, and bladder neck contracture after prostatectomy; among women, the primary causes of bladder stones are cystoceles and obstruction after incontinence surgery (100,101). The encrustation of foreign bodies may lead to stone formation, as has been reported with retained ureteral stents, long-term indwelling Foley catheters (especially for patients with spinal cord injury), suture material, and materials placed by the patient. Patients may have no symptoms or may experience suprapubic pain, dysuria, frequency, hesitancy, terminal gross hematuria, or sudden interruption of voiding by the obstructing stone. Examination may provide evidence of suprapubic fullness, enlarged prostate, cystocele, neurologic abnormalities, or elevated postvoid residual volumes (100). Workup includes urinalysis for detecting infection (with subsequent antibiotic therapy if the results are positive); kidney, ureter, and bladder radiography

582

Bell et al.

to detect calcifications; and possibly renal–bladder ultrasound, spiral noncontrast computed tomography (CT), or cystourethroscopy for confirmation (100). Initial medical therapy with urinary alkalinization may be attempted for stone dissolution, but operative intervention will probably be necessary. Before surgical treatment is planned, the physician must recognize the underlying cause of stone formation. For example, a middle-aged man with a bladder stone probably has some degree of prostatic obstruction that may or may not require transurethral resection of the prostate (TURP) or open prostatectomy.

Transurethral Cystolitholapaxy and Shockwave Lithotripsy The initial approach for most patients is transurethral bladder stone destruction (100). In the operating room, a cystoscope can be placed and subsequent litholapaxy can be accomplished with a variety of lithotrites, such as mechanical lithotripsy, the Swiss Lithoclast1, electrohydrolic lithotripsy, ultrasound, and laser energy (102). The use of the holmium:YAG laser has been shown to be effective, technically feasible, and safe, even for bladder calculi larger than 4 cm in diameter (102,103). Potential perioperative complications associated with transurethral procedures are urinary infection or urosepsis, bleeding, and bladder perforation (100). Prophylactic antibiotic coverage is essential because many of these stones are inherently infected. Excessive postoperative hematuria can be prevented by careful instrumentation and avoidance of torque on the prostate; when this condition occurs, it can be managed with continuous vigorous bladder irrigation by hand or via a three-way catheter. Bladder perforations can be managed according to the procedures discussed above for TURBT. Late complications include urethral stricture after prolonged cystoscopic manipulation and recurrence of the bladder stone after inadequate treatment of the underlying cause (100). If concurrent TURP is to be performed, the lithotripsy portion should be performed first so that the potential complications of excessive blood loss and hypotonic fluid resorption associated with TURP can be avoided (100). Endoscopic treatment of bladder stones is reported to be safe and effective for children after augmentation cystoplasty (104). An alternative, minimally invasive procedure for treating bladder stones is shockwave lithotripsy (SWL). Adequate stone clearance with minimal complications has been reported when calculi are less than 3 cm in diameter and patients are treated in the prone position (105). SWL has also been used to treat larger calculi (4–6 cm in diameter) with patients in the seated position (106). Bladder outlet obstruction resulting from prostatic enlargement or urethral stricture must be treated initially so that fragment impaction can be prevented after SWL.

Percutaneous and Open Stone Removal An alternative to transurethral stone destruction, especially for children and patients with surgically reconstructed bladders, is percutaneous suprapubic cystolitholapaxy (107–111). A shorter instrument with a larger diameter is used for this procedure; this instrument allows rapid stone destruction and evacuation of stone fragments. Different means of percutaneous access have been described; the important step is avoidance of bowel injury, especially for patients who have previously undergone lower abdominal or pelvic surgery (107–111). The lithotrites used in this procedure are similar to those used in the transurethral approach; combinations of the two may also be used. When the bladder stone is accompanied by a large adenomatous prostate gland (more than 100 g) or a large bladder diverticulum, open suprapubic cystotomy with possible diverticulectomy may be the procedure of choice (100). Bladder calculi are removed with 100% accuracy before surgical removal of the enlarged prostatic lobe. Recuperation may be somewhat longer for open surgery. Prevention and treatment of potential postoperative complications such as sepsis and hemorrhage are similar to those for complications of transurethral cystolitholapaxy. Fistula formation is always a concern when the bladder is explored; its incidence can be reduced by meticulous two- or three-layer closure of the cystotomy site. Placing a drain may help detect any urinary leakage.

VESICOURETERAL REFLUX AND URETEROCELES VUR and ureteroceles are two common urologic anomalies that occur most frequently among children. VUR is the retrograde flow of urine from the bladder through the ureter, the kidney, or both (112). VUR may be primary, due to a congenital lack of longitudinal muscle of the intravesical ureter, or it may be secondary, due to elevated bladder pressure with bladder outlet obstruction or voiding dysfunction (113). A ureterocele is a congenital dilatation of the distal segment of the ureter. Ureteroceles have been traditionally categorized as intravesical or ectopic in location, unilateral or bilateral, and associated with a single or a duplicated collecting system (114). Surgical intervention with its associated potential complications is often necessary for treating both disease processes. Prenatal hydronephrosis often leads to VUR after birth. Alternatively, when newborns or young children exhibit symptoms of a UTI, renal–bladder ultrasonography and voiding cystourethrography will help confirm the presence of hydronephrosis and urinary reflux (113). Repeated bouts of infection may lead to renal scarring and subsequent loss of renal function. Thus, the primary objective of treating children with VUR is preventing these episodes of pyelonephritis and preserving renal function (112).

Chapter 47:

Initial antibiotic prophylaxis is the medical treatment of choice for most grades of VUR because the reflux often spontaneously resolves. Indications for surgical intervention include breakthrough UTIs, high-grade reflux associated with renal scarring, noncompliance with antibiotic prophylaxis, persistence of reflux in girls approaching puberty, and worsening in the grade of reflux over time (113). The mainstay of surgical treatment is ureteral reimplantation, which can be performed as an open or a laparoscopic procedure and either intravesically or extravesically (113). Subureteral polytetrafluoroethylene injection has been described as an alternative endoscopic approach for correcting VUR, but the likelihood of durable results may be lower than that associated with primary reimplantation (115). The primary complication of ureteral reimplantation is subsequent urinary obstruction, mainly as a result of the technical intricacies of the procedure. However, when ureteral reimplantation is performed by experienced pediatric urologists, the associated rate of obstruction is relatively low (113). Postoperative stenting may aid in preventing early, transient ureteral obstruction by allowing edema and spasm to resolve. Late ureteral obstruction may be a result of tissue ischemia or ureteral angulation and compression during bladder filling. In such cases, surgical revision is often necessary (113). Ureteroceles may be diagnosed by prenatal ultrasonography; alternatively, young infants may exhibit bladder outlet obstruction or urosepsis (114). In most cases, intravesical ureteroceles may be initially treated with endoscopic incision, but such treatment may not be possible for extravesical (ectopic) ureteroceles because controversy exists about the initial therapeutic intervention (114,116). Ectopic ureteroceles are often associated with an ipsilateral duplicated collecting system, and the upper pole moiety is often substantially dilated as a result of the distal obstruction. Initial transurethral puncture may temporarily stabilize renal function, but as many as 70% of patients will require secondary reconstructive surgery (117). Complete recovery of renal function is usually not possible even after the obstruction has been relieved; therefore, one initial surgical option is upper pole partial nephrectomy with or without ureterectomy (114,116). In addition to the upper pole obstruction, as many as 50% of patients will have VUR of the lower pole moiety or the contralateral kidney (116). An alternative surgical option, therefore, is a combination of upper tract surgery (e.g., heminephrectomy) with lower tract surgery (e.g., ureteral reimplantation) (114,116). Because not all patients with VUR will require reimplantation, patient selection is important and is based on the degree of reflux, the presence or absence of renal scarring, and the patient’s age (118). Potential complications of partial nephrectomy are acute vascular injury to the functioning lower pole; hemorrhage; the formation of urinary leak,

Complications of Bladder and Prostate Surgery

583

urinoma, or both; and UTI. Careful dissection, knowledge of renal vascular anatomy, and placement of a drain postoperatively will aid in preventing such complications. Incomplete excision or inaccurate puncture of the ureterocele may cause the formation of an obstructing urethral flap, whereas excessive excision may damage the urinary sphincter, thereby causing incontinence (114).

URINARY INCONTINENCE Urinary incontinence is frequently a reason for patients’ visits to a urology clinic; this condition may be a result of dysfunctional bladder filling or emptying. A complete review of incontinence is beyond the scope of this chapter, and the forms of incontinence (stress, urge, overflow, and mixed) have been meticulously described in textbooks and in the medical literature. A careful history and physical examination will usually uncover the underlying cause. Medical therapy with agents such as tolterodine and oxybutynin is useful in treating the ‘‘overactive’’ bladder by relieving the symptoms of urgency and frequency. Various options for surgical treatment of urinary incontinence have been advocated. The administration of periurethral bulking agents may be used as an initial attempt at treating incontinence related to intrinsic urinary sphincter deficiency if the patient is reluctant to undergo an open procedure. Alternatively, the administration of such agents may be attempted as an adjuvant treatment for persistent leakage after an open procedure (119). The agents are usually placed transurethrally under direct cystoscopic vision. Agents approved by the U.S. Food and Drug Administration include bovine collagen, autologous fat, and carbon beads (Durasphere1) (119). Bulking agents are an attractive alternative for patients because their placement is relatively simple, the associated morbidity rates are low, and patients often notice immediate results. Although bulking agents do not usually provide a permanent cure, repeated injections may leave many patients satisfied with the results (119). Temporary postoperative urinary retention is the most frequently encountered complication; it probably results from urethral edema along the injection site or from sphincter spasm. Treatment usually involves intermittent self-catheterization for a few days until the edema has resolved. Symptomatic UTIs may occur and are best prevented by ensuring that the results of preoperative urine cultures are negative for infection and by maintaining sterile technique through out the performance of the procedure. Open surgical procedures for urinary incontinence include transvaginal bladder neck suspensions, retropubic urethropexy, pubovaginal sling, and artificial urinary sphincters (120). Retropubic urethropexy and its associated complications are discussed in chapter 48 of this textbook. An in-depth comparison

584

Bell et al.

of specific anti-incontinence procedures, with a review of expected results and reasons for selection, is not the focus of this chapter. Therefore, documented complications associated with the surgical treatment of urinary incontinence will be discussed as a whole. Intraoperative complications may be recognized during the procedure or shortly thereafter. Injuries to the bladder or the urethra may occur during dissection or when a needle or trocar is passed to position the suspending material (121). Cystoscopy will often rule out such injuries. Some authors advocate immediate repair or repositioning the needle or trocar and continuing the procedure (121). Failure to recognize such an injury may result in vesicovaginal or urethrovaginal fistulas. When transvaginal fistulas occur, their repair may be early or delayed, with similar results (122,123). In addition, any retained material, such as permanent suture, may result in recurrent UTI or bladder stone formation. Inadvertent bowel injury may occur during trocar placement, suprapubic dissection, or placement of a suprapubic catheter, and the subsequent risk of infection is increased when artificial materials are used (120). Mild bleeding is frequently the result of transvaginal dissection, but serious hemorrhage and severe pelvic or vaginal hematomas are relatively uncommon (121). Infiltration of the anterior vaginal wall with normal saline will facilitate development of the proper plane for dissection. Direct pressure may be temporarily applied, but sutureligature of the bleeding vessels under direct vision may be necessary if the procedure is to continue. Postoperative vaginal packing often controls venous oozing. Most patients are placed in the dorsal lithotomy position for anti-incontinence procedures; such positioning creates the risk of postoperative nerve damage (121). Careful patient positioning and padding of pressure points may reduce the risk of damage to the peroneal nerve; this damage manifests itself postoperatively by ipsilateral foot drop. Acute and chronic postoperative complications are also well known. One of the most frequently experienced complications after any type of anti-incontinence procedure is postoperative urinary retention (121). This condition often results from transient inflammation and edema at the dissection site, the effects of general anesthesia, or the patient’s perception of pain on urination immediately after the procedure (121). In addition, suture or sling material may have been fashioned or secured too tightly, and this problem results in compression of the urethra or the bladder neck (120). Placing a suprapubic catheter during the procedure helps prevent acute retention. Alternatively, patients may need to learn to perform intermittent self-catheterization. If outlet obstruction persists, urethrolysis may be indicated (124). Approximately 5% to 10% of patients experience persistent stress urinary incontinence, which may occur if the sling material is tied too loosely, is positioned improperly, is separated from its anchoring tissue, or undergoes breakage or degradation (120). Approximately 10% to 30% of

patients may experience de novo urgency symptoms as the result of a sudden change in bladder outlet resistance induced by the anti-incontinence procedure itself (120). Symptoms are usually transient and are best managed with usual anticholinergic therapy. Finally, some patients who undergo artificial urinary sphincter implantation may require revision for erosion, malfunction, or infection (125).

LAPAROSCOPY Some physicians have advocated increased use of the laparoscopic approach to bladder surgery because laparoscopy is associated with less postoperative pain, shorter hospital stays, and improved cosmetic results (126). In fact, several of the previously described procedures have been performed laparoscopically, such as pelvic lymphadenectomy, simple and radical cystectomy, ileal conduit diversion, bladder augmentation, ureteral reimplantation, and bladder neck suspension (126–130). Complication rates decrease as operative experience increases; however, certain complications are unique to laparoscopy (128). More than a decade ago, the laparoscopic procedure most commonly performed was pelvic lymph node dissection before radical prostatectomy. At that time, if the results of intraoperative frozen-section examination were positive for lymph node involvement, consideration was given to not performing prostatectomy. Changes in clinical management have limited the role of isolated laparoscopic pelvic lymph node dissection, but many of the reports of complications in the literature include this procedure for analysis and discussion. In one review of 372 patients who underwent laparoscopic pelvic lymph node dissection, vascular injuries were the most common complication, followed by visceral injury to the bowel, bladder, and ureter (129). The mechanism of injury was usually traumatic trocar insertion or sharp dissection. The epigastric vessels, distal aorta, and common iliac vessels are at greatest risk of injury, and these vessels can best be avoided by careful midline insertion of the Veress needle or the Hasson cannula (128,129,131). When a vascular injury is immediately recognized, control and repair may be attempted laparoscopically, but conversion to laparotomy should occur without hesitation if initial attempts are unsuccessful (132). Inspecting the abdomen with lower insufflation pressures, both before and at the end of the case, will help identify possible injuries. When the operation proceeds near small or large bowel, the use of bipolar coagulation or ultrasonic dissectors and limiting the use of monopolar coagulation may decrease the risk of injury or perforation to the bowel (128). A complication that is unique to laparoscopy is carbon dioxide (CO2) absorption. Elevated CO2 absorption during laparoscopic pelvic procedures occurs when subcutaneous emphysema is present,

Chapter 47:

Complications of Bladder and Prostate Surgery

585

Figure 6 Control of bleeding from anterior wall vessel injured during port placement. Source: From Ref. 132.

when extraperitoneal insufflation is performed, and when the duration of insufflation is prolonged (133). Increased absorption may place patients with pulmonary compromise (chronic obstructive pulmonary disease, obesity, etc.) at an added risk of intraoperative or postoperative complications. Meticulous fascial closure of all port sites larger than 5 mm will help reduce the risk of postoperative incisional hernia (Fig. 6) (128). Finally because the likelihood of complications associated with laparoscopic surgery is higher when the procedure is performed by inexperienced surgeons, formal urologic residencies and fellowship training should include training in such skills (128,134).

will also have a ruptured bladder (137). If the patient’s condition is hemodynamically stable, the key radiographic study is retrograde urethrography plus cystography or CT-guided cystography (135–141). When performed properly, static cystography involves instilling 300 to 500 cc of a water-soluble contrast solution under gravity, with slight overinjection at the end

BLADDER TRAUMA An important topic of discussion regarding bladder complications is traumatic bladder injury. Again, a substantial volume of literature about urologic injury is available for review; therefore, this chapter will contain only a brief overview of mechanisms of injury, diagnosis, treatment, and potential complications. Traumatic injuries to the bladder include contusions, extraperitoneal rupture, and intraperitoneal rupture. Such injuries account for approximately 22% of all urologic injuries (135). Blunt trauma (e.g., motor vehicle accidents, falls, and assault) to the pelvis and abdomen is the most common mechanism of bladder injury, accounting for 60% to 85% of injuries, whereas penetrating bladder trauma (e.g., gunshot wounds and stabbings) accounts for the remainder (136). Approximately 95% of patients with a ruptured bladder will exhibit gross hematuria, and 50% to 85% of the perforations will be extraperitoneal (137). Other important findings include suprapubic pain, anuria, blood at the urethral meatus, perineal swelling and hematoma, or the presence of a high riding prostate detected by digital rectal examination. Careful consideration must be given to an injury in the lower urinary tract when a severe pelvic fracture has been detected because 5% to 10% of patients with a fractured pelvis

Figure 7 Intraoperative cystogram showing (A) filling and (B) drainage films.

586

Bell et al.

CASE STUDY An 82-year-old woman visited a urologist. She had a history of low-grade, lowstage papillary transitional cell carcinoma of the bladder with several recurrences over a 10-year period. The disease had not progressed during this time period. She had missed several scheduled appointments for routine follow-up. At the time of the current presentation, she was not experiencing gross hematuria or other lower urinary tract symptoms. Given her history, flexible cystoscopy was performed in the physician’s office. Several large (4–5 cm) bladder tumors with a papillary appearance were detected on the right lateral wall and the dome of the bladder. The patient was scheduled for TURBTs. In the operating room, the patient was given spinal anesthesia and was placed into the lithotomy position. The entire bladder was thoroughly inspected with the 12
and the 70
lens. The previously identified bladder tumors were readily visible. Transurethral resection was begun with the tumor located along the lateral wall of the bladder. An attempt was made to resect the tumor at its base and along the bladder wall. After resection, perivesical fat was readily identified, a finding suggestive of bladder perforation. Further tumor resection was discontinued, and intraoperative gravity cystography was performed. The contrast material surrounded the exterior of the bladder but did not enter the peritoneal cavity (Fig. 7). An extraperitoneal bladder perforation was diagnosed. Hemostasis was achieved, and a 20-Fr Foley catheter was placed for drainage. The patient remained in the hospital overnight for observation and was discharged home the following morning. Oral antibiotics were prescribed. Twelve days later, follow-up formal cystography showed no evidence of extravasation. The Foley catheter was removed, and the patient was free of symptoms. She was scheduled for subsequent TURBT.

of filling under pressure and appropriate anterior– posterior, oblique (if feasible), and postdrainage views (137). Extravasation confined to the pelvis appears as whisks, streaks, or sunbursts and indicates extraperitoneal bladder rupture, whereas extravasation that diffuses throughout the abdominal cavity and outlines bowel or settles in the paracolic gutters indicates intraperitoneal bladder rupture. Most penetrating injuries require surgical exploration so that associated injuries to other organs and structures can be detected (140). When intraperitoneal bladder rupture is diagnosed after blunt injury, surgical exploration and repair are usually indicated because concomitant visceral injury is usually present (137,141). For extraperitoneal bladder rupture, nonoperative conservative management is recommended because minimal morbidity has been reported (137–142). Therapy consists of Foley catheter drainage and administration of broad-spectrum antibiotics for 7 to 10 days, at which time a follow-up cystogram is performed for documenting complete bladder healing before the catheter is removed. Persistent gross hematuria with obstructing clots, evidence of urosepsis, or persistent extraperitoneal extravasation should alert the clinician to the need for surgical exploration. Surgical principles of repair of traumatic bladder rupture include dissection through the peritoneum to the bladder dome, inspection of the entire bladder for injury, identification of the course of the ureters, avoidance of exploration of pelvic hematomas,

water-tight closure of the bladder in two or three layers with 2–0 or 3–0 absorbable sutures, placement of a suprapubic catheter through a separate cystotomy incision, and possible placement of a pelvic drain (136,137). Again, administering antibiotics will decrease the risk of pelvic abscess, especially if a pelvic hematoma or orthopedic hardware is present. Potential complications that may result from the trauma itself or during the recovery period include persistent urinary extravasation, hemorrhage, pelvic infection, wound dehiscence, small capacity bladder, de novo urge incontinence, and erectile dysfunction (136,143). A large volume of output from the pelvic drain may indicate a persistent urinary leak, usually due to unrecognized lacerations or leakage along the suture line. This leakage usually resolves with continued catheter drainage but may occasionally place the fascia at risk of separation. During surgical exploration, avoiding pelvic hematomas will decrease the risk of serious bleeding. Antibiotic coverage is essential for preventing pelvic infection or UTI. Excessive debridement of the bladder injury can result in loss of functional bladder capacity and the development of bothersome lower urinary tract symptoms. Finally, neurovascular injury resulting from the trauma or from radical surgical exploration can lead to erectile dysfunction. If erectile dysfunction occurs, its evaluation should be delayed until all serious injuries have been addressed and stabilized (143).

Chapter 47:

CONCLUSIONS A wide variety of indications exist for bladder surgery. Inherent to the techniques described above are potential associated complications. Thorough knowledge of patient selection, pathophysiology, anatomy, surgical technique, and potential complications will reduce the morbidity and mortality rates associated with bladder surgery.

REFERENCES 1. Walsh PC, Retik AB, Vaughan ED, Wein AJ. Urothelial tumor of the urinary tract. In: Walsh PC, Wein AJ, Retik AB, Vaughan ED, Vaughan ED Jr, eds. Campbell’s Urology. 7th ed. Vol. 3. Philadelphia: Saunders, 1998: 2327–2410. 2. Holzbeierlein JM, Smith JA Jr. Surgical management of noninvasive bladder cancer (stages Ta/T1/CIS). Urol Clin North Am 2000; 27:15–24, vii–viii. 3. Mydlo JH, Weinstein R, Shah S, Solliday M, Macchia RJ. Long-term consequences from bladder perforation and/or violation in the presence of transitional cell carcinoma: results of a small series and a review of the literature. J Urol 1999; 161:1128–1132. 4. Smith JA Jr. Laser surgery for transitional-cell carcinoma. Technique, advantages, and limitations. Urol Clin North Am 1992; 19:473–483. 5. Smith JA Jr. Laser treatment of urologic cancers. Semin Surg Oncol 1989; 5:30–37. 6. Lerner SP, Skinner E, Skinner DG. Radical cystectomy in regionally advanced bladder cancer. Urol Clin North Am 1992; 19:713–723. 7. Herr HW. Transurethral resection in regionally advanced bladder cancer. Urol Clin North Am 1992; 19:695–700. 8. Hinman F Jr. Partial cystectomy. In: Hinman F, Donley S, Stempen PH, eds. Atlas of Urologic Surgery. 2nd ed. New York: Elsevier, 1998:501–520. 9. Droller MJ. Bladder cancer: state of the art care. CA Cancer J Clin 1998; 48(5):269–284. 10. Bales GT, Kim H, Steinberg GD. Surgical therapy for locally advanced bladder cancer. Semin Oncol 1996; 23:605–613. 11. Gschwend JE, Fair WR, Vieweg J. Radical cystectomy for invasive bladder cancer: contemporary results and remaining controversies. Eur Urol 2000; 38:121–130. 12. Turner WH, Studer UE. Cystectomy and urinary diversion. Semin Surg Oncol 1997; 13:350–358. 13. Frazier HA, Robertson JE, Paulson DF. Complications of radical cystectomy and urinary diversion: a retrospective review of 675 cases in 2 decades. J Urol 1992; 148:1401–1405. 14. Skinner DG, Crawford ED, Kaufman JJ. Complications of radical cystectomy for carcinoma of the bladder. J Urol 1980; 123:640–643. 15. Johnson DE, Lamy SM. Complications of a single stage radical cystectomy and ileal conduit diversion: review of 214 cases. J Urol 1977; 117:171–173. 16. Brannan W, Fuselier HA Jr, Ochsner M, Randrup ER. Critical evaluation of 1-stage cystectomy—reducing morbidity and mortality. J Urol 1981; 125:640–642.

Complications of Bladder and Prostate Surgery

587

17. Neulander EZ, Hawke CK, Soloway MS. Incidental appendectomy during radical cystectomy: an interdepartmental survey and review of the literature. Urology 2000; 56:241–244. 18. Hardeman SW, Soloway MS. Urethral recurrence following radical cystectomy. J Urol 1990; 144:666–669. 19. Freeman JA, Esrig D, Stein JP, Skinner DG. Management of the patient with bladder cancer. Urethral recurrence. Urol Clin North Am 1994; 21:645–651. 20. Braslis KG, Soloway MS. Management of ureteral and renal pelvic recurrence after cystectomy. Urol Clin North Am 1994; 21:653–659. 21. Walsh PC, Retik AB, Vaughan ED, Wein AJ. Use of intestinal segments and urinary diversion. In: Campbell’s Urology. Vol. 3. 1998:3121. 22. Hendren WH. Historical perspective of the use of bowel in urology. Urol Clin North Am 1997; 24: 703–713. 23. Fisch M, Wammack R, Mu¨ller SC, Hohenfellner R. The Mainz pouch II (sigma rectum pouch). J Urol 1993; 149:258–263. 24. Spence HM, Hoffman WW, Fosmire GP. Tumour of the colon as a late complication of ureterosigmoidostomy for exstrophy of the bladder. Br J Urol 1979; 51: 466–470. 25. Husmann DA, Spence HM. Current status of tumor of the bowel following ureterosigmoidostomy: a review. J Urol 1990; 144:607–610. 26. Sohn M, Fuzesi L, Deutz F, Lagrange W, Kirkpatrick JC, Braun JC. Signet ring cell carcinoma in adenomatous polyp at site of ureterosigmoidostomy 16 years after conversion to ileal conduit. J Urol 1990; 143:805–807. 27. Gittes RF. Carcinogenesis in ureterosigmoidostomy. Urol Clin North Am 1986; 13:201–205. 28. Bricker EM. Bladder substitution after pelvic evisceration. Surg Clin North Am 1950; 30:1511–1521. 29. Kock NG, Nilson LO, Nilsson LJ, Norlen LJ, Philipson BM. Urinary diversion via a continent ileal reservoir: clinical results in 12 patients. J Urol 1982; 128:469–475. 30. Rowland RG, Mitchell ME, Bihrle R, Kahnoski RJ, Piser JE. Indiana continent urinary reservoir. J Urol 1987; 137:1136–1139. 31. Razor BR. Continent urinary reservoirs. Semin Oncol Nurs 1993; 9:272–285. 32. Benson MC, Olsson CA. Urinary diversion. Urol Clin North Am 1992; 19:779–795. 33. Killeen KP, Libertino JA. Management of bowel and urinary tract complications after urinary diversion. Urol Clin North Am 1988; 15:183–194. 34. Lieskovsky G, Skinner DG, Boyd SD. Complications of the Kock pouch. Urol Clin North Am 1988; 15: 195–205. 35. Stein JP, Freemen JA, Esrig D, et al. Complications of the afferent antireflux valve mechanism in the Kock ileal reservoir. J Urol 1996; 155:1579–1584. 36. Cohen TD, Streem SB. Minimally invasive endourologic management of calculi in continent urinary reservoirs. Urology 1994; 43:865–868. 37. Boyd SD, Everett RW, Schiff WM, Fugelso PD. Treatment of unusual Kock pouch urinary calculi with extracorporeal shock wave lithotripsy. J Urol 1988; 139:805–806. 38. Seaman EK, Benson MC, Shabsigh R. Percutaneous approach to treatment of Indiana pouch stones. J Urol 1994; 151:690–692.

588

Bell et al.

39. Jarrett TW, Pound CR, Kavoussi LR. Stone entrapment during percutaneous removal of infection stones from a continent diversion. J Urol 1999; 162:775–776. 40. Sait K, Stuart G, Nation J, Ghatage P. Urolithiasis following formation of a continent urostomy: case report and review of the literature. Gynecol Oncol 2000; 77:330–333. 41. L’Esperance JO, Lakshmanan Y, Trainer AF, Jiang Z, Blute RD Jr., Ayvazian PA. Adenocarcinoma in an Indiana pouch after cystectomy for transitional cell carcinoma. J Urol 2001; 165:901–902. 42. Wang DS, Silverman ML, Bihrle W III. Stomal recurrence of bladder carcinoma after cystectomy. J Urol 1999; 162:157. 43. Schmidt JD, Hawtrey CE, Flocks RH, Culp DA. Complications, results and problems of ileal conduit diversions. J Urol 1973; 109:210–216. 44. Cornud F, Lefebvre JF, Chretien Y, Helenon O, Moreau JF. Percutaneous transrenal electro-incision of ureterointestinal anastomotic strictures: long-term results and comparison of fluoroscopic and endoscopic guidance. J Urol 1996; 155:1575–1578. 45. Sakano S, Yoshihiro S, Joko K, Kawano H, Naito K. Adenocarcinoma developing in an ileal conduit. J Urol 1995; 153:146–148. 46. Camey M, Botto H, Richard E. Complications of the Camey procedure. Urol Clin North Am 1988; 15:249–255. 47. Elmajian DA, Stein JP, Esrig D, et al. The Kock ileal neobladder: updated experience in 295 male patients. J Urol 1996; 156:920–925. 48. Elmajian DA, Stein JP, Skinner DG. Orthotopic urinary diversion: the Kock ileal neobladder. World J Urol 1996; 14:40–46. 49. Hautmann RE, Egghart G, Frohnberg D, Miller K. The ileal neobladder. J Urol 1988; 139:39–42. 50. Flohr P, Hefty R, Paiss T, Hautmann R. The ileal neobladder—updated experience with 306 patients. World J Urol 1996; 14:22–26. 51. Ghoneim MA, Kock NG, Lycke G, el-Din AB. An appliance-free, sphincter-controlled bladder substitute: the urethral Kock pouch. J Urol 1987; 138:1150–1154. 52. Studer UE, Danuser H, Hochreiter W, Springer JP, Turner WH, Zingg EJ. Summary of 10 years’ experience with an ileal low-pressure bladder substitute combined with an afferent tubular isoperistaltic segment. World J Urol 1996; 14:29–39. 53. Reddy PK, Lange PH, Fraley EE. Total bladder replacement using detubularized sigmoid colon: technique and results. J Urol 1991; 145:51–55. 54. Benson MC, Seaman EK, Olsson CA. The ileal ureter neobladder is associated with a high success and a low complication rate. J Urol 1996; 155:1585–1588. 55. Cancrini A, De Carli P, Pompeo V, et al. Lower urinary tract reconstruction following cystectomy: experience and results in 96 patients using the orthotopic ileal neobladder substitution of Studer et al. Eur Urol 1996; 29:204–209. 56. Schiff SF, Lytton B. Incontinence after augmentation cystoplasty and internal diversion. Urol Clin North Am 1991; 18:383–392. 57. Haupt G, Pannek J, Knopf HJ, Schulze H, Senge T. Rupture of ileal neobladder due to urethral obstruction by mucous plug. J Urol 1990; 144:740–741. 58. Melman A. Iatrogenic causes of erectile dysfunction. Urol Clin North Am 1988; 15:33–39.

59. Lue TF. Impotence after radical pelvic surgery: physiology and management. Urol Int 1991; 46:259–265. 60. Walsh PC, Mostwin JL. Radical prostatectomy and cystoprostatectomy with preservation of potency. Results using a new nerve-sparing technique. Br J Urol 1984; 56:694–697. 61. Kirby R, Carson C, Goldstein I. Treatment. In: Kirby RS, Carson CC III, Goldstein I, eds. Erectile Dysfunction: A Clinical Guide. London: Taylor and Francis, 1999: 49–87. 62. McDougal WS. Metabolic complications of urinary intestinal diversion. J Urol 1992; 147:1199–1208. 63. Stampfer DS, McDougal WS, McGovern FJ. The use of bowel in urology. Metabolic and nutritional complications. Urol Clin North Am 1997; 24:715–722. 64. Mills RD, Studer UE. Metabolic consequences of continent urinary diversion. J Urol 1999; 161:1057–1066. 65. Sweeney P, Kursh ED, Resnick MI. Partial cystectomy. Urol Clin North Am 1992; 19:701–711. 66. Weinstein RP, Grob BM, Patcher EM, Soloway S, Fair WR. Partial cystectomy during radical surgery for nonurological malignancy. J Urol 2001; 166(1): 79–81. 67. Cranidis A, Nestoridis G. Bladder augmentation. Int Urogynecol J Pelvic Floor Dysfunct 2000; 11:33–40. 68. Lapides J, Diokno AC, Silber SJ, Lowe BS. Clean, intermittent self-catheterization in the treatment of urinary tract disease. J Urol 1972; 107:458–461. 69. McGuire EJ, Woodside JR, Borden TA, Weiss RM. Prognostic value of urodynamic testing in myelodysplastic patients. J Urol 1981; 126:205–209. 70. Goldwasser B, Webster GD. Augmentation and substitution enterocystoplasty. J Urol 1986; 135:215–224. 71. Duel BP, Gonzalez R, Barthold JS. Alternative techniques for augmentation cystoplasty. J Urol 1998; 159(3):998–1005. 72. Churchill BM, Aliabadi H, Landau EH, et al. Ureteral bladder augmentation. J Urol 1993; 150:716–720. 73. Bellinger MF. Ureterocystoplasty update. World J Urol 1998; 16:251–254. 74. Duel BP, Gonzalez R, Barthold JS. Alternative techniques for augmentation cystoplasty. J Urol 1998; 159: 998–1005. 75. Nguyen DH, Mitchell ME. Gastric bladder reconstruction. Urol Clin North Am 1991; 18:649–657. 76. Kurzrock EA, Baskin LS, Kogan BA. Gastrocystoplasty: is there a consensus? World J Urol 1998; 16:242–250. 77. Carr MC, Mitchell ME. Gastrocystoplasty. Sci World J 2004; 4(suppl 1):48–55. 78. Gosalbez R Jr., Woodard JR, Broecker BH, Warshaw B. Metabolic complications of the use of stomach for urinary reconstruction. J Urol 1993; 150:710–712. 79. Mingin GC, Stock JA, Hanna MK. Gastrocystoplasty: long-term complications in 22 patients. J Urol 1999; 162:1122–1125. 80. Nguyen DH, Bain MA, Salmonson KL, Ganesan GS, Burns MW, Mitchell ME. The syndrome of dysuria and hematuria in pediatric urinary reconstruction with stomach. J Urol 1993; 150:707–709. 81. Bramble FJ. The clam cystoplasty. Br J Urol 1990; 66: 337–341. 82. Shekarriz B, Upadhyay J, Demirbilek S, Barthold JS, Gonzalez R. Surgical complications of bladder augmentation: comparison between various enterocystoplasties in 133 patients. Urology 2000; 55:123–128.

Chapter 47:

83. Blyth B, Ewalt DH, Duckett JW, Snyder HM III. Lithogenic properties of enterocystoplasty. J Urol 1992; 148:575–579. 84. Palmer LS, Franco I, Kogan SJ, Reda E, Gill B, Levitt SB. Urolithiasis in children following augmentation cystoplasty. J Urol 1993; 150:726–729. 85. Nurse DE, McInerney PD, Thomas PJ, Mundy AR. Stones in enterocystoplasties. Br J Urol 1996; 77: 684–687. 86. Gonzalez R, Buson H, Reid C, Reinberg Y. Seromuscular colocystoplasty lined with urothelium: experience with 16 patients. Urology 1995; 45:124–129. 87. Jednak R, Schimke CM, Barroso U Jr, Barthold JS, Gonzalez R. Further experience with seromuscular colocystoplasty lined with urothelium. J Urol 2000; 164:2045–2049. 88. Cendron M, Gearhart JP. The Mitrofanoff principle. Technique and application in continent urinary diversion. Urol Clin North Am 1991; 18:615–621. 89. Kaefer M, Retik AB. The Mitrofanoff principle in continent urinary reconstruction. Urol Clin North Am 1997; 24:795–811. 90. Filipas D, Fisch M, Leissner J, Stein R, Hohenfellner R, Thuroff JW. Urinary diversion in childhood: indications for different techniques. BJU Int 1999; 84:897–904. 91. Bauer SB, Hendren WH, Kozakewich H, et al. Perforation of the augmented bladder. J Urol 1992; 148:699–703. 92. Couillard DR, Vapnek JM, Rentzepis MJ, Stone AR. Fatal perforation of augmentation cystoplasty in an adult. Urology 1993; 42:585–588. 93. Kryger JV, Gonzalez R, Barthold JS. Surgical management of urinary incontinence in children with neurogenic sphincteric incompetence. J Urol 2000; 163:256–263. 94. Herschorn S, Hewitt RJ. Patient perspective of long-term outcome of augmentation cystoplasty for neurogenic bladder. Urology 1998; 52:672–678. 95. Golomb J, Klutke CG, Lewin KJ, Goodwin WE, deKernion JB, Raz S. Bladder neoplasms associated with augmentation cystoplasty: report of 2 cases and literature review. J Urol 1989; 142:377–380. 96. Filmer RB, Spencer JR. Malignancies in bladder augmentations and intestinal conduits. J Urol 1990; 143:671–678. 97. Gregoire M, Kantoff P, DeWolf WC. Synchronous adenocarcinoma and transitional cell carcinoma of the bladder associated with augmentation: case report and review of the literature. J Urol 1993; 149:115–118. 98. Fernandez-Arjona M, Herrero L, Romero JC, Nieto S, Martin R, Pereira I. Synchronous signet ring cell carcinoma and squamous cell carcinoma arising in an augmented ileocystoplasty. Case report and review of the literature. Eur Urol 1996; 29:125–128. 99. Lane T, Shah J. Carcinoma following augmentation ileocystoplasty. Urol Int 2000; 64:31–32. 100. Lipke M, Schulsinger D, Sheynkin Y, Frischer Z, Waltzer W. Endoscopic treatment of bladder calculi in post-renal transplant patients: a 10-year experience. J Endourol 2004; 18(8):787–790. 101. Schwartz BF, Stoller ML. The vesicle calculus. Urol Clin North Am 2000; 27:333–346. 102. Teichman JM, Rogenes VJ, McIver BT, Harris JM. Holmium:yttrium-aluminum-garnet laser cystolithotripsy of large bladder calculi. Urology 1997; 50:44–48. 103. Grasso M. Experience with the holmium laser as an endoscopic lithotrite. Urology 1996; 48:199–206.

Complications of Bladder and Prostate Surgery

589

104. Palmer LS, Franco I, Reda EF, Kogan SJ, Levitt SB. Endoscopic management of bladder calculi following augmentation cystoplasty. Urology 1994; 44:902–904. 105. Bhatia V, Biyani CS. Vesical lithiasis: open surgery versus cystolithotripsy versus extracorporeal shock wave therapy. J Urol 1994; 151:660–662. 106. Husain I, el-Faqih SR, Shamsuddin AB, Atassi R. Primary extracorporeal shockwave lithotripsy in management of large bladder calculi. J Endourol 1994; 8:183–186. 107. Badlani GH, Douenias R, Smith AD. Percutaneous bladder procedures. Urol Clin North Am 1990; 17:67–73. 108. Ikari O, Netto NR Jr, D’Ancona CA, Palma PC. Percutaneous treatment of bladder stones. J Urol 1993; 149: 1499–1500. 109. Franzoni DF, Decter RM. Percutaneous vesicolithotomy: an alternative to open bladder surgery in patients with an impassable or surgically ablated urethra. J Urol 1999; 162:777–778. 110. Salah MA, Holman E, Toth C. Percutaneous suprapubic cystolithotripsy for pediatric bladder stones in a developing country. Eur Urol 2001; 39:466–470. 111. Elder JS. Percutaneous cystolithotomy with endotracheal tube tract dilation after urinary tract reconstruction. J Urol 1997; 157:2298–2300. 112. Elder JS, Peters CA, Arant BS Jr, et al. Pediatric Vesicoureteral Reflux Guidelines Panel summary report on the management of primary vesicoureteral reflux in children. J Urol 1997; 157:1846–1851. 113. Ellsworth PI, Cendron M, McCullough MF. Surgical management of vesicoureteral reflux. AORN J 2000; 71:498–505, 508–513, 517–520, 523–524. 114. Coplen DE, Duckett JW. The modern approach to ureteroceles. J Urol 1995; 153:166–171. 115. Engel JD, Palmer LS, Cheng EY, Kaplan WE. Surgical versus endoscopic correction of vesicoureteral reflux in children with neurogenic bladder dysfunction. J Urol 1997; 157:2291–2294. 116. Coplen DE, Barthold JS. Controversies in the management of ectopic ureteroceles. Urology 2000; 56:665–668. 117. Jayanthi VR, Koff SA. Long-term outcome of transurethral puncture of ectopic ureteroceles: initial success and late problems. J Urol 1999; 162:1077–1080. 118. Shekarriz B, Upadhyay J, Fleming P, Gonzalez R, Barthold JS. Long-term outcome based on the initial surgical approach to ureterocele. J Urol 1999; 162:1072–1076. 119. Chapple RW, Wein AJ, Brubaker L, et al. Stress incontinence injection therapy: what is best for our patients? Eur Urol 2005; 48(4):552–565. 120. Choe JM. Suprapubic sling adjustment: minimally invasive method of curing recurrent stress incontinence after sling surgery. J Urol 2002; 168(5):2059–2062. 121. Kelly MJ, Zimmern PE, Leach GE. Complications of bladder neck procedures. Urol Clin North Am 1991; 18:339–348. 122. Little NA, Juma S, Raz S. Vesicovaginal fistulae. Semin Urol 1989; 7:78–85. 123. Woo HH, Rosario DJ, Chapple CR. The treatment of vesicovaginal fistulae. Eur Urol 1996; 29:1–19. 124. Sweeney DD, Leng WW. Treatment of postoperative voiding dysfunction following incontinence surgery. Curr Urol Rep 2005; 6(5):365–370. 125. Stanton SL. Surgical treatment of sphincteric incontinence in women. World J Urol 1997; 15:275–279. 126. Parra RO, Boullier JA. Laparoscopic surgery for bladder carcinoma. Semin Surg Oncol 1996; 12:145–152.

590

Bell et al.

127. Anderson KR, Clayman RV. Laparoscopic lower urinary tract reconstruction. World J Urol 2000; 18: 349–354. 128. Fahlenkamp D, Rassweiler J, Fornara P, Frede T, Loening SA. Complications of laparoscopic procedures in urology: experience with 2,407 procedures at 4 German centers. J Urol 1999; 162:765–771. 129. Kavoussi LR, Sosa E, Chandhoke P, et al. Complications of laparoscopic pelvic lymph node dissection. J Urol 1993; 149:322–325. 130. Mendoza D, Newman RC, Albala D, et al. Laparoscopic complications in markedly obese urologic patients (a multi-institutional review). Urology 1996; 48:562–567. 131. Nordestgaard AG, Bodily KC, Osborne RW Jr, Buttorff JD. Major vascular injuries during laparoscopic procedures. Am J Surg 1995; 169:543–545. 132. McGinnis DE, Strup SE, Gomella LG. Management of hemorrhage during laparoscopy. J Endourol 2000; 14: 915–920. 133. Wolf JS Jr, Clayman RV, Monk TG, McClennan BL, McDougall EM. Carbon dioxide absorption during laparoscopic pelvic operation. J Am Coll Surg 1995; 180:555–560. 134. See WA, Cooper CS, Fisher RJ. Predictors of laparoscopic complications after formal training in laparoscopic surgery. JAMA 1993; 270:2689–2692.

135. Godec CJ. Genitourinary trauma. Urol Radiol 1985; 7:185–191. 136. Gomez RG, Ceballos L, Coburn B, et al. Consensus statement on bladder injuries. BJU Int 2004; 94(1):27–32. 137. Bodner DR, Selzman AA, Spirnak JP. Evaluation and treatment of bladder rupture. Semin Urol 1995; 13: 62–65. 138. Spirnak JP. Pelvic fracture and injury to the lower urinary tract. Surg Clin North Am 1988; 68:1057–1069. 139. Coburn M. Damage control for urologic injuries. Surg Clin North Am 1997; 77:821–834. 140. Baniel J, Schein M. The management of penetrating trauma to the urinary tract. J Am Coll Surg 1994; 178:417–425. 141. Carroll PR, McAninch JW. Major bladder trauma: mechanisms of injury and a unified method of diagnosis and repair. J Urol 1984; 132:254–257. 142. Corriere JN Jr, Sandler CM. Management of the ruptured bladder: seven years of experience with 111 cases. J Trauma 1986; 26:830–833. 143. Matthews LA, Herbener TE, Seftel AD. Impotence associated with blunt pelvic and perineal trauma: penile revascularization as a treatment option. Semin Urol 1995; 13:66–72. 144. Golomb J, Klutke CG, Raz S. Complications of bladder substitution and continent urinary diversion. Urology 1989; 34:329–338.

48 Urethral, Scrotal, and Penile Surgery Angelo E. Gousse and Robert R. Kester Department of Urology, Miller School of Medicine at the University of Miami, Miami, Florida, U.S.A. Patricia M. Byers Division of Trauma, Burns, and Critical Care, The DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

The avoidance, recognition, and management of complications in the lower genitourinary tract require a thorough understanding of the anatomy of the pelvis and genitalia. The focus in the female is primarily the avoidance of fistulae and incontinence. In the male, due to the shared function and location of the urethra, the focus is to maintain continence, fertility, and erectile function.

Injuries to the Posterior Urethra Presentation and Classification of Injuries Posterior urethral injuries are the most serious injuries to the lower urinary tract. Such injuries are generally caused by a severe external force, such as a highspeed blunt injury or a penetrating injury (1). The injury tears the attachments of the prostate, the puboprostatic ligaments, from the pelvic floor and often

URETHRA Anatomy of the Urethra The male urethra can be divided into four areas: the prostatic, membranous, bulbous, and penile or pendulous urethra (1). For the purposes of treatment, urethral injuries are classified as either posterior or anterior. Posterior injuries include prostatic and membranous injuries above or including the urogenital diaphragm, and anterior injuries affect the bulbous or penile urethra (Fig. 1). The female urethra consists of a 4-cm tube of inner epithelium surrounded by an outer muscularis layer (2). The muscularis includes both smooth muscle in continuity with the trigonal musculature and striated muscle oriented circularly. The circular striated muscle fibers are most prominent in the middle-third of the urethra. As is true of the male urethra, the female urethra contains an inner mucosal layer composed of transitional cells at the bladder neck and squamous cells at the meatus. Additionally, the female urethra has posterior and anterior portions; the demarcation between the two portions is the urogenital diaphragm. The region above the urogenital diaphragm is the posterior portion, and the region below the diaphragm is the anterior portion.

Figure 1 Male genitourinary system. 1, bladder; 2, pubic bone; 3, penis; 4, corpus cavernosa; 5, glans penis; 6, prepuce; 7, urethral opening; 8, sigmoid colon; 9, rectum; 10, seminal vesicle; 11, ejaculation duct; 12, prostate gland; 13, cowper gland; 14, anus; 15, vas deferens; 16, epididymis; 17, testicle; and 18, scrotum.

592

Gousse et al.

tears the urethra. Pelvic fractures cause more than 90% of traumatic urethral ruptures. Patients may experience an inability to urinate. Physical examination rarely demonstrates substantial genital swelling because neither Buck fascia nor Colles’ fascia is violated. Most patients experience gross hematuria, but the absence of this symptom does not rule out serious injury. When the prostate has been completely disrupted, digital rectal examination will disclose a boggy fluid collection, composed of blood and urine, in the normal location of the prostate. In such cases, the prostate and bladder will ascend above their normal anatomic location. When a patient is uninjured and is in stable condition, urologic diagnostic testing is directed by the type of suspected injury: external injuries are detected first, followed by urethral injury, bladder injury, and finally ureteral and renal injury (3). Retrograde urethrography (RUG) is indicated whenever urethral injury is suspected. Initial anteroposterior radiographs of the pelvis will demonstrate any associated pelvic fractures. Injection studies are performed after 20 to 25 mL of a water-soluble contrast agent has been injected into the male urethra while the patient is in a 25 to 30 oblique position. The injection is facilitated by holding the tip of the injection syringe in the penile meatus, inserting a small Foley catheter no deeper than the fossa navicularis with the balloon partially inflated with 1 to 2 mL of water, or using a specialized clamp such as the Brodney clamp. When properly performed, RUG will clearly diagnose injuries to the male urethra. Such injuries are classified by types. Type I injuries, the mildest, involve elongation of the prostatic urethra without actual rupture. Type II injuries involve partial or complete rupture of the prostatomembranous urethra, with extravasation of contrast material confined below the urogenital diaphragm, as demonstrated by RUG. Type III injuries, the most common posterior urethral injuries, involve partial or complete rupture of the prostatomembranous urethra and disruption of the urogenital diaphragm. In such cases, RUG demonstrates extravasation of contrast material above the urogenital diaphragm, in the pelvis or the peritoneal cavity (4). Posterior urethral obliterative stricture is a late complication associated with trauma, transurethral resection of the prostate, radical retropubic prostatectomy for prostate cancer, or radiation therapy for either prostate or bladder cancer. Clearly, such injuries occur much more frequently among men than among women. When they do occur, their diagnosis may be more difficult when only conventional radiographic contrast studies such as RUG, voiding cystourethrography (VCUG), and double-balloon catheter urethrography are used. Such studies do not clearly demonstrate the anatomy of the posterior urethra or the anatomic derangement of adjacent structures. Because the anatomic details of both the urethra and the periurethral tissues can be evaluated noninvasively with magnetic resonance (MR) imaging, this

method can be used as an adjunctive tool for evaluation of urethral abnormalities (5). In cases of female urethral trauma, MR imaging is helpful in assessing the presence and the extent of anterior or posterior urethral injury and injury to adjacent structures.

Treatment For Type I injuries, urethral Foley catheterization is recommended for three to five days so that acute urinary retention can be avoided. These injuries will resolve spontaneously. The management of Type II or Type III posterior urethral injuries remains controversial and difficult. Initial goals should include designing a treatment plan that will minimize the long-term complications of urethral stricture, incontinence, and, for male patients, erectile dysfunction. For less severe injuries, when possible, urethral Foley catheterization is used for 10 to 14 days. VCUG is performed before the catheter is removed. For more serious injures, particularly those for which placing a Foley is impossible, the standard of care has become suprapubic catheter placement and delayed urethral reconstruction (6). However, primary repair may be advocated in certain situations, such as when there is concomitant vascular or rectal injury or when there is severe prostatic laceration or dislocation (7,8). In addition, some authors advocate passing a urethral catheter over a guidewire at the time of surgery by cystoscopic techniques. This procedure has been shown to reduce the incidence of delayed stricture formation that requires formal urethroplasty from 89% in cases managed by suprapubic catheter alone to 23% in cases managed with cystoscopically guided catheter placement (4). For female patients, a recent study (9) advocates early drainage via cystostomy and deferred surgical reconstruction when immediate surgical repair is precluded by life-threatening clinical conditions or extensive traumatized tissue in the affected area (9). For children, initial treatment with open vesicostomy has been advocated (10). Delayed repair of disruption of the male posterior urethra is carried out after a minimum of six to eight weeks of suprapubic drainage. The repair is made via a midline perineal approach and involves isolation and removal of the diseased segment and direct endto-end suturing of the distal urethral segment to the membranous urethra. Patients are counseled about the possibility of postoperative incontinence and erectile dysfunction. Although delayed open repair by perineal urethroplasty is the standard of care, short strictures (less than 3 cm) can be managed by cystoscopic incision and postoperative urethral self-dilatation (11,12).

Injuries to the Anterior Male Urethra Presentation and Diagnostic Imaging Injuries to the anterior male urethra are more common than posterior urethral injuries. They commonly

Chapter 48: Urethral, Scrotal, and Penile Surgery

result from straddle injuries of the perineum, such as those caused by falling astride a fence rail or a bicycle crossbar (7). Unlike posterior urethral injuries, anterior urethral injuries are seldom caused by pelvic fractures. Less common causes of anterior urethral injury are associated penile fractures during sexual intercourse, gunshot and other penetrating wounds, iatrogenic causes, and self-mutilation. Urethral disruption contained by Buck fascia causes a sleeve-like swelling of the penile shaft, with no swelling of the scrotum. More extensive penile extravasation can pass beyond Buck fascia and be limited by Colles’ fascia; in such cases, the classic perineal butterfly hematoma configuration occurs (1). The mainstay of diagnosis is RUG.

Treatment The management of blunt anterior urethral contusions without laceration involves observation and urethral Foley catheter drainage for three to five days. If any laceration is present, however, immediate surgical exploration and primary repair are indicated, whether the injury is blunt or penetrating in origin. For proximal injuries, a perineal approach is favored. For injuries distal to the proximal pendulous urethra, a circumcision incision is created and the penile shaft skin is degloved proximally. A partial injury is surgically repaired over a urethral Foley catheter that is left in place for 7 to 10 days. For complete injuries, direct end-to-end anastomosis is preferred and the catheter is left in place for 10 to 14 days. Before the catheter is removed after either type of repair, urethrography is performed alongside the catheter, so that additional extravasation can be excluded.

Reconstructive Surgery of the Male Urethra Anterior Urethral Strictures Urethral stricture disease refers to the formation of scar tissue within the spongy erectile tissue of the anterior urethra (13). Such strictures may be caused by any disease process of this spongy tissue or of the urethral epithelium overlying it. Although these strictures are most commonly caused by straddle or iatrogenic injuries due to instrumentation, they may also be caused by infection or may be idiopathic. The diagnosis of anterior urethral stricture begins with clinical suspicion when a male patient exhibits obstructive voiding symptoms, particularly if urinary infection is present. RUG and cystoscopy are indicated. With the patient under anesthesia, the stricture is gently dilatated and the urethra proximal to it is examined. This evaluation is important for appropriately staging the extent of the disease because missed proximal strictures can result in postoperative recurrent scarring proximal to the original stricture. In such cases, the proximal scar was present but was held open hydrostatically by the dilatation proximal to the original stricture.

593

Urethral strictures are treated surgically. Urethral dilatation is the oldest and simplest treatment for strictures. The use of metal urethral sounds or balloon dilators has been successful. Another option is internal urethrotomy, which involves transurethral incision of the stricture, most commonly with a cold endoscopic Collins knife. The goal of therapy after any of these approaches is the maintenance of the enlarged urethral lumen after complete healing. This goal may be accomplished by leaving an indwelling urethral Foley catheter in place for three days and then requiring a program of selfcatheterization for three to six months. The most common surgical complication of operative treatment is recurrent stricture. The stricture recurs after 65% to 80% of internal urethrotomy procedures (14,15). Because of the high rate of stricture recurrence associated with urethrotomy, alternative energy sources have been sought, including various types of lasers such as the CO2 laser, the potassium titanyl phosphate laser, the neodymium yttriumaluminum-garnet laser, the argon laser, and the holmium: yttrium-aluminum-garnet laser. However, the results of laser therapy have been disappointing (13). The most dependable technique for repairing anterior urethral strictures is excision and primary reanastomosis. This technique, which is suitable for shorter strictures of 1 to 2 cm, depends on adequate urethral mobilization and tension-free suturing. Durable success for more than one year has been reported in as many as 95% of cases (16). For longer strictures, free grafts have been successful. Tissues that can be used for such grafts are full-thickness and partial-thickness skin, bladder epithelium, and buccal mucosa. Tubularized-free grafts should be avoided because they are associated with a high rate of stricture recurrence. Another treatment option for longer, more severe strictures is a two-staged repair. During the first stage, a urethral plate is constructed by using a meshed split-thickness skin graft. At a later date, another surgical procedure is performed for tubularizing the graft, thereby forming a neourethra (17). An alternative to skin grafts is the genital skin island flap based on the dartos fascia of the penis or on the tunica dartos of the scrotum (18). The success rate associated with onlay island flaps is higher than that associated with either tubularizedfree grafts or island procedures.

Urinary Incontinence Urinary incontinence is a well-recognized and serious health problem for 3% to 60% of patients after radical prostatectomy (19–21). The cause of the incontinence is the radical removal of the passive continence mechanism with its internal sphincter at the level of the bladder neck. This mechanism is removed as part of the surgical specimen, and there is an associated loss of cooptation of the membranous urethra. Historically, such incontinence has been treated with passive and

594

Gousse et al.

active surgical procedures, culminating in the artificial urinary sphincter (AUS), first developed by Kaufman in 1972 as a silicone gel prosthesis (22,23). The Artificial Urinary Sphincter

The AUS is an implantable prosthetic device that is placed surgically and can restore urinary control for patients with sphincteric incontinence. Via an open abdominal approach, the AUS is placed either circumferentially around the urethra at the level of the bulbar urethra or around the bladder neck. A control valve is placed in the scrotum and a reservoir is placed posterior to the rectus abdominis musculature (24). The AUS currently in use is the American Medical Systems (AMS, Minnetonka, Minnesota, U.S.) AMS 800TM, first introduced in 1982. Compared with its predecessor, the AMS 791TM, it incorporates improved features. These features include a control valve and a reservoir as a single component, and a narrow-backed cuff design that decreases the incidence of tissue pressure atrophy and cuff erosion. The AMS 800 is well tolerated and offers men who are incontinent after prostatectomy a reasonable chance for obtaining long-term urinary control. Satisfactory continence is reported by 73% to 83% of men after device implantation (25,26). A recent telephone survey of patients found that the narrow-backed design was associated with a statistically significant decrease in the need for surgical revision due to cuff erosion (27). Congenital and neurogenic conditions may also cause incontinence among men, women, and children. As long as the bladder volume is adequate, an AUS is indicated for any case of sphincteric incontinence. In cases that involve reduction in bladder volume, concomitant procedures may be required for augmenting bladder volume. For men, the cuff is placed at the bulbar urethra or the bladder neck; for children and women, the cuff is placed only around the bladder neck. For children, an open abdominal approach is recommended for placing the cuff around the vesical neck. For women, either an abdominal or a transvaginal approach can be used for placing the cuff around the vesical neck (24). The cuff is left deactivated in the open position for six to eight weeks postoperatively. The cuff is then activated in the physician’s office by the application of firm pressure that compresses the pump. The deactivation pin ‘‘pops’’ into the activated position; its positioning can be monitored by radiography because diluted contrast agent is used as the internal fluid. After activation, patients are taught how to cycle the device by compressing the pump, thereby opening the urethral cuff for voiding. After three to five minutes of micturation, the cuff will automatically refill, closing the urethra and restoring continence. Complications of the Artificial Urinary Sphincter

The AUS may be associated with complications. Hematomas may occur and, when large, may require

surgical drainage. In addition, the patient may experience early urinary retention during the immediate postoperative period. In such cases, it is necessary to confirm that the pump mechanism is in the open, deactivated position. If the incontinence is not due to a mechanical problem with the device, the patient may require gentle, intermittent urethral catheterization with a 10- or 12-Fr straight or Coude catheter. Delayed urinary retention due to recurrent bladder neck contracture is also possible. Infections may occur days, months, or years after the insertion of an AUS despite the appropriate use of perioperative antibiotics. Early infections are commonly due to skin flora or airborne organisms, whereas late infections are more likely to be due to gram-negative bacteria of urinary tract origin. After an AUS has been placed, patients will require the administration of prophylactic antibiotics before undergoing any surgical, urological, or dental procedure, so that the risk of bacterial seeding can be decreased. Unfortunately, treating a prosthetic infection almost always requires removal of the device; reimplantation can be attempted three to six months later, although the likelihood of infection or cuff erosion is increased in such cases (24). Cuff erosion may also occur either immediately or as a delayed complication; however, this condition most commonly occurs three to four months postoperatively. Symptoms of cuff erosion are pain and swelling of the perineum or scrotum, urinary infection, bloody discharge, and recurrent urinary incontinence. When cuff erosion is detected during the immediate postoperative period, the most likely cause is inadvertent and unrecognized iatrogenic injury at the time of surgery. The use of properly selected low-pressure and narrow-backed cuffs has lessened the incidence of the delayed form of this complication. Persistent or recurrent incontinence may occur in association with an AUS. This complication is most often due to mechanical failure, tissue atrophy, unrecognized bladder urodynamic instability, cuff infection, or cuff erosion. The five-year reliability rate of the AMS 800 device is estimated at more than 90%. Mechanical failure due to fluid loss can be verified by radiography, although surgical exploration is necessary for locating the site of leakage. According to Barrett and Licht, the most common sites of leakage are the cuff, the tubing connectors, and the pump (24). Tissue atrophy is another important cause of postoperative incontinence; this condition can be diagnosed by urodynamic studies and cystoscopy. Tissue atrophy may be treated by reoperation for increasing the pressure of the balloon reservoir pressure or for decreasing the size of the cuff, initially by 0.5 cm. Postoperative incontinence can also be caused by failure to recognize involuntary bladder contractions before an AUS is placed. Urodynamic studies are crucial for diagnosing this problem. Contractions may then be managed with pharmacologic therapy. However, if the problem is poor bladder

Chapter 48: Urethral, Scrotal, and Penile Surgery

compliance, surgery may be necessary for increasing bladder capacity.

Reconstructive Surgery of the Female Urethra Pelvic Anatomy The pelvic diaphragm is divided into the coccygeus muscles and the levator ani with its pubococcygeus, iliococcygeus, and ischiococcygeus muscles. These muscles form the primary inferior support for the urethra, vagina, and rectum, all of which pass through a hiatus formed by the pubococcygeus muscle. The lateral margins of these muscles are formed by the arcus tendineus, which extends from the posterior-inferior pubic ramus to the ischial spines. The endopelvic fascia covers the pelvic organs and is composed of an intrapelvic abdominal leaf and an extrapelvic vaginal leaf, which come together laterally and fuse into a common insertion along the arcus tendineus (28). The levator fascia is central in the pelvic floor and consists of four specialized condensations: the pubourethral ligament, which anchors the urethra to the inferior pubic ramus; the urethropelvic ligaments, which attach the urethra and bladder neck to the arcus tendineus; the vesicopelvic fascia, which attaches the bladder base to the arcus tendineus and the pelvic sidewall; and the cardinal ligaments, the most posterior condensation of levator fascia, which are continuous with the vesicopelvic fascia and attach the uterine isthmus to the lateral pelvic wall. The cardinal ligaments are important because their laxity can lead to cystocele, uterine prolapse, or enterocele when the uterus has been surgically removed. Adequate surgical correction of any weakness of the cardinal ligament is required concomitant with grade IV cystocele repair so that recurrent pelvic floor prolapse can be prevented. The three main support structures for the uterus are the cardinal ligaments, the sacrouterine ligaments, and the broad ligaments. The sacrouterine ligaments are located posteriorly and attach the cervix contiguous with the cardinal ligaments to either side of the sacrum. The broad ligaments are two superior peritoneal folds that contain the fallopian tubes, the round ligaments, the ovarian ligaments, and the ovarian vessels.

Stress Urinary Incontinence Clinical Manifestations

Stress urinary incontinence among women is defined as the spontaneous loss of urine upon maneuvers such as coughing or straining that allow for the transmission of abdominal pressure within the bladder. The cause may be urethral hypermobility or intrinsic sphincter deficiency. Stress urinary incontinence should be corrected only when it adversely affects the patient’s daily activities, personal hygiene, social interactions, financial status, or psychological wellbeing (29–32).

595

Corrective surgery for stress urinary incontinence among women is performed via either a vaginal or a retropubic approach and requires a comprehensive plan for pelvic reconstruction. Preoperative considerations are the type and severity of the stress incontinence, the degree of any associated cystocele, and the presence of other pelvic floor abnormalities, including enterocele, uterine hypermobility, or rectocele (28). Severe grades of cystocele can paradoxically cause urinary retention because of the hyperacute change in the vesicourethral angle. Vaginal Repair of Stress Urinary Incontinence

When the vaginal approach is used for the surgical repair of stress urinary incontinence, the surgeon can also repair an associated enterocele or rectocele or can perform a vaginal hysterectomy, as indicated. In addition, complex pelvic prolapse can be corrected, either by sacrospinal fixation or by vaginal sacrocolporrhaphy (McCall culdoplasty procedure). Currently, vaginal repair is an area of evolving clinical development; many types of repairs are possible, including anterior colporrhaphy, an anterior vaginal wall sling, a pubovaginal sling (PVS) incorporating autologous or cadaveric fascia, or porcine dermal grafts (Pelvicol1) (33–36). In addition, synthetic materials such as tension-free vaginal tape (TVT) and SPARCTM have been fashioned into grafts that are placed suburethrally via minimally invasive techniques (37,38). TVT is placed with vaginal needles, whereas SPARC is placed with retropubic needles. The advent of newer techniques has seen a decline in the frequency of bladder neck suspension procedures such as the Peyrera, Stamey, and Raz procedures (39,40). The overall cure rate of 67% and the improvement rate of 82% associated with these older procedures are inferior to those associated with the PVS procedure or the retropubic approaches. One study found that for PVS procedures the rate of cure is 83% and the rate of improvement is 87% after 48 months (41). Another found that for TVT the cure rate is 85% and the improvement rate is 96% at 56 months (42). Long-term data for SPARC are not yet available. In addition to the low cure and improvement rates cited above, complications may also be associated with any procedure used to suspend the female urethra. Patients may develop pain, bleeding, infection, de novo urgency incontinence, prolongation of urinary retention, and secondary prolapse such as enterocele. A theoretical advantage of a synthetic material, such as TVT, is the ease with which it can be released if urinary retention is prolonged. In one study of TVT, the urinary retention rate was 3%; these cases resolved, with complete bladder emptying by 100% of the patients, after surgical release of the TVT (43). Retropubic Repair of Stress Urinary Incontinence

Although its long-term results compare favorably with those of PVS techniques, open retropubic urethropexy

596

Gousse et al.

is a much more invasive procedure. It can be performed either as a Marshall-Marchetti-Kranz procedure or as a Burch procedure (44,45). For open retropubic urethropexy, the rate of cure is 84% and the rate of improvement is 90% for more than two years (41). As is true of PVS procedures, open retropubic urethropexy may be complicated by pain, bleeding, infection, de novo urgency incontinence, secondary prolapse such as enterocele, and prolongation of urinary retention. Patients who experience a recurrence of urinary incontinence should undergo urodynamic studies. If a postoperative intrinsic sphincter deficiency is found, an attempt at surgical repair is indicated (46). Repair of Urethrovaginal Fistula

The spectrum of anatomic defects in the female urethra ranges from small urethrovaginal fistulas, which cause vaginal voiding, to the loss of the entire urethra and bladder neck, which causes total incontinence (47). The most common cause of urethrovaginal fistula is previous gynecologic or urologic surgery; anterior colporrhaphy and urethral diverticulectomy are the most common antecedent procedures (48,49). Preoperative evaluation includes a thorough history and physical examination and an evaluation of the extent of urethral loss. Careful cystoscopy usually confirms the diagnosis. The extent of urethral loss will dictate the type of procedure needed for surgical repair. Distal fistulae beyond the external sphincter can be managed with simple excision. Extensive loss or very large fistulae will require complete urethral reconstruction. Anti-incontinence procedures are often performed simultaneously with extensive repairs. The operation is begun by placing the patient in the lithotomy position. A 14-Fr urethral Foley catheter and a 24-Fr suprapubic Foley catheter are placed with a Lowsley tractor clamp. An inverted ‘‘U’’ incision is created, the fistula is circumscribed, and the scarred margins are used to provide a secure closure of the fistula tract. A vaginal advancement flap is used to prevent an overlapping suture line. When there is concern about the quality of the vaginal repair because of previous irradiation, a Martius labial fat pad may be used to bolster the repair (50). Large defects of the female urethra present a challenging urological problem. An abdominal approach allows omental interposition and is preferred when ureteral implantation is required. The usefulness of the vaginal approach has been well described and has the advantage of allowing closure of all urethral fistulas without the attendant risks of major abdominal surgery (48–55). With urethral loss, the preferred approach is the use of vaginal flaps for transvaginal reconstruction of the urethra. Again, the patient is placed in the lithotomy position. An inverted ‘‘U’’ incision is created by making two parallel incisions on either side of the meatus and

extending them distally. Flaps are mobilized laterally and then medially; such mobilization allows tubularization of the neourethra around a 14-Fr catheter. The suture line is reinforced with a Martius graft, and the vagina and labia are closed. If a PVS was previously performed for incontinence, it must be secured before the vaginal mucosa is closed. Postoperatively, the Foley catheter is kept in place for 7 to 10 days. The catheter is then removed and VCUG is performed. If extravasation or retention is noted, a suprapubic catheter is placed for one to two weeks of drainage. Complications of urethrovaginal fistula repair include elevated urinary residuals and urinary retention. Urinary retention is more common when a concomitant PVS procedure is performed. Prolonged drainage with a suprapubic catheter or long-term intermittent catheterization may be necessary. Another potential complication is urinary incontinence as a consequence of stress incontinence, detrusor instability, or urethral fistula recurrence. If stress urinary incontinence occurs, it should be treated in the standard fashion. Some patients may benefit from periurethral injection of collagen (56). Repair of Urethral Diverticulum

The diagnosis of urethral diverticulum is often overlooked for women with lower urinary tract symptoms (57,58). The mean age of women with this condition is 45 years. The population incidence of urethral diverticulum among adult women ranges from 1.4% to 5% (59,60). Many cases are asymptomatic. The most widely held theory is that these lesions are acquired, and they probably develop from infected periurethral glands (58,61). Urethral diverticula may be complicated by infection, bladder outlet obstruction, paradoxical urinary incontinence, and malignancy, most commonly adenocarcinoma. Periurethral stones form in 1% to 10% of cases (62). The common presentation is referred to as the three Ds: dysuria, dribbling, and dyspareunia. Frequency, urgency, and dysuria are the most common symptoms, occurring in approximately 50% of cases (47). Because these symptoms commonly result in visits to the urologist’s office, urethral diverticulum should be considered whenever a patient’s symptoms persist and do not respond therapy. In 63% of cases, the diagnosis can be made with a careful physical examination (63). In some cases, tenderness will be found without a palpable mass. Preoperative evaluation should include cystoscopy performed with a 0 lens. VCUG is the most helpful imaging study and will be useful in diagnosing 95% of cases (47). Intravenous pyelography should be performed preoperatively for ruling out the diagnosis of ectopic ureterocele (64). Urodynamic testing is indicated for any woman with symptoms of stress urinary incontinence because these symptoms occur among 72% of patients with urethral diverticulum (58). If stress incontinence is present, it should be

Chapter 48: Urethral, Scrotal, and Penile Surgery

treated with a simultaneous anti-incontinence procedure at the time of diverticulectomy. Multiple surgical procedures have been developed for treating urethral diverticulum. One method, transurethral saucerization, is generally reserved for distal diverticula (65). Transvaginal repair with excision of the diverticulum and the use of vaginal flaps is another recommended approach (66,67). With the patient in the dorsal lithotomy position, an 18-Fr suprapubic Foley catheter is placed with a Lowsley tractor clamp. A 12-Fr urethral Foley catheter is also placed. An inverted ‘‘U’’ incision is made in the anterior vaginal vault, with its apex distal to the urethral meatus. The flap is raised and the diverticulum is dissected circumferentially. The diverticulum sac is freed and excised by sharp dissection. At this point, the catheter is seen in the lumen of the urethra. The operculum of the diverticulum is closed in a longitudinal fashion over the urethral Foley catheter, and the periurethral tissues are closed transversely. At this point, if the quality of the repair is questionable, a Martius labial graft can be used between the periurethral and vaginal wall layers. The final vaginal wall layer is closed by reapproximating the inverted ‘‘U’’ incision. Postoperatively, the Foley catheter is left for drainage for 10 to 14 days, after which time VCUG is performed. The suprapubic catheter is removed when the patient can void successfully. Postoperative complications include pain, bleeding, and infection. When bladder spasms occur, anticholinergic medications are indicated. In addition, stress incontinence may be unmasked postoperatively. Urethral strictures may occur if too much urethral wall is removed, particularly in the dissection of a large diverticulum. Recurrent diverticulum and urethrovaginal fistula formation are recognized late complications.

SCROTUM History and Physical Examination The adult scrotum is a loose sac containing testicles and spermatic cord structures (Fig. 1). Swelling can occur as the result of traumatic, infectious, or other inflammatory conditions. Pain can be local in the testes, the epididymides, or the scrotum itself, or it can be referred in origin. Referred pain can result from an incarcerated inguinal hernia or from colic caused by the passage of a renal calculus through the ureter. Evaluating and treating chronic orchalgia may be difficult (68). During scrotal examination, the physician should palpate the testicles, epididymides, and spermatic cords. This examination can also detect an inguinal hernia. If a scrotal mass is present, transillumination is helpful in determining whether the mass is solid or cystic. A solid mass is consistent with a diagnosis of tumor, whereas a cystic mass is consistent with a diagnosis of hydrocele or spermatocele (69).

597

Scrotal Conditions Scrotal Injury The scrotum is susceptible to trauma because of its external position (13). Degloving injuries of the scrotum occur when the skin is trapped and stripped from its underlying structures; such injuries are commonly associated with like injuries to the penis. With partial injuries, the remaining scrotal skin can be mobilized for quite a distance for use in coverage (1). Although some complete degloving injuries require burying the testicles in the thigh and later performing skin grafting, most of these injuries can be managed with immediate or delayed split-thickness skin grafting. For burn injuries, the ability to reconstruct the damage depends on how well the normal structures have been maintained. In cases in which the urethra is also burned and the phallus’s integrity is severely compromised, a gracilis myocutaneous flap transfer with staged skin grafting or a complete phalloplasty may be required.

Infection The scrotum contains sebaceous and apocrine sweat glands that are susceptible to infection. In addition, the moist intertriginous space between the scrotum and the medial thigh promotes tinea cruris infections. These infections are readily treated with local care and antimicrobial agents. Necrotizing fasciitis or Fournier’s gangrene is a serious infection of the male genitalia, usually the scrotum and perineum. First described in 1883, it was initially characterized as a fulminating genital gangrene of idiopathic origin among healthy young patients (70). Today it primarily affects older patients. In 95% of cases, a predisposing factor can be identified, such as diabetes mellitus, local trauma, periurethral extravasation of urine, paraphimosis, or perirectal or perianal infections. It is also a complication of surgical procedures such as circumcision or herniorrhaphy (71). Patients often have a history of recent perianal trauma, urological instrumentation, urethral stricture associated with a sexually transmitted disease, or urethrocutaneous fistula. The infection commonly starts as a cellulitis near the site of entry. Fournier gangrene is a true urological emergency. Early in the course of the infection, pain, fever, and toxicity occur as deeper tissues become involved. Skin crepitus and necrosis rapidly ensue. Laboratory findings often point to a septic cause; the diagnosis is confirmed by radiography or ultrasonography, which will show subcutaneous air in the scrotum and adjacent areas of the abdominal wall and thighs (Fig. 2). Successful treatment depends on prompt diagnosis. The presence of toxicity out of proportion to the clinical findings should raise clinical suspicion of the diagnosis of Fournier gangrene. Intravenous fluids and antibiotics are given in preparation for

598

Gousse et al.

saline solution. Failure to close the tunica albuginea will result in the continued extrusion of seminiferous tubules. A small Penrose drain is placed beneath the tunica vaginalis for 24 to 36 hours, and antibiotics are administered for preventing postoperative infection.

Orchiectomy The reasons for orchiectomy are surgical treatment of testicular tumor, torsion, abscess, or advanced prostate cancer. The therapeutic benefits of bilateral orchiectomy and androgen ablation for metastatic prostate cancer were first recognized by Huggins and associates at the University of Chicago (79). For the excision of testicular tumor, an inguinal incision is preferred; for cases of torsion, abscess, or prostate cancer, a scrotal incision is preferred. The primary complications after orchiectomy are bleeding and infection.

Hydrocele Figure 2 Radiograph showing subcutaneous emphysema in a patient with Fournier gangrene.

surgery. Wound cultures generally yield multiple organisms, and this finding indicates anaerobic– aerobic synergy (72). Immediate and extensive wide debridement is essential, and the incision should extend beyond the areas of involvement. Often, a suprapubic catheter is left in place. Postoperatively, aggressive fluid and antibiotic management is mandatory. A second procedure is almost always necessary within 24 to 48 hours of the first procedure. Some authors have reported that hyperbaric oxygen therapy is beneficial (73). Even with maximum medical and surgical intervention, the overall mortality rate is 20%, ranging from 7% to 75% (72,74,75). The fatality rates are higher among patients with diabetes or alcoholism and those for whom treatment has been delayed.

TESTICLE Anatomy The testicles are paired structures located in the scrotum. Each is invested in a vestige of the parietal and visceral peritoneum, which forms the tunica vaginalis testis. The testicle receives its blood supply from vessels within the spermatic cord structures, including the gubernaculum.

Traumatic Injury Testicular rupture can be caused by either blunt or penetrating injury (76,77). Physical examination reveals swelling, tenderness, and signs of hematoma. Ultrasonography can be performed if the diagnosis is in doubt, but its sensitivity is only 64% and its specificity is only 75% (78). Immediate exploration is indicated; any extruded or necrotic tissue should be debrided and irrigated with copious amounts of warm

A hydrocele is a fluid collection within the tunica vaginalis testis of the testicle. This condition is not uncommon among newborn boys, especially when the infant is premature; it is caused by a patent processus vaginalis of peritoneum. Generally, in such cases, the hydrocele will resolve spontaneously by the end of the first year of life (80). Hydrocele may also occur after inguinal hernia repair, varicocele repair, or vasectomy; it may also occur spontaneously later in life. Communicating hydroceles occur when there is a persistent communication between the peritoneum and the tunica vaginalis testis. Surgical repair is indicated in all of these cases. An inguinal incision with high ligation of the hydrocele tract is the preferred technique for repairing a communicating hydrocele. For hydrocele without communication, a scrotal approach with a plication or bottle technique is preferred. Complications associated with surgical repair of hydrocele are bleeding, infection, and recurrence. The most common complication is hematoma, which is minimized by meticulously suturing all of the raw edges of the hydrocele sac and draining the scrotum whenever necessary. Inadvertent injury of adjacent scrotal structures may also occur and may adversely affect male fertility (81).

Varicocele The spermatic veins that drain the testis can dilate to form a varicocele. This process nearly always occurs on the left side, presumably because of incompetence of the spermatic vein as it drains into the left renal vein. This mass of veins has been described as a ‘‘bag of worms.’’ Varicocele is graded along a scale from 0, which is a subclinical grade, through 4, in which the varicocele is very prominent on visual inspection. Normally, blood flows through either spermatic vein into a low-pressure venous system; thus, a varicocele will almost always decrease or

Chapter 48: Urethral, Scrotal, and Penile Surgery

disappear when the patient is placed in the supine position. In contrast, when the varicocele is caused by obstruction of venous drainage, as is the case with left renal cell cancer with renal vein extension, the vein will not drain and there will be no change when the patient is placed in the supine position. Varicocele can be associated with infertility and is the most common surgically correctable cause of male infertility (82). Complications associated with varicocele repair are bleeding, infection, and recurrence. The most common complication is the formation of a secondary hydrocele because of lymphatic obstruction (83). The intraoperative use of magnification can reduce the risk of this complication (84). Another serious complication is injury or inadvertent ligation of the testicular artery, which may result in testicular atrophy and infertility.

EPIDIDYMIS AND VAS DEFERENS Anatomy The vas deferens is the main secretory duct of the testicle. Sperm cells from the testicle pass through a collection area called the epididymis and then travel via the vas deferens to the seminal vesicles, entering the ejaculatory duct in the prostatic urethra (Fig. 1).

Traumatic Injury Direct trauma to the epididymis and the vas deferens can occur at the same time as injury to the testicle. Generally, in cases of epididymal injury, precise reanastamosis of the microtubular structures is not possible. Optimal surgical care consists of appropriate debridement, control of bleeding, and reapproximation of tissue edges. Neither vasoepididymostomy, which connects the vas deferens to the epididymis, nor vasovasostomy, which reconnects the vas deferens, is recommended in a contaminated field. Delayed repair is performed several months later with the goal of a precise, unobstructed, water-tight repair.

Vasectomy Bilateral vasectomy is an office-based procedure for creating male sterility. Traditionally, two small incisions are made, and the right and left vas deferens are delivered through the ipsilateral incisions. A recent advance in the technique of vasectomy is the no-scalpel technique (85). Bleeding and infection can occur after vasectomy. Hematoma is the most common complication; its average incidence is 2% (86). The surgeon’s experience is the most important factor related to the postoperative occurrence of hematoma. Sperm granuloma is caused by an inflammatory response to sperm leaking from the cut end of the vas deferens or from sperm extravasating from the rete testes as the result of an increase in intratubular pressure. Long-term complications associated with vasectomy are chronic

599

epididymal pain and testicular pain. Antisperm antibodies develop among 60% to 80% of patients, but no clinical disease has been documented as associated with these antibodies (87).

PENIS Penile Anatomy The penile shaft is composed of three erectile bodies: two corpora cavernosa and one corpora spongiosum, which contains the urethra (Fig. 1). Covering these structures are fascial layers, blood vessels, lymphatic vessels, nerves, and skin. One of the fascial layers consists of Buck fascia, which encloses the deep dorsal structures of the penis. The other important fascial layer is Colles’ fascia, which makes up the anterior triangle and attaches to the perineum at the perineal body. The cavernosal artery supplies blood for erections and arises from the common penile artery, a branch of the internal pudendal artery. Venous drainage follows the arterial supply. The lymphatic vessels generally go deep to Buck fascia dorsally and drain into the deep inguinal lymph nodes. Somatic innervation is supplied by the pudendal nerves (motor and sensory). Autonomic innervation is supplied by the pelvic plexus, which consists of preganglionic afferent parasympathetic fibers from the sacral center (S2-S4) and preganglionic and visceral afferent fibers from the thoracolumbar center (T11–L2) (13). The skin includes the prepuce that forms a fold over the glans penis.

Penile Trauma Injuries to the penis are classified as superficial, deep, or degloving. Superficial injuries include those to the skin and subcutaneous structures such as lacerations and thermal injuries. Deep injuries occur when structures deep to Buck fascia are injured and include deep lacerations, gunshot wounds, and spontaneous rupture of the corpora cavernosa that may occur during sexual intercourse. Degloving injuries occur when the skin of the penis is caught in machinery; degloving injuries to the penis often occur at the same time as degloving injuries to the scrotum (1). Treatment generally includes operative debridement, copious irrigation with saline solution, and meticulous reapproximation of healthy tissues. Often, it is wise to place a suprapubic tube for ensuring proper bladder drainage. Deep injuries may result in erectile dysfunction. Degloving injuries will often require staged procedures. The penis can be buried into healthy subcutaneous tissue or covered with split-thickness skin grafts, either during the initial surgical procedure or subsequently.

Circumcision Circumcision, the removal of the prepuce, is generally performed for social or religious reasons. Circumcision

600

Gousse et al.

CASE REPORT A 62-year-old man with adult-onset diabetes mellitus underwent an uneventful repair of a left inguinal hernia. At his return visit one week after surgery, he complained of ongoing pain and requested renewal of his prescription for narcotics. The wound was neither draining nor erythematous. Three days later, he called the surgeon, again complaining of increased pain in the wound; after a telephone interview, he was told to continue the narcotic therapy. The next day he was found unresponsive at home and was taken to the emergency department by ambulance. He was unresponsive, cold, clammy, and anuric. On physical examination, the hernia incision was noted; palpation demonstrated diffuse crepitus in the left lower quadrant, perineum, left flank, and left thigh. Radiography demonstrated subcutaneous gas, a finding consistent with a diagnosis of Fournier gangrene (Fig. 2). Endotracheal intubation, ventilator support, and large-bore intravenous access were initiated, and the patient was resuscitated with intravenous fluids. Broad-spectrum intravenous antibiotics were administered. Although a bed in the intensive care unit was reserved for the patient, he was taken directly to the operating room for wide debridement. The patient lost the full-thickness skin covering of the entire perineum, left flank, and left thigh. The anesthesia team continued aggressive resuscitation with fluids and blood, with careful attention to the patient’s central pressures. The penis and testicles were wrapped in Vaseline1 gauze, and the patient was taken to the intensive care unit. On the following day, further operative debridement was performed. On the next day, the wounds were found to be clean, and the patient’s condition stabilized. One week later, he underwent surgery for coverage of the genitalia. The right testicle was buried into the right thigh, and the left testicle and the penis were covered with a split-thickness skin graft. The rest of the wounds were eventually covered with skin grafts, and the patient was discharged to a rehabilitation unit.

is perhaps the most common surgical procedure performed in the United States (88). For neonates, a bell clamp (Gomco clamp) is used without any sutures, whereas for older children and adults, a double-sleeve technique including sutures is used. There is controversy about the indications for pediatric circumcision. Those who support routine circumcision point to a reduction in the risk of neonatal urinary tract infection and the prevention of penile cancer. However, an absolute contraindication to circumcision is the presence of congenital disorders such as hypospadias because the intact prepuce is crucial for penile reconstruction. For older men, the most common indication for circumcision is the presence of phimosis. Complications occur in as many as 2% of cases; the frequency of their occurrence is negatively correlated with the experience of the surgeon (89). Bleeding and infection are common. A devastating complication is the removal of excessive skin and deep structures such as the urethra or corporal bodies. In these situations, reconstruction efforts should be undertaken without genital reassignment.

Penectomy Penectomy is indicated for penile cancer. Partial penectomy is performed for distal cancers, whereas total penectomy is performed for proximal cancers.

Complications of penectomy are bleeding, infection, and urethral strictures at the neourethra.

Penile Prosthesis Indications Penile prosthesis surgery is indicated for patients who desire to regain erectile function. In such cases, lesser treatments such as oral ingestion or penile injection of vasoactive substances have usually already failed. Occasionally, these other treatments are contraindicated and penile prosthesis surgery is selected as the primary therapy. Either semirigid or inflatable devices may be chosen for implantation. Device selection is dictated by the preferences of the surgeon and the patient. Mechanically, the semirigid device is simpler in design and is therefore easier to place than the modern inflatable prosthesis, which is composed of three or four different pieces. In addition, the semirigid device is associated with fewer complications than the inflatable device (90).

Technical Considerations The operative placement of a penile prosthesis begins with either a penoscrotal or an infrapubic incision. The corpora cavernosa are then dissected. The corporal bodies are dilated, and the cylinders of the

Chapter 48: Urethral, Scrotal, and Penile Surgery

prosthesis, either rigid or inflatable, are placed. As an inflatable device, a mechanical pump is placed in the scrotum. Finally, a hydraulic reservoir is placed either in a retropubic position (AMS) or with the pump as a single unit within the scrotum (Mentor, Santa Barbara, California, U.S.).

Complications In general, complications associated with penile prosthesis surgery can be divided into four distinct categories: intraoperative technical problems, infections, postoperative problems, and mechanical failure (91). A common intraoperative problem is perforation of the corpora cavernosa during dilation; this problem typically occurs at the proximal crura. Management involves closing the perforation or fashioning a sleeve of artificial material (such as Gore-Tex1) into a ‘‘sock’’ that is placed at the proximal aspect of the prosthesis (92,93). One of the most dreaded complications associated with penile prosthesis surgery is infection; it occurs in 0.6% to 8.9% of all cases (94). Infection is more common among patients with diabetes mellitus, spinal cord injuries, or recent urinary tract infections and among those undergoing an operation for placement of a replacement device. When a prosthetic infection occurs, the device is removed, either partially or, in the case of multicomponent devices, completely. The surgeon must then decide whether to replace the prosthesis while performing a salvage procedure or to remove the entire prosthesis and place drains within the corporal chambers (95). Other postoperative surgical complications are problems with positioning, pain, and pressure necrosis that can result in erosion of the prosthesis. Mechanical complications have improved with each new generation of prosthetic devices. Presently, a 5% rate of mechanical failure is expected within 5 to 10 years after implantation (91). Revision surgery is more difficult than the original procedure because of scar tissue. If the patient has experienced leakage of hydraulic fluid, the connector sites should be explored first because these sites are frequently associated with pressure-induced failure. The second most common site of fluid leakage is the cylinders themselves, at the point at which the tubing attaches to the cylinder. However, if mechanical failure occurs more than five years after implantation, the entire device should be replaced.

SUMMARY Complications of genitourinary surgery include fistulae, incontinence, infertility, and erectile dysfunction. Expertise in the anatomy and physiology of this organ system, coupled with the knowledge of the rapidly evolving technology, which is available, can decrease the long-term morbidity in patients with these complications.

601

REFERENCES 1. Saglowsky AI, Peters PC. Genitourinary trauma. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, eds. Campbell’s Urology. 8th ed. Philadelphia: Saunders, 2002:3085–3120. 2. Staskin DR, Hadley HR, Leach GE, Schmidbauer CP, Zimmern P, Raz S. Anatomy for vaginal surgery. Semin Urol 1986; 4:2–6. 3. Dreitlein DA, Suner S, Basler J. Genitourinary trauma. Emerg Med Clin North Am 2001; 19:569–590. 4. Herschorn S, Thijssen A, Radomski SB. The value of immediate or early catheterization of the traumatized posterior urethra. J Urol 1992; 148:1428–1431. 5. Ryu J, Kim B. MR imaging of the male and female urethra. Radiographics 2001; 21:1169–1185. 6. Morehouse D, Belitsky P, Mackinnon K. Rupture of the posterior urethra. J Urol 1972; 107:255–258. 7. Corriere JN. Trauma to the lower urinary tract. In: Gillenwater JY, Grayhack JT, Howards SS, Mitchell ME, eds. Adult and Pediatric Urology. 4th ed. Lippincott Williams and Wilkins, 2001:400–513. 8. Devine CJ, Jordan GH, Schlossberg SM. Surgery of the penis and urethra. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, eds. Campbell’s Urology. 8th ed. Philadelphia: Saunders, 2002:2957–3032. 9. Podesta ML, Jordan GH. Pelvic fracture urethral injuries in girls. J Urol 2001; 165:1660–1665. 10. Boone TB, Wilson WT, Husmann DA. Postpubertal genitourinary function following posterior urethral disruptions in children. J Urol 1992; 148:1232–1234. 11. Quint HJ, Stanisic TH. Above and below delayed endoscopic treatment of traumatic posterior urethral disruption. J Urol 1993; 149:484–487. 12. Spirnak JP, Smith EM, Elder JS. Posterior urethral obliteration treated by endoscopic reconstitution, internal urethrotomy and temporary self-dilation. J Urol 1993; 149:766–768. 13. Jordan GH, Schlossberg SM, Devine CJ. Surgery of the penis and urethra. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, eds. Campbell’s Urology. 8th ed. Philadelphia: Saunders, 2002:3316–3394. 14. McAninch JW, Laing FC, Jeffrey RB Jr. Sonourethrography in the evaluation of urethral strictures: a preliminary report. J Urol 1988; 139:294–297. 15. Pansadoro V, Emiliozzi P. Internal urethrotomy in the management of anterior urethral strictures: long-term follow-up. J Urol 1996; 156:73–75. 16. Morey AF, McAninch JW, Duckett CP, Rogers RS. American Urological Association symptom index in the assessment of urethroplasty outcomes. J Urol 1998; 159:1192–1194. 17. Schreiter F, Noll F. Mesh graft urethroplasty using split thickness skin graft or foreskin. J Urol 1989; 142:1223–1226. 18. Quartey JK. One-stage penile/preputial cutaneous island flap urethroplasty for urethral stricture: a preliminary report. J Urol 1983; 129:284–287. 19. Fowler FJ Jr, Barry MJ, Lu-Yao G, Roman A, Wasson J, Wennberg JE. Patient-reported complications and follow-up treatment after radical prostatectomy. The National Medicare Experience: 1988–1990 (updated June 1993). Urology 1993; 42:622–629. 20. Litwiller SE, Djavan B, Klopukh BV, Richier JC, Roehrborn CG. Radical retropubic prostatectomy for localized carcinoma of the prostate in a large metropo-

602

21. 22.

23. 24.

25. 26.

27.

28. 29.

30.

31. 32. 33. 34.

35. 36.

37.

38.

Gousse et al.

litan hospital: changing trends over a 10-year period (1984–1994). Dallas Outcomes Research Group for Urological Disorders. Urology 1995; 45:813–822. Steiner MS, Morton RA, Walsh PC. Impact of anatomical radical prostatectomy on urinary continence. J Urol 1991; 145:512–515. Madjar S, Raz S, Gousse AE. Fixed and dynamic urethral compression for the treatment of post-prostatectomy urinary incontinence: is history repeating itself? J Urol 2001; 166:411–415. Kaufman JJ. Urethral compression operations for the treatment of post-prostatectomy incontinence. J Urol 1973; 110:93–96. Barrett DM, Licht MR. Implantation of the artificial genitourinary sphincter in men and women. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, eds. Campbell’s Urology. 8th ed. Philadelphia: Saunders, 2002:1121–1134. Fishman IJ, Shabsigh R, Scott FB. Experience with the artificial urinary sphincter Model AS800 in 148 patients. J Urol 1989; 141:307–310. Montague DK, Angermeier KW, Paolone DR. Longterm continence and patient satisfaction after artificial sphincter implantation for urinary incontinence after prostatectomy. J Urol 2001; 166:547–549. Gousse AE, Madjar S, Lambert MM, Fishman IJ. Artificial urinary sphincter for post-radical prostatectomy urinary incontinence: long-term subjective results. J Urol 2001; 166:1755–1758. Dupont MC, Albo ME, Raz S. Diagnosis of stress urinary incontinence. An overview. Urol Clin North Am 1996; 2323:407–415. Stothers L. Reliability, validity and gender differences in the quality of life index of the SEAPI-QMM incontinence classification system. Neurourol Urodyn 2004; 23:223–228. Fonda D, Woodward M, D’Astoli M, Chin WF. Sustained improvement of subjective quality of life in older community-dwelling people after treatment of urinary incontinence. Age Ageing 1995; 24:283–286. Ekelund P, Grimby A, Milsom I. Urinary incontinence. social and financial costs high. BMJ 1993; 306:1344. Diokno AC. Epidemiology and psychosocial aspects of incontinence. Urol Clin North Am 1995; 22:481–485. Beck RP, McCormick S. Treatment of urinary stress incontinence with anterior colporrhaphy. Obstet Gynecol 1982; 59:269–274. Raz S, Stothers L, Young GP, et al. Vaginal wall sling for anatomical incontinence and intrinsic sphincter dysfunction: efficacy and outcome analysis. J Urol 1996; 156:166–170. Dwyer NT, Kreder KJ. An update on slings. Curr Opin Urol 2005; 15:244–249. Abdel-Fattah M, Barrington JW, Arunkalaivanan AS. Pelvicol pubovaginal sling versus tension-free vaginal tape for treatment of urodynamic stress incontinence: a prospective, randomized three-year follow-up study. Eur Urol 2004; 46:629–635. Ulmsten U, Henriksson L, Johnson P, Varhos G. An ambulatory surgical procedure under local anesthesia for treatment of female urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct 1996; 7:81–86. Tseng LH, Wang AC, Lin YH, Li SJ, Ko YJ. Randomized comparison of the suprapubic arc sling procedure versus tension-free vaginal taping for stress incontinent

39.

40.

41.

42.

43.

44. 45. 46. 47.

48.

49. 50. 51. 52. 53.

54. 55.

women. Int Urogynecol J Pelvic Floor Dysfunct 2005; 16(3):230–235. O’Leary MP, Gee WF, Holtgrewe HL, et al. 1999 American Urologic Association Gallup Survey: changes in physician practice patterns, treatment of incontinence and bladder cancer, and impact of managed care. J Urol 2000; 164:1311–1316. Kim HL, Gerber GS, Patel RV, Hollowell CM, Bales GT. Practice patterns in the treatment of female urinary incontinence: a postal and internet survey. Urology 2001; 57:45–48. Leach GE, Dmochowski RR, Appell RA, et al. Female stress urinary incontinence. Clinical Guidelines Panel summary report on surgical management of female stress urinary incontinence. The American Urological Association. J Urol 1997; 158:875–880. Nilsson CG, Kuuva N, Falconer C, Rezapour M, Ulmsten U. Long-term results of tension-free vaginal tape (TVT) procedure for surgical treatment of female stress urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct 2001; 12(suppl 2):S5–S8. Klutke C, Siegel S, Carlin B, Paszkiewicz E, Kirkemo A, Klutke J. Urinary retention after tension-free vaginal tape procedure: incidence and treatment. Urology 2001; 58:697–701. Stamey TA. Endoscopic suspension of the vesical neck for urinary incontinence. Surg Gynecol Obstet 1973; 136:547–554. Raz S. Modified bladder neck suspension for female stress incontinence. Urology 1981; 17:82–85. McGuire EJ. Urodynamic findings in patients after failure of stress incontinence operations. Prog Clin Biol Res 1981; 78:351–360. Leach GE, Trockman BA. Surgery for vesicovaginal and urethrovaginal fistula and urethral diverticulum. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, eds. Campbell’s Urology. 8th ed. Philadelphia: Saunders, 2002:1135–1153. Patil U, Waterhouse K, Laungani G. Management of 18 difficult vesicovaginal and urethrovaginal fistulas with modified Ingelman-Sundberg and Martius operations. J Urol 1980; 123:653–656. Blaivas JG. Vaginal flap urethral reconstruction: an alternative to the bladder flap neourethra. J Urol 1989; 141:542–545. Leach GE. Urethrovaginal fistula repair with Martius labial fat pad graft. Urol Clin North Am 1991; 18:409–413. Goodwin WE, Scardino PT. Vesicovaginal and ureterovaginal fistulas: a summary of 25 years of experience. J Urol 1980; 123:370–374. Webster GD, Sihelnik SA, Stone AR. Urethrovaginal fistula: a review of the surgical management. J Urol 1984; 132:460–462. Zimmern PE, Hadley HR, Leach GE, Raz S. Transvaginal closure of the bladder neck and placement of a suprapubic catheter for destroyed urethra after long-term in dwelling catheterization. J Urol 1985; 134:554–557. Blaivas JG. Treatment of female incontinence secondary to urethral damage or loss. Urol Clin North Am 1991; 18:355–363. Elkins TE, Ghosh TS, Tagoe GA, Stocker R. Transvaginal mobilization and utilization of the anterior bladder wall to repair vesicovaginal fistulas involving the urethra. Obstet Gynecol 1992; 79:455–460.

Chapter 48: Urethral, Scrotal, and Penile Surgery

56. Lockhart JL, Walker RD, Vorstman B, Politano VA. Periurethral polytetrafluoroethylene injection following urethral reconstruction in female patients with urinary incontinence. J Urol 1988; 140:51–52. 57. Boyd SD, Raz S. Ectopic ureter presenting in midline urethral diverticulum. Urology 1993; 41:571–574. 58. Ganabathi K, Leach GE, Zimmern PE, Dmochowski R. Experience with the management of urethral diverticulum in 63 women. J Urol 1994; 152:1445–1452. 59. Aldridge CW Jr, Beaton JH, Nanzig RP. A review of office urethroscopy and cystometry. Am J Obstet Gynecol 1978; 131:432–437. 60. Robertson JR. Genitourinary Problems in Women. Springfield, IL: Charles C. Thomas, 1978. 61. Peters W III, Vaughan ED Jr. Urethral diverticulum in the female. Etiologic factors and postoperative results. Obstet Gynecol 1976; 47:549–552. 62. Leach GE, Bavendam TG. Female urethral diverticula. Urology 1987; 30:407–415. 63. Davis HJ, TeLinde RW. Urethral diverticula: an assay of 121 cases. J Urol 1958; 80:34–39. 64. Blacklock AR, Shaw RE, Geddes JR. Late presentation of ectopic ureter. Br J Urol 1982; 54:106–110. 65. Lapides J. Transurethral treatment of urethral diverticula in women. J Urol 1979; 121:736–738. 66. Burrows LJ, Howden NL, Meyn L, Weber AM. Surgical procedures for urethral diverticula in women in the United States, 1979–1997. Int Urogynecol J Pelvic Floor Dysfunct 2005; 16:158–161. 67. Leach GE, Schmidbauer CP, Hadley HR, Staskin DR, Zimmern P, Raz S. Surgical treatment of female urethral diverticulum. Semin Urol 1986; 4:33–42. 68. Granitsiotis P, Kirk D. Chronic testicular pain: an overview. Eur Urol 2004; 45:430–436. 69. Junnila, Lassen P. Testicular masses. Am Fam Physician 1998; 57:685–692. 70. Fournier JA. Jean-Alfred Fournier 1832–1914. Gangrenefoudroyante de la verge (overwhelming gangrene) Sem Med 1883; 3:345. Dis Colon Rectum 1988; 31:984–988. 71. Yeniyol CO, Suelozgen T, Arslan M, Ayder AR. Fournier’s gangrene: experience with 25 patients and use of Fournier’s gangrene severity index score. Urology 2004; 64:218–222. 72. Quatan N, Kirby RS. Improving outcomes in Fournier’s gangrene. BJU Int 2004; 93:691–692. 73. Paty R, Smith AD. Gangrene and Fournier’s gangrene. Urol Clin North Am 1992; 19:149–162. 74. Baskin LS, Carroll PR, Cattolica EV, McAninch JW. Necrotising soft tissue infections of the perineum and genitalia. Bacteriology, treatment and risk assessment. Br J Urol 1990; 65:524–529. 75. Clayton MD, Fowler JE Jr, Sharifi R, Pearl RK. Causes, presentation and survival of fifty-seven patients with necrotizing fasciitis of the male genitalia. Surg Gynecol Obstet 1990; 170:49–55.

603

76. Aboseif S, Gomez R, McAninch JW. Genital selfmutilation. J Urol 1993; 150:1143–1146. 77. Altarac S. Management of 53 cases of testicular trauma. Eur Urol 1994; 25:119–123. 78. Corrales JG, Corbel L, Cipolla B, et al. Accuracy of ultrasound diagnosis after blunt testicular trauma. J Urol 1993; 150:1834–1836. 79. Huggins C, Hodges CV. Studies on prostatic cancer: I. The effects of castration, of estrogen and of androgen injection of serum phospholipases in metastatic carcinoma of the prostate 1941. J Urol 2002; 168:9–12. 80. Benjamin K. Scrotal and inguinal masses in the newborn period. Adv Neonatal Care 2002; 2:140–148. 81. Ross LS, Flom LS. Azoospermia: a complication of hydrocele repair in a fertile population. J Urol 1991; 146:852–853. 82. Aafjes JH, van der Vijver JC. Fertility in men with and without a varicocele. Fertil Steril 1985; 43:901–904. 83. Szabo R, Kessler R. Hydrocele following internal spermatic vein ligation: a retrospective study and review of the literature. J Urol 1984; 132:924–925. 84. Goldstein M, Gilbert BR, Dicker AP, Dwosh J, Gnecco C. Microsurgical inguinal varicocelectomy with delivery of the testis: an artery and lymphatic sparing technique. J Urol 1992; 148:1808–1811. 85. Li SQ, Goldstein M, Zhu J, Huber D. The no-scalpel vasectomy. J Urol 1991; 145:341–344. 86. Kendrick J, Gonzales B, Huber D, Grubb GS, Rubin GL. Complications of vasectomy in the United States. J Fam Pract 1987; 25:245–248. 87. Schuman LM, Coulson AH, Mandel JS, Massey FJ Jr, O’Fallon WM. Health Status of American Men – A study of postvasectomy sequelae. 1993; 46:697–958. 88. Thompson HC, King LR, Knox E, Korones SB. Report of the ad hoc task force on circumcision. Pediatrics 1975; 56:610–611. 89. Williams N, Kapila L. Complications of circumcision. Br J Surg 1993; 80:1231–1236. 90. Lewis RW. Long-term results of penile prosthetic implantation. Urol Clin North Am 1995; 22:847–856. 91. Lewis R. Surgery for erectile dysfunction. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, eds. Campbell’s Urology. 7th ed. Philadelphia: Saunders, 1998:1215–1235. 92. Fishman IJ. Corporeal reconstruction procedures for complicated penile implants. Urol Clin North Am 1989; 16:73–90. 93. Mulcahy JJ. A technique of maintaining penile prosthesis position to prevent proximal migration. J Urol 1987; 137:294–296. 94. Carson CC. Diagnosis, treatment and prevention of penile prosthesis infection. Int J Impot Res 2003; 15(suppl 5):S139–S146. 95. Wilson SK, Delk JR II. Inflatable penile implant infection: predisposing factors and treatment suggestions. J Urol 1995; 153:659–661.

49 Complications of Genitourinary Trauma Yekutiel Sandman Department of Urology, Jackson Memorial Medical Center, University of Miami Miller School of Medicine, Miami, Florida, U.S.A. Peter P. Lopez Division of Trauma and Surgical Critical Care, DeWitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine, Miami, Florida, U.S.A.

Injury to the genitourinary (GU) system is a common, but rarely life-threatening, injury after major trauma. Urologic trauma is associated with other, more serious, injuries. Overall, about 10% of all traumas have GU involvement (1). When presented with a trauma patient, the clinician needs to be attentive to potential GU trauma by knowing the mechanism of injury, anatomy, and signs and symptoms of GU trauma. Following initial evaluation according to advanced trauma life support (ATLS) standards, the possibility of GU injury should be addressed if certain signs and symptoms are present (2). Among these signs are hematuria, hypotension, pelvic fracture, and flank hematoma. An increased blood urea nitrogen (BUN) to creatinine ratio is consistent with spillage of urine into the peritoneal cavity, though many types of traumatic GU injuries have a normal ratio (3). When GU trauma is suspected (Fig. 1), the first decision is whether to proceed to the operating room (OR) or to obtain imaging studies. If the patient is stable, radiologic evaluation may begin; however, if the patient is unstable, the patient should be taken directly to the OR. If the patient does go to the OR, the next decision is whether to obtain intraoperative imaging of the GU tract. Intraoperative images are obtained by injecting 2 mL/kg of contrast and acquiring a film 10 minutes after injection (4). Intraoperative images allow assessment of the GU tract in an unstable patient without delaying critical surgery. This data rules out serious renal injury, thereby negating the need for renal exploration in most circumstances and documents the presence of a second functioning kidney (which may be of medicolegal importance). However, intraoperative films are not always necessary. In a stable patient, imaging is indicated with (i) gross hematuria; (ii) microhematuria and signs of shock; (iii) penetrating trauma consistent with injury to the GU system; or (iv) signs of renal trauma such as flank ecchymosis, pulsatile flank mass, lower rib fractures, or facture of the transverse spinous processes (2,5). In the pediatric patient, microscopic hematuria with greater than

50 red blood cells (RBCs) per high-powered field (HPF) should lead to evaluation of the GU tract even without signs of shock (6). The lower threshold in children is due to less perinephric fat protecting the kidney, the fact that children maintain normal pressure for longer time even with severe fluid loss, and the relatively larger size of their kidneys, which increases the possibility of renal trauma. Ultrasound, though a popular imaging technique for abdominal trauma (7), is less helpful in imaging the GU system (8). In the past two decades, computerized tomography (CT) has replaced intravenous pyelogram (IVP) for imaging the GU tract when trauma is suspected (9). Among the advantages of CT, is the ability to distinguish patients requiring operative intervention from those who would benefit from conservative management (10). In addition, CT provides a baseline for later comparison. Spiral CT scans have been criticized by many for not having an excretory phase nephrogram due to the rapidity of the imaging (11,12); obtaining delayed images assures the GU tract is adequately assessed (13,14). Those patients that do not require imaging, i.e., those without hematuria or adults with microscopic hematuria and stable blood pressure, may safely be followed with a standard urinalysis, three weeks after the initial trauma, to assure that there is no persistent hematuria (15).

RENAL Epidemiology and Diagnosis The kidneys are injured in about 5% to 10% of all trauma (16). Presenting signs, symptoms and clues include flank or abdominal pain, flank ecchymosis, rib fractures, penetrating injuries (usually bullet or stab wounds) to the flank, abdomen or lower chest and, like most other GU trauma, hematuria. However, there is no correlation between the amount of hematuria and the severity of the injury to the kidney. Moreover, pedicle injury, which potentially results in renal loss, may not have any hematuria at all as the

606

Sandman and Lopez

Trauma and hematuria or symptoms suspicious for GU trauma

unstable

stable

To OR 1. Gross hematuria 2. Microscopic hematuria and shock 3. Hematuria and symptoms of renal injury 4. Penetrating injury suspicious for GU injury 5. Child with more than 50 RBCs/HPF

1. Microscopic hematuria & stable vital signs in adult 2. Child with less than 50 RBCs/HPF

Follow up urinanalysis 3 weeks Conservative management

CT scan (or IVP)

Conservative management

Explore

Intra-op IVP

To OR

injury is outside the collecting system. The most common mechanism of injury (10) to the kidney (about 90% of renal trauma) is blunt trauma. Blunt injury with rapid deceleration may cause injury to the renal vessels. Penetrating injuries account for only 10% of traumatic renal injuries, though urban trauma centers may see a 20% to 30% incidence of penetrating renal trauma. In a stable patient with any of the indications discussed above, CT imaging of the GU tract is paramount because most renal injuries are managed nonoperatively (see below). CT scan allows radiographic grading of the injury assisting with formulation of a treatment plan. Subtle differences exist between various authors on how to best perform a CT scan for GU trauma. According to Wah and Spencer (5), the scan should start from the dome of the diaphragm and continue to the iliac crests; patients receive contrast by mouth, as well as intravenous (IV) contrast. The scan begins at 65 seconds after the administration of 150 mL of 300 mg/mL nonionic IV contrast to obtain early phase films. Wah also obtains delayed films when renal injury is suspected, though others advocate regularly obtaining delayed films at three minutes to evaluate the renal system, regardless of suspicion of renal injury (17). To evaluate the collecting system, late films at 15 to 20 minutes are routinely done.

Management The most accepted classification for renal trauma is the one developed by the American Association for the Surgery of Trauma (Fig. 2 and Table 1) (18,20). A review of 2483 patients with renal trauma at San Francisco General Hospital showed (19) that Grade I and II injuries (injuries of the kidney including hematoma and superficial lacerations) are managed

Explore

Figure 1 Flowchart. Abbreviations: GU, genitourinary; RBCs, red blood cells; HPF, high-powered field; IVP, intravenous pyelogram; OR, operating room; CT, computerized tomography.

nonoperatively as are most (deeper) Grade III injuries (21). Grade IV injuries, (those with laceration to the kidney involving the collecting system or thrombosis of the main renal vessels), are managed operatively if there is hemodynamic instability. Other operative criteria include intractable renal bleeding (expanding renal hematoma, or a pulsatile renal hematoma), injury to the renal vessels, or the presence of a nonviable kidney segment associated with other injuries (22,23). Extravasation of urine alone has been an indication for operative intervention, although currently it is commonly handled nonoperatively (16,24). Grade V injuries, (avulsed renal pedicle or shattered kidney) should be managed operatively except in rare, select cases (when the criteria above are not met despite having a Grade V injury) (25). Nicol and Theunissen advocate exploring all penetrating renal trauma, if the patient is undergoing laparotomy for other injuries because it allows direct visualization and does not lead to higher rates of nephrectomy (26). Margenthaler et al. advocate nonoperative management of all renal trauma in children unless the patient is hemodynamically unstable (27). Nonoperative management consists of bed rest until gross hematuria ceases. If hematuria recurs upon ambulation, the patient should be reconfined to bed. Once the patient can ambulate without hematuria, he or she can be safely discharged with follow-up (28). Surgical options include primary repair, partial nephrectomy or, as a last resort, total nephrectomy. The approach to the kidney (29) is via the abdomen. After examining the abdominal contents and repairing any intra-abdominal injury, attention should be turned to the kidneys. The transverse colon is superiorly reflected. The small bowel is reflected off the field superiorly and to the right. At this point, the posterior peritoneum is incised (Fig. 3). Obtaining

Chapter 49: Complications of Genitourinary Trauma

607

Figure 2 American Association for the Surgery of Trauma Renal Trauma Grading Score and computerized tomography of injuries by grade. Source: From Refs. 17–19.

608

Sandman and Lopez

Table 1 American Association for the Surgery of Trauma Classification of Renal Trauma Grade I II III IV V

Type of injury

Management

Hematoma Laceration < 1 cm Laceration > 1 cm but not involving the collecting system Laceration extending into the collecting system or main renal vessel thrombosis Multiple lacerations extending into the collecting system or devascularized kidney

Nonoperative Nonoperative Nonoperative At times nonoperative but usually operative Almost always operative

control of the aorta and vena cava at this point is the accepted next step (Fig. 4); however, the utility of this maneuver has recently come into question (30). To obtain control of the renal vessels, one should gently retract the left renal vein superiorly with a vessel loop. The left renal artery is identified and controlled with a nonocclusive vessel loop followed by identification and control of the right renal artery, which is found between the aorta and vena cava after retracting the left renal vein cranially and after incising between the aorta and vena cava. Finally, the right renal vein is identified and controlled. The vessels are not occluded unnecessarily since the warm ischemia time of the kidney is only 30 minutes. Next, the colon is reflected medially, Gerota’s fascia incised, and the kidney and all its vessels should be explored. Primary operative repair (Fig. 5) is an option if there is a Grade I, II, III, or IV injury to the kidney. It consists of individual control of parenchymal renal vessels with suture ligation (typically with 4-0 chromic sutures) and watertight closure of the collecting system with a running absorbable suture (such as 4-0 chromic). Partial nephrectomy follows the same

principles and includes the sharp debridement of nonviable tissue. In both procedures, the surgeon should cover the kidney with intrinsic renal fascia or with the omentum in order to decrease extravasation and increase hemostasis (29). Total nephrectomy is indicated in select cases: an unstable patient with a normal second kidney, if the patient’s instability is due to low body temperature and poor coagulation, a nonrepairable kidney, and warm ischemic time greater than six hours (31).

Figure 3 Exposure for renal trauma. Source: From Ref. 29.

Figure 4 Method of renovascular control. Source: From Ref. 29.

Complications Complications of renal trauma, in addition to the morbidity and mortality attached to any traumatic injury, are either immediate or delayed (16). Immediate complications include extravasation, which can lead to urinoma, which, in turn, can lead to infection and abscess. Whitney and Peterson (32), in a series of 81 penetrating trauma victims who underwent operative intervention, found a urinary leak rate of 2% in minor injuries (injuries in which the surgical intervention proved needless) and 33% with major injuries (those where the operative intervention proved essential). Simple extravasation may be treated via internal stenting (33). In fact, in a series of 46 patients with Grade IV or V lesions that were treated nonoperatively, Matthews et al. found that 31 patients had extravasation. However, the leak resolved without any intervention in all but four patients. Those patients were treated with ureteral stenting. There were no complications

Chapter 49: Complications of Genitourinary Trauma

609

Table 2 Complication Rate in Grade IV Injuries by Concurrent Intra-abdominal Injury and Management Injury/management Intra-abdominal injury No intra-abdominal Injury

Explored

Not explored

3/13 (23%) 6/38 (38%)

12/14 (85%) N/A

Abbreviation: N/A, not applicable.

Figure 5 Primary renal repair: (A) Debridement of nonviable tissue. (B) Closure of collecting system (after tying off of parenchymal vessels). (C) Omental flap. Source: From Ref. 29.

with nonoperative management (no increased risk of hypertension or renal failure) and all 31 patients received prophylactic antibiotics (24). The incidence of urinoma (Fig. 6) and infection can be stratified by grade of injury. The overall risk of surgery versus nonoperative complications falls in favor of nonintervention for Grades I to III (19). For Grade IV injuries, Husmann et al. (22) demonstrated that there is an 85% chance of complication including infected urinoma, hypertension, and hemorrhage if an

abdominal injury coexists with a devitalized segment of kidney. In contrast, there was only a 23% risk of complication if the same type of patient was managed surgically. If there was no associated intra-abdominal injury, the likelihood of complication was 38%; however, the majority of these complications (five of six) were amenable to nonoperative management (Table 2). Urinoma is diagnosed by CT scan obtained with delayed images, which allows the contrast to pass through the kidney and into the collecting system. Treatment for urinoma is based on the length of time since the operation and/or trauma and whether an abscess or fistula is present. If it has been a week or less and there is no active infection, operative drainage and repair is indicated. If the patient is actively infected or if the urinoma has been present for seven days or more, the patient should be managed nonoperatively as the inflammation will likely make for a difficult dissection and poorly healing tissue (28). Percutaneous drainage of the collection and placement of a nephroureteral stent is the treatment of choice in these instances (34–36). Infection should be treated with broad-spectrum antibiotics and other supportive treatment as needed. Definitive repair is undertaken when the patient is more stable. Knudson et al. (37) found an overall 23% complication rate with renovascular injuries (all Grade IVor V

Figure 6 (A) Perinephric collection in a case of a ureteropelvic junction distraction injury on the right with (B) extension into the Psoas muscle (solid black arrow).

610

Sandman and Lopez

by definition). These complications included delayed nephrectomy, persistent hypertension, renal failure, and diminished renal function. Renovascular injuries have not shown to have statistically significant differences in repair versus nephrectomy except in Grade V arterial injuries. More patients develop complications in attempted repair of such an injury than do those that undergo nephrectomy (odds ratio of 15% increase in above complications). Knudson et al., therefore, recommend nephrectomy in Grade V renal artery injuries and repair in Grade IV renal artery injuries. Hypertension has also been reported as a delayed complication of renal trauma (38). In a review of 158 patients with vascular and parenchymal renal trauma, Montgomery et al. found a 4.4% incidence of hypertension attributable to renal trauma. The onset of symptoms ranged from immediate (when still hospitalized for the trauma) to eight months after the initial event. The symptoms included headache, chest pain or tightness, and nosebleeds. There were several patients with no symptoms or vague complaints as well. The need to obtain vascular control prior to exploration of the kidney for trauma has come into question recently. It has been thought that vascular control decreases the rate of nephrectomy while adding minimal time to the operation (39). However, recent studies have questioned the advantages of vascular control. For instance, Gonzalez et al. found no statistical difference in blood loss, rate of nephrectomy, or transfusions in a trial of patients randomized to either vascular control versus no early vascular control (30). The rate of nephrectomy was 31% and 30% in the vascular control and no-control groups, respectively. The no-control patients were transfused an average of 5.2 units of packed RBCs versus 5.5 units in the control group, while the blood loss was 0.9 and 1.1 L in the each of the above groups. A limitation of the above study was its small sample size. Our experience has been to gain vascular control if repair of the kidney is planned. However, if the patient is unstable and bleeding, we do not take the additional time required gaining vascular control; rather, we proceed directly to evaluation of the kidney. Delayed bleeding has been noted following renal trauma as well (40). Carroll et al. reviewed 92 of the renal trauma cases that came into San Francisco General Hospital and found two cases that had delayed bleeding (41). Both of these cases were operatively reexplored. Neither required nephrectomy to control the bleeding. Teigen et al. described two cases of delayed hematuria in children where pseudoaneurysm developed in an area of the kidney that was devitalized (40). They advocate early angiography for evaluation of delayed hematuria and possible treatment via embolization of any pseudoaneurysm. Delayed bleeding occurs in adults as well and is usually managed by interventional radiologic techniques.

URETERAL Epidemiology and Diagnosis Traumatic ureteral injury is rare, usually occurring in coincidence with damage to other intra-abdominal structures (42). Iatrogenic injuries occur most commonly during ureteroscopic and gynecological procedures (43). Though ureteral injury is often accompanied by hematuria, typically microscopic, in one series, 15% of these injuries occurred without hematuria (44). Diagnosing ureteral trauma correctly depends on a high index of suspicion, which should be raised when the injury makes ureteral damage possible (a bullet’s trajectory in proximity to the ureter) or when an intervention has occurred in proximity to the ureter. When such injury is missed, the patient will often present with a flank mass, or the signs and symptoms of an infected urinoma (e.g., rigors, tenderness, erythema, fevers, and hypotension). Regardless of the cause of ureteric injury, the principles of diagnosis and repair remain the same. The most common mode of noniatrogenic injury to the ureter is gunshot wounds (GSW); however, only about 2.5% to 5% of GSW are accompanied by ureteral injury (45). Some GSW are not due to direct transection, but rather secondary to a ‘‘blast injury.’’ This blast effect occurs when the force of the bullet is transmitted through the abdomen; the ureter is devascularized and becomes nonviable (28). Eventually, there is necrosis and breakdown of the ureter, which results in ureteral stricture or rupture. Another possible etiology of ureteral injury due to external trauma is ureteralpelvic disruption secondary to violent hyperextension of the trunk, e.g., as during a motor vehicle crash. This type of injury occurs more frequently in children (46). Those patients who have suffered external trauma with ureteral damage, regardless of etiology, are often critically ill with multiple injuries (47). Imaging of the ureters allows identification of injury (48). If CT scanning is used, it is paramount that a nephric excretory phase is obtained to make the diagnosis. It is in these instances that spiral CTs are criticized, due to their speed and the possibility that a ureteric injury may be missed due to the lack of an excretory phase. Therefore, when a CT scan is used for diagnosis of a possible GU injury, it is imperative that delayed films (at least five to eight minutes after the injection of contrast) are obtained (49). Extravasation of contrast (Fig. 7) if the ureter is disrupted or nonpassage of contrast if the ureter has ceased functioning without formal disruption (as in the first stages of a ‘‘blast’’ injury) would be seen. A ureteral contusion can be missed with intraoperative IVP (4). Therefore, Palmer et al. (42) as well as Medina et al. (50) maintain that diagnosis of ureteral injury should be made by direct observation when exploring the abdomen in the OR. They point out that methylene blue can be injected into the renal pelvis with subsequent observation for leaking to assist with the diagnosis. Care should be taken to use a large gauge

Chapter 49: Complications of Genitourinary Trauma

Figure 7 Extravasation of contrast after right ureteral injury. Source: From Ref. 48.

needle, such as a 25 gauge needle, to prevent iatrogenic damage to the collecting system.

Management The first critical decision to make after diagnosing a ureteral injury is whether the patient is stable enough to undergo repair. As the consequences of GU damage are rarely life threatening, patients who are hemodynamically unstable will likely not benefit from an extended repair of a noncritical injury. In these instances, planned reoperation is the most reasonable course of action. If time permits, the ureter may be externalized by ureterostomy (Fig. 8) and corrected at a later point (52). If the patient is stable, immediate repair should be done. The ureter is divided into three parts: proximal, mid and distal. All parts share the same basic

Figure 8 Pediatric feeding tube in proximal ureter brought out through the abdominal wall. Source: From Ref. 51.

611

principles of repair; the only difference is the type of anastomosis performed. The goal is to create a tension-free but watertight anastomosis (Fig. 9). Basic principles include mobilizing a large amount of ureter while giving a wide swath to the adventitia, dissecting the ureter back to its viable, bleeding edge, spatulating the ureter at the anastomosis, isolating the repair with omentum, especially if other organs are injured, and draining the area of repair (35,53). In addition, routine stenting is beneficial (31). These principles hold true whether the injury is due to penetrating or iatrogenic trauma. In proximal injury, an end-to-end ureteroureterostomy, (reconnecting the ureter with primary closure), is usually performed if there is a simple injury with viable ureter, for example, caused by a stab wound (48). In cases of profound ureteral loss with preservation of only the most proximal ureter, options include extending the length of the ureter by interposition of ileum or decreasing the ureteral length needed by either mobilization of the kidney, autotransplantation, or via a transureteroureterostomy (connecting the injured ureter to the noninjured ureter on the contralateral side). Autotransplantation for severe proximal ureteral loss, advocated by Meng et al., has been met with success in six of seven attempts. Two of the six attempts had a direct ureterocystostomy (implanting the ureter into the bladder) and four had ureteropyeloplasties (fixing the ureter-renal pelvis area) (54). However, autotransplantation and ileal interposition have been criticized as too time-consuming to undertake in an acute trauma setting, and an expeditious transureteroureterostomy (Fig. 10) is advocated (53). In addition, Guerriero recommends a transureteroureterostomy to avoid fecal spillage in the case of contamination of the retroperitoneum (35). Midureteral injuries are treated by ureteroureterostomy. Other options, if the mobilized segment does not allow for a tension-free anastomosis, include mobilization of the kidney or creation of either a Psoas bladder hitch or Boari flap (48), though the Boari flap has been criticized for decreasing the functional capacity of the bladder (55). These options are also possible with distal ureteral injuries, although a direct ureterocystostomy is preferable if there is sufficient ureteral length (52). Due to the ability to span a large segment of nonfunctional ureter and the relative lack of complications, the Psoas hitch (Fig. 11) has gained widespread approval (55,57). It includes mobilization of the contralateral obliterated umbilical artery and the ipsilateral umbilical artery if needed, making a cystotomy to aid with pexing the bladder to the Psoas tendon, affixing the bladder to the Psoas with nonabsorbable sutures while avoiding the genitofemoral nerve and finally reanastomosing the ureter to the bladder. The ureter should be stented (56). A Boari flap (Fig. 12) consists of taking an ipsilateral bladder flap, after having filled the bladder to near capacity, and anastomosing the ureter to the flap. The flap/ureter area is

612

Sandman and Lopez

Figure 9 Ureteroureterostomy. (A) Exposure of ureteral injury. (B) Mobilization of ureter with wide swath given to adventitia to assure adequate blood supply. (C) Debridement to viable, bleeding edge. (D) Suturing of spatulated ureteral ends. (E) Final appearance of repaired ureter. Source: From Ref. 35.

then fixed to the Psoas muscle and the flap is tubularized and a stent is placed (56).

Complications Complications of ureteral injury include stricture, which occurs if there is a devitalized segment. Stricture may eventually lead to permanent renal damage because of increased pressure. Treatment of a ureteral stricture depends on several factors including the length and the vascularity of the segment that is strictured. In a study by Richter et al., strictures that were short and vascularized were found to have a balloon dilatation success rate of 89% (58). The same study found that devascularized strictures had a 40% or less success rate while strictures with increased length, at the ureteral pelvic junction or at an ureteroenteric anastomosis, had a poor success rate regardless of vascularity. If there is a leak from the anastomotic site or, more commonly, an unrecognized damaged ureter, a urinoma (Fig. 6) and all the consequences of a urinoma may occur; these include abscess formation, sepsis, and possible ureterocutaneous fistula (59). A urinoma requires drainage, as would an abscess. Access to drain them is often attained percutaneously.

Though rare, ureterovaginal fistulas or ureterocutaneous fistulas occur after unrecognized ureteral trauma as well (60). Stenting across the fistula is the treatment of choice for this type of complication (35,36). Selzman et al. found that stenting alone was adequate in patients who had the stent in place for a sufficient amount of time (seven of seven patients) (61). They found no difference in the results of two patients whose stent was placed in an anterograde fashion. Ahn and Loughlin (57) reported that only 1 of 17 patients had any postoperative complications to their Psoas hitch method. This patient was 1 of 14 with a refluxing (nontunneled) ureteral-bladder anastomosis. This patient developed urosepsis that resolved with medical treatment and had no recurrences in the 16 months of follow-up and intact renal function as well.

BLADDER Epidemiology and Diagnosis Bladder rupture due to trauma is generally categorized as either intraperitoneal or extraperitoneal. The etiology and the management of each of these are unique, as are the ramifications of the particular

Chapter 49: Complications of Genitourinary Trauma

Figure 10 Transureteroureterostomy. (A) Injured ureter pulled (behind colon) to uninjured ureter. (B) Anastomosis. (C) Final appearance. Source: From Ref. 35.

613

injury. Close to 100% of blunt trauma bladder injuries are associated with pelvic fractures (62), though only one out of 10 pelvic fractures have an associated bladder injury (63). Fifteen percent of blunt trauma associated bladder injuries are intraperitoneal while 85% are extraperitoneal (64). The causes of bladder rupture into either of these areas is due to the fact that the peritoneal lining sits above the dome of the bladder and the force of the injury may be directed superiorly, yielding an intraperitoneal rupture, or elsewhere resulting in an extraperitoneal injury. Extraperitoneal bladder rupture is due to bursting of a full bladder, shearing from its areas of attachment or perforation of the bladder by bony spicules after pelvic fracture (65). Intraperitoneal bladder injury need not be associated with pelvic fracture; the blunt trauma force transmitted to a full bladder can be enough to result in intraperitoneal rupture (66). The key to diagnosis of a bladder injury is hematuria. When associated with pelvic fracture, workup of the lower urinary tract (bladder and distal) is warranted (67). Injuries to associated areas, such as perineal hematoma or rectal bleeding, are also indicative of bladder injury (68). Morey et al. state the only absolute indication for cystography in the setting of blunt trauma is gross hematuria associated with pelvic fracture. They found 85% to 100% of patients with intra- or extraperitoneal bladder rupture had these two findings. They maintain that patients with gross hematuria without a pelvic fracture, microhematuria with a pelvic fracture, and isolated microhematuria only need to be imaged to rule out bladder rupture in the face of clinical indicators such as those mentioned above (69). Diagnosis of bladder rupture is made by a stress retrograde cystourethrogram. The key element of this study is filling the bladder to the point where it is fully distended. This ensures the entire bladder is filled and there is no collapsed segment masking a tear that can

Figure 11 Psoas hitch: (A) Contralateral bladder freed, cystotomy made, and the bladder pexed. (B) Ureter anastomosed at area of psoas hitch (NB ureter yet to be stented). Source: From Ref. 56.

614

Sandman and Lopez

Figure 12 Boari flap: (A) Ipsilateral bladder flap identified, (B) raised and ureter reimplanted, and (C) bladder flap tabularized. Source: From Ref. 56.

yield a false negative test result. Deck et al. performed CT stress cystourethrogram in three steps (14). First, 100 mL of contrast was instilled into the bladder in a retrograde fashion and initial images obtained. Next, the bladder was filled to maximum capacity tolerated (with nonresponsive patients the bladder was filled to a water pressure of 40 cm and 350 mL of contrast) and repeat images were obtained. Last, postdrainage films were obtained. Plain films with anterior–posterior, oblique views, and postvoid films are acceptable (62). The advantage of CT is that it allows evaluation of the bladder concurrently, with the imaging commonly obtained after abdominal and GU trauma. On the other hand, plain films are significantly less expensive than CT scans.

On plain films, extraperitoneal rupture, a flame-like extravasation, would be seen, while intraperitoneal rupture distinguishes itself as superior extravasation from the dome of the bladder. On CT, extraperitoneal bladder rupture will outline the various extraperitoneal fascial planes (Figs. 13 and 14) while intraperitoneal rupture will outline abdominal contents such as bowel (Figs. 15 and 16) (71). Ultrasound imaging is often nondiagnostic as extraperitoneal rupture is not seen via sonography and intraperitoneal rupture is often missed (72). Diagnostic peritoneal lavage does not provide enough

Figure 13 Extraperitoneal bladder rupture. Note wisp of contrast exiting left side of bladder (arrow) and feathery appearance of contrast.

Figure 14 Contrast extravasating between the layers of the rectus and layers of the abdominal wall (arrows). Source: From Ref. 71.

Chapter 49: Complications of Genitourinary Trauma

615

Figure 15 Intraperitoneal bladder rupture seen on CT scan with delayed imaging. Note the contrast exiting the dome of the bladder (thick arrow in A) and the multiple bowel loops outlined and contrast collections in pericolic gutters (B). Abbreviation: CT, computed tomography.

information to rule out bladder injury either as it does not sample the extraperitoneal area.

Management Surgical intervention is generally not indicated in extraperitoneal rupture and urethral catheter drainage is the standard of care according to Corriere and Sandler, though they advocate surgical repair if the patient is going to the OR for other injuries (65,73). In one study (73), they present 39 patients with extraperitoneal bladder trauma due to blunt injury who were treated with either urethral catheter alone (30) or both a urethral and suprapubic catheter (SPC) (9). Extravasation was resolved in all the patients, 34 within 10 days, the other five ranging

Figure 16 Contrast outlining small bowel loops (arrows) and the posterior peritoneal fascia (arrowheads). Source: From Ref. 71.

between 14 and 90 days. There were no complications and no prophylactic antibiotics were used. A follow-up cystogram was obtained after 10 to 14 days with nonoperative management and again after three weeks if there was persistent extravasation of contrast. When surgical closure was performed, the follow-up cystogram was done after one week. In both cases, the catheter was removed when there was no extravasation of contrast (74). Some authors believe all patients with bladder rupture should be started on empiric antibiotics despite the results in the study discussed above (1,75). Intraperitoneal bladder ruptures are repaired with absorbable sutures in a two-layer, running fashion (Fig. 17). Operative repair is necessary due to the possibility of uroascitis and the concomitant electrolyte disturbances that occur with intraperitoneal urine (73). Catheter drainage alone is believed to be insufficient because the urine will preferentially drain through the larger rent in the bladder. Standard approach to the bladder (70) is through a low vertical midline incision (Fig. 17). Taking care to avoid any possible pelvic hematoma, a midline cystotomy is created. Following inspection and identification of any lacerations, the outermost layer is repaired with an absorbable suture, followed by the same type of repair in the additional one or two layers (Fig. 18). If the ureteral orifices are in proximity to the wound, a stent should be placed. Placing both suprapubic and urethral catheters has been the standard practice; however, it is now believed that postoperative urethral catheterization alone may suffice with no increase in posttraumatic complications (77,78). All penetrating trauma to the bladder should be explored, debrided, and repaired with absorbable sutures in a two-layer, running fashion. Treatment includes postoperative bladder decompression via suprapubic and urethral catheters or just urethral catheters, as described above (70).

616

Sandman and Lopez

Figure 17 (A and B) The relationship of the rectus asdominis muscle to the linea alba and linea semicircularis. (C and D) The transversal is fascia is opened, and the peritoneum is reflected cranially. Source: From Ref. 76.

Complications A persistent leak occurs when there is incomplete closure of the bladder or misdiagnosis of an intraperitoneal rupture as extraperitoneal, or in the case of insufficient catheter drainage. With extraperitoneal bladder injuries, the morbidity associated with prolonged drainage is largely the inconvenience of prolonged catheterization and the possibility of bladder calculi (75). However, intraperitoneal bladder rupture may lead to uroascites and complications including peritonitis, increased breakdown of gastrointestinal anastomosis, respiratory difficulty, and sepsis (74). A pelvic hematoma may become infected, leading to septic complications. The chance of infection may be decreased by administration of broad-spectrum antibiotics if given prophylactically, though they are not always given after bladder repair or catheter treatment (75). Fistula formation, either vesicocutaneous or vesicovaginal, is another possible complication. Kotkin and Koch (75) reported that 2 of their 36 patients with extraperitoneal rupture developed fistulas, though Corriere and Sandler (73) reported no fistulas in their 111 patients. With intraperitoneal rupture, the fistula

Figure 18 Repair of an intraperitoneal bladder rupture; (A) Debridement of devitalized tissue; (B) A simple, full-thickness running closure of the incision with a 3-0 absorbale stich; (C) A second, running Lambert suture is used to invert the anastomosis in a watertight fashion. Source: From Ref. 76.

may develop along the tract of the suprapubic tube. Peters advocates looking for causes of fistula formation (foreign body, infected tract) if the tract does not close with adequate bladder emptying (66). Definitive treatment for fistulas resistant to simple urinary diversion via urethral catheter is the excision of the tract and suturing the bladder in two layers (60). Newer options include fibrin sealant to close fistulous tracts after excision (79). Incontinence, another possible complication of bladder injury, is usually not permanent after bladder laceration, unless the bladder neck is involved. Incontinence is usually transient, a result of irritation of the bladder mucosa, due to the catheter balloon, the catheter, or the laceration itself (60).

URETHRA Epidemiology and Diagnosis The urethra, from a trauma injury perspective, is divided into anterior, referring to the urethra distal to the pelvic diaphragm, and posterior, proximal to

Chapter 49: Complications of Genitourinary Trauma

617

Figure 20 Complete rupture of the urethra with no contrast entering the bladder. Source: From Ref. 83.

would demonstrate extravasation with bladder filling (Fig. 21) (68).

Management

Figure 19 Proximal and distal urethra in male. Source: From Ref. 80.

the diaphragm. In the male, anterior includes the membranous and bulbous urethra, while the posterior includes the prostatic urethra (Fig. 19). Due to its short length in the female, the urethra is treated as a single entity. Urethral injuries are more common in males (64). Anterior urethral injuries are usually due to iatrogenic causes, for example, catheterization, straddle injuries, or penetrating injuries (81). Posterior urethral injuries most often occur with pelvic injuries; in fact, 80% to 90% have an associated pelvic fracture (64). Presenting symptoms include blood at the penile meatus and an inability to void despite having the sensation of a ‘‘full’’ bladder. Anterior urethral injuries may also be associated with a butterfly (i.e., perineal) hematoma or a penile (sleeve) hematoma. Posterior urethral injury is associated with a ‘‘boggy’’ prostate caused by a hematoma occupying the prostatic fossa. In addition, a pelvic fracture associated with blood at the penile meatus should raise the suspicion of a posterior urethral fracture (82). Diagnosis of urethral injury is made by retrograde cystourethrogram. Complete interruption would demonstrate no contrast in the bladder and extravasation into the pelvis (Fig. 20), while incomplete disruption

Urethral stretch injuries should be managed by Foley catheterization for three to four days (83). Incomplete rupture is treated with Foley catheterization. As long as there is a mucosal bridge, the urethra should regenerate and heal. After 10 to 14 days, the Foley catheter is removed and a voiding cystourethrogram to confirm

Figure 21 Retrograde cystourethrogram with extravasation of contrast as well as contrast in the bladder. Source: From Ref. 84.

618

Sandman and Lopez

healing should be obtained (85). Complete anterior disruption is treated by immediate repair, or delayed reconstruction with an SPC providing immediate urinary diversion. With most anterior urethral injuries, primary realignment is the method of choice, given the patient’s clinical situation allows for repair and the tissue loss is not extensive enough to prevent a tension-free anastomosis (86). Straddle injuries to the anterior urethra and any other injury that obliterates the urethra are not amenable to primary repair because there is insufficient length to provide a tension-free anastomosis despite mobilization of the urethra. In these cases, the urine is drained via SPC and the urethra is reconnected at least six weeks after the trauma (81). Management of posterior urethral injuries is more controversial. The goals of management are to correct the defect while minimizing complications, namely, incontinence, stricture, and impotence. The two most common strategies, primary realignment versus delayed reconstruction with urinary diversion via an SPC, minimize certain negative results while increasing others. In a study of 100 males with pelvic fractures and urethral disruption and a review of 771 patients discussed in the literature, Koraitim (87) found that formation of a stricture with the SPC method is almost guaranteed (97% stricture rate), though the rates of incontinence and impotence were the lowest of the various methods discussed (4% and 19%, respectively). The rate of stricture is markedly decreased with primary realignment (53%) and the rate of incontinence was similar to that of SPC (5%); however, patients were at a greater risk of impotence, with a rate of 36%. Primarily, reanastomosing the posterior urethra via direct vision and suturing was found to be unacceptable because this method had much higher rates of impotence (56%) and incontinence (21%) and a virtually identical rate of stricture (49%) when compared to primary realignment (Table 3). Given this data, Koraitim (83) recommends SPC in situations where the patient is unstable, where the defect in the urethra is minimal or incomplete, or when the procedure is technically difficult either due to inexperience of the surgeon or due to anatomic issues. Primary realignment is appropriate in instances of widely separated urethral ends or associated rectal or bladder neck injury. Morey et al. echo these recommendations (85). Recently, performing endoscopic realignment after patient stabilization has been suggested (88,89). This follows the damage control philosophy: stabilization and immediate alleviation of any potential

life-threatening concerns, resuscitation, and then finally reoperation. SPC is done via an open cystotomy, which allows for inspection of the bladder as well (85). Realignment has classically been via a ‘‘railroading’’ technique. This consists of placing one catheter in an antegrade manner through the cystotomy and a urethral via the urethral meatus, suturing the two together in the retropubic space, and bringing a second catheter into the bladder to bridge the defect (Fig. 22) (90). Another option is the sound technique, where one passes an antegrade Davis interlocking sound and has it meet a retrograde sound placed via the meatus. At that point, the antegrade sound is guided out the meatus and facilitates retrograde catheter placement. More recently, endoscopic placement of a bridging catheter to achieve realignment has gained popularity (91,92). It is believed this method decreases disruption of any pelvic hematoma and less trauma to the nerves around the urethra, which, in turn, decreases the risk of infection or impotence. One endoscopic method, described by Kielb et al. (92), begins with performing a flexible cystoscopy to visualize the bladder and the ‘‘puckered opening’’ found after complete disruption of the posterior urethra. The bladder is entered with the cystoscope, a guidewire is passed into the bladder, and the scope is removed. Using the Seldinger technique, a Council

Table 3 Incidence of Complication and Type of Repair After Posterior Urethral Trauma Complication/procedure Suprapubic catheter Primary realignment Primary suturing

Stricture (%)

Incontinence (%)

Impotence (%)

97 53 49

4 5 21

19 36 56

Figure 22 Railroading of catheters to realign a distracted urethra. Source: From Ref. 51.

Chapter 49: Complications of Genitourinary Trauma

619

tip catheter (or a Foley catheter with the tip cut off) is passed over a wire and left in place. If the operator is unable to find the bladder opening, a cystotomy is made and a second scope placed in an antegrade fashion, facilitating recognition of the bladder opening (92). In women, Hemal et al. (93) found that primary repair had acceptable results requiring no reoperation in 17 of 17 patients, and with stricture amenable to nonsurgical resolution in 4 of 17 patients. SPC, on the other hand, was found to be a less desirable treatment modality as it has a high rate of urethrovaginal fistula (two of eight patients) as well as stricture requiring a second, delayed operation (eight of eight patients). They, therefore, advocate performing either realignment or a primary end-to-end reanastomosis. If these options are not feasible, due to the extensive urethral disruption or the patient’s tenuous clinical state, the method of choice is SPC and eventual correction of the inevitable stricture and possible fistula that will follow.

Complications As stated above, the complications of urethral injury include impotence, incontinence, and stricture, and the incidence of each is dependent on the immediate management modality taken (Table 2). The decisive factor leading to posttraumatic impotence seems to be the severity of the initial injury rather than management choice between SPC versus primary realignment. For instance, Asci et al. (94) found no statistically significant difference between initial treatment by SPC versus primary realignment for incontinence or impotence, though there was a difference for stricture (83.3% and 45.0%, respectively). In addition, Jenkins et al. (95) have shown that 10% of patients with small urethral displacement were impotent after suffering trauma, while 25% of those with a widely separated urethra were impotent, lending weight to the fact that initial injury is the determining factor for urethral injury complications. Controversy remains whether the predominant etiology of posttraumatic impotence is due to neural or vascular injury (96–98). The significance of this difference is that the nature of the injury may dictate both initial and subsequent treatment; if neural factors determine later potency, primary realignment, which may damage the neurovascular bundle, would not be done; if there is extensive neural injury, there would be little gained from penile revascularization. Presently, treatment for impotence after urethral trauma tends to be conservative, including intracavernosal therapy and venoocclusive rings with vacuum devices. If evidence of vasculogenic impotence exists, the patient may benefit from penile revascularization (98). If a stricture does develop, regardless of initial treatment, management depends on the length of the stricture. To determine the length of a stricture (99), retrograde urethrography along with antegrade cystourethrography may be performed (Fig. 23).

Figure 23 Cystogram and retrograde urethrogram showing large defect due to a stricture. Source: From Ref. 90.

Alternatively, sonourethrography can determine anterior stricture length, possibly more accurately than an urethrogram. For posterior strictures, magnetic resonance imaging in conjunction with retrograde urethrography and antegrade cystourethrography is helpfultodeterminestricturelengthandpelvicanatomy. If the length is short, reconstructive options are either endoscopic or open. Open treatment consists of primary urethral reanastomosis, in the manner described above. Endoscopically, various options are available, including the ‘‘cut to the light’’ technique, with simultaneously placed antegrade and retrograde rigid cystoscopes, one with a urethrotome blade cutting to the second light source after assuring proper position fluoroscopically (100). Also available is the core technique consisting of passing a guidewire (or needle and then guidewire through the needle) through the stricture and then placing a urethral catheter in a retrograde fashion. The catheter is left in place for three to four weeks and then repeated endoscopy and urethrotomy is done when removing the catheter in order to stabilize the stricture walls (101). There is disagreement about the success of each method. Levine and Wessells (100) do not believe that endoscopic intervention improves upon the results of open urethroplasty. They point to the 100% repeat procedure requirement of the endoscopic group versus

620

Sandman and Lopez

Figure 24 Cystogram and retrograde urethrogram showing a large defect due to a stricture. Source: From Ref. 102.

the 22% in the open group. On the other hand, Goel et al. (101) state the avoidance of an open procedure and its inherent morbidity, coupled with the long-term success rate of endoscopic procedures, regardless of minor reprocedures needed, justifies their use. Open repair, generally undertaken three to six months after the initial injury, consists of several steps (90). First, an antegrade urethral sound is passed into the bladder and then into the urethra with the tip in the proximal stricture. After palpating the sound, the pelvic floor is opened and the sound is exposed. The prostatic urethra is spatulated posteriorly, exposing the verumontanum. If sufficient length is available to perform a tension-free, spatulated anastomosis, it is done at this time (Fig. 24). If not, to get more length, one can free up the distal urethra as far as the ligament of the penis circumferentially. If sufficient length still has not been achieved, the next option is to dissect between the corporal bodies and lay the urethra between the two. To obtain still more length, one may resect the pubis and lay the urethra in the space previously occupied by the bone. Between these three maneuvers (Fig. 25), an additional 7 cm or so can be obtained. Various grafts can be used to bridge the defect as well. The anastomosis should be spatulated and large enough to accommodate a 40-French catheter. Incontinence is another potential complication after pelvic fracture and urethral injury. In men, continence is maintained by two mechanisms: the

Figure 25 Diagram of various maneuvers to increase urethral length. Source: From Ref. 90.

Chapter 49: Complications of Genitourinary Trauma

internal sphincter, composed of the bladder neck, and the external sphincter, composed of intrinsic and extrinsic muscle. After urethral distraction defects, there is often compromise of the external urinary sphincter, either due to the initial insult or due to its subsequent repair (103,104). If compromise occurs, continence is maintained by the bladder neck only (Fig. 26). If the bladder neck is significantly compromised, continence will be affected adversely as well. Iselin and Webster (103) evaluated 15 men with pelvic fractures and four of five patients with incontinence had a bladder neck opening greater or equal to 1.5 cm prior to urethroplasty (the fifth had an opening 0.8 cm long). The six continent patients all had bladder neck lengths less than 1.5 cm. They concluded there was a higher rate of incontinence with greater bladder neck compromise. Treatment for incontinence due to trauma consists of repair of the bladder neck and freeing it from any surrounding fibrotic tissue that may be holding the bladder neck open. Artificial urinary sphincter placement is an option as well (90). Pelvic hematoma is a likely occurrence after pelvic fracture (104). If contaminated through repeated attempts at catheterization, the hematoma may become infected. This in turn increases the likelihood of abscess or microabscess. Microabscess, in turn, cause fistula formation. Management of abscess is usually via percutaneous drainage. Fistulous tracts have to be excised entirely as described above (see section on ‘‘Complications’’ under ‘‘Bladder’’).

MALE REPRODUCTIVE ORGANS Penis Amputation (105) Penile amputation may occur under various circumstances; however, most are self-inflicted. Selfamputation warrants a psychiatric evaluation. Regardless of etiology, the amputated penis has a warm ischemia time of about two to six hours. Management consists of determining whether the possibility of

621

reanastomosis exists. If so, a microvascular repair technique is used and the urine is diverted away from the anastomosis (usually via a suprapubic cystostomy catheter). If the penile remnant is unavailable or not amenable to reattachment, one should debride the penile stump to the bleeding edge with preservation of excess skin. The urethra should be spatulated and the excess skin brought down to cover the distal end of the penile stump (creating a ‘‘neo-corona’’).

Fracture (105) Penile fracture occurs when an erect penis strikes a hard object, most commonly against a partner during vigorous intercourse, and bends at an acute angle resulting in a tear in Buck’s fascia. The patient will experience a sudden ‘‘popping’’ sound and/or sensation accompanied by immediate detumescence of the penis. Treatment includes degloving the penis, evacuation of the hematoma, and primary repair of Buck’s fascia. In addition, as a significant number of penile fractures are associated with urethral damage, evaluation of the urethra is necessary. This may include a urinalysis to evaluate for hematuria, radiologic evaluation, or direct inspection at the time of surgery.

Penetrating Injuries (105) Penetrating injuries, regardless of etiology, should be managed by limited debridement of the devitalized area and primarily reanastomosing the urethra in a spatulated fashion. If the urethra cannot be reattached in a tension-free fashion, it should be marsupialized cutaneously at the distal limit of the proximal urethra. Further debridement should be undertaken as needed when the viable margins declare themselves. Prophylactic antibiotics should be used in all cases. Penetrating urethral trauma should be managed as stated above.

Scrotum and Testicle Avulsion (105)

Figure 26 Normal and compromised sphincter in males. Abbreviations: BN, bladder neck; IUM, intrinsic urethral mechanism; EM, extrinsic muscles. Source: From Ref. 103.

Scrotal avulsion can occur as a result of rotating machine accidents and/or after car or motorcycle accidents (as a form of ‘‘road rash’’). Usually the external spermatic fascia is spared and only the dartos is affected. Initial management is conservative, waiting for the nonviable area to declare itself. At that point, usually 12 to 24 hours later, debridement is undertaken and assessment is made whether to primarily repair the scrotum or to perform a split thickness skin graft. While waiting for the demarcation, the exposed areas should be kept moist and cool with saline packs. There is no need to bury the penis or testicles in thigh pouches. The urethra should be evaluated for injury as well, in the method described above.

622

Sandman and Lopez

Fracture (105) Blunt or penetrating trauma may cause fracture of the testicles. On ultrasound examination, disruption of the tunica albuginea and tubules in a hematocele are seen. Operative management is indicated if the above-mentioned findings are confirmed or if lack of injury to the testicle is not assured. Treatment is surgical drainage of any peri- or intratesticular hematoma, debridement of any extruded or devitalized seminiferous tubules, and closure of the tunica albuginea with an absorbable suture. Conservative management has a significantly higher rate of orchiectomy and is, therefore, not an optimal choice. As a fresh hematocele is difficult to distinguish from a traumatic hydrocele, all traumatic hydroceles should be followed and observed to definitively differentiate it from hematocele.

REFERENCES 1. Skinner EC, Parisky YR, Skinner DG. Management of complex urologic injuries. Surg Clin N Am 1996; 76(4): 861–878. 2. Miller KS, McAninch JW. Radiographic assessment of renal trauma: our 15 year experience. J Urol 1995; 154(8):352–355. 3. Sullivan MJ, Lackner H, Banowsky LHW. Intraperitoneal extravasation of urine: BUN/serum creatinine disproportion. JAMA 1972; 221(5):491–492. 4. Morey AL, McAninch JW, Tiller BK, et al. Single shot intraoperative excretory urography for the immediate evaluation of renal trauma. J Urol 1999; 161:1088–1092. 5. Wah TM, Spencer JA. The role of CT in the management of adult urinary tract trauma. Clin Radiol 2001; 56(4):268–277. 6. Perez-Brayfield MR, Gatti JM, Smith EA, et al. Blunt traumatic hematuria in children. Is a simplified algorithm justified? J Urol 2002; 167(6):2543–2546; discussion 2546–2547. 7. Hoff WS, Holevar M, Nagy KK, et al. Practice Management Guidelines for the Evaluation of Blunt Abdominal Trauma. Eastern Association for the Surgery of Trauma, 2001. 8. Goldman SM, Sandler CM. Upper urinary tract trauma—current concepts. World J Urol 1998; 16: 62–68. 9. Sandler CM, et al. Diagnostic Approach to Renal Trauma. American College of Radiology ACR Appropriateness Criteria, 1998:727–731. 10. Carpio F, Morey AF. Radiographic staging of renal injuries. World J Urol 1999; 17(2):66–70. 11. Haas CA, Brown SL, Spirnak JP. Limitations of routine spiral computerized tomography in the evaluation of bladder trauma. J Urol 1999; 162(1):51–52. 12. Blankenship JC, Gavant ML, Cox CE, Chauhan RD, Gingrich JR. Importance of delayed imaging for blunt renal trauma. World J Surg 2001; 25(12):1561–1564. 13. Brown SL, Hoffman DM, Spirnak JP. Limitations of routine spiral computerized tomography in the evaluation of blunt renal trauma. J Urol 1998; 160(6 Pt 1): 1979–1981.

14. Deck AJ, Shaves S, Talner L, Porter JR. Computerized tomography cystography for the diagnosis of traumatic bladder rupture. J Urol 2000; 164(1):43–46. 15. Ahn JH, Morey AF, McAninch JW. Workup and management of traumatic hematuria. Emerg Med Clin N Am 1998; 16(1):145–164. 16. Brandes SB, McAninch JW. Renal trauma: a practical guide to evaluation and management. Digital Urol J 1996 (http://www.duj.com/Article/McAninch/ McAninch.html). 17. Kawashima A, Sandler CM, Corl FM, et al. Imaging of renal trauma: a comprehensive review. Radiographics 2001; 21(3):557–574. 18. Santucci RA, McAninch JW, Safir M, Mario LA, Service S, Segal MR. Validation of the American Association for the Surgery of Trauma Organ Injury Severity Scale for the Kidney. J Trauma 2001; 50(2): 195–200. 19. Santucci RA, McAninch JW. Grade IV renal injuries: evaluation, treatment and outcome. World J Surg 2001; 25:1565–1572. 20. Moore E, Shackford S, Packter H, et al. Scaling: spleen, liver, and kidney. J Trauma 1989; 29:1664–1666. 21. Thall EH, Stone NN, Cheng DL, et al. Conservative management of penetrating and blunt type III renal injuries. Br J Urol 1996; 77(4):512–517. 22. Husmann DA, Gilling PJ, Perry MO, et al. Major renal lacerations with devitalized fragments following blunt abdominal trauma: a comparison between nonoperative (expectant) versus surgical management. J Urol 1993; 150:1774–1777. 23. Moudouni SM, Patard JJ, Manunta A, Guiraud P, Guille F, Lobel B. A conservative approach to major blunt renal lacerations with urinary extravasation and devitalized renal segments. BJU Int 2001; 87(4): 290–294. 24. Matthews LA, Smith EM, Spirnak JP. Nonoperative treatment of major blunt renal lacerations with urinary extravasation. J Urol 1997; 157(6):2056–2058. 25. Altman AL, Haas C, Dinchman KH, Spirnak JP. Selective nonoperative management of blunt grade 5 renal injury. J Urol 2000; 164(1):27–30. 26. Nicol AJ, Theunissen D. Renal salvage in penetrating kidney injuries: a prospective analysis. J Trauma 2002; 53(2):351–353. 27. Margenthaler JA, Weber TR, Keller MS. Blunt renal trauma in children: experience with conservative management at a pediatric trauma center. J Trauma 2002; 52(5):928–932. 28. Wessells H, McAninch JW. Update on Upper Urinary Tract Trauma. AUA Update Series. Vol. XV. Lesson 14, 1996. 29. Meng MV, Brandes SB, McAninch JW. Renal trauma: indications and techniques for surgical exploration. World J Urol 1999; 17(2):71–77. 30. Gonzalez RP, Falimirski M, Holevar MR, Evankovich C. Surgical management of renal trauma: is vascular control necessary? J Trauma 1999; 47(6):1039–1042; discussion 1042–1044. 31. Kuo RL, Eachempati SR, Makhuli MJ, Reed RL II. Factors affecting management and outcome in blunt renal injury. World J Surg 2002; 26(4):416–419. 32. Whitney RF, Peterson NE. Penetrating renal injuries. Urology 1976; 7(1):7–11.

Chapter 49: Complications of Genitourinary Trauma

33. Moudouni SM, Hadj Slimen M, Manunta A, et al. Management of major blunt renal lacerations: is a nonoperative approach indicated? Eur Urol 2001; 40(4): 409–414. 34. Toporoff B, Sclafani S, Scalea T, et al. Percutaneous antegrade ureteral stenting as an adjunct for treatment of complicated ureteral injuries. J Trauma 1992; 32(4): 534–538. 35. Guerriero WG. Ureteral injuries. Urol Clin N Am 1989; 16(2):237–248. 36. Al-Ali AF, Haddad LF. The late treatment of 63 overlooked or complicated ureteral missile injuries: the promise of nephrostomy and role of autotransplantation. J Urol 1996; 156(6):1918–1921. 37. Knudson MM, Harrison PB, Hoyt DB, et al. Outcome after major renovascular injuries: a Western Trauma Association Multicenter report. J Trauma 2000; 49(6): 1116–1122. 38. Montgomery RC, Richardson JD, Harty JI. Posttraumatic renovascular hypertension after occult renal injury. J Trauma 1998; 45(1):106–110. 39. McAninch JW, Carroll PR. Renal trauma: kidney preservation through improved vascular control—a refined approach. J Trauma 1982; 22(4):285–290. 40. Teigen CL, Venbrux AC, Quinlan DM, Jeffs RD. Late massive hematuria as a complication of conservative management of blunt renal trauma in children. J Urol 1992; 147(5):1333–1336. 41. Carroll PR, Klosterman PW, McAninch JW. Surgical management of renal trauma: analysis of risk factors, technique and outcome. J Trauma 1998; 28(7):1071– 1077. 42. Palmer LS, Rosenbaum RR, Gershbaum MD, Kreutzer ER. Penetrating ureteral trauma at an urban trauma center: 10-year experience. Urology 1999; 54(1):34–36. 43. Preston JM Iatrogenic ureteric injury: common medicolegal pitfalls. BJU International 86(3):313–317. 44. Perez-Brayfield MR, Keane TE, Krishnan A, Lafontaine P, Feliciano DV, Clarke HS. Gunshot wounds to the ureter: a 40-year experience at Grady Memorial Hospital. J Urol 2001; 166(1):119–121. 45. Campbell EW Jr., Filderman PS, Jacobs SC. Ureteral injury due to blunt and penetrating trauma. Urology 1992; 40(3):216–220. 46. Ghali AM, El Malik EM, Ibrahim AI, Ismail G, Rashid M. Ureteric injuries: diagnosis, management, and outcome. J Trauma 1999; 46(1):150–158. 47. Azimuddin K, Milanesa D, Ivatury R, Porter J, Ehrenpreis M, Allman DB. Penetrating ureteric injuries. Injury 1998; 29(5):363–367. 48. Armenakas NA. Current methods of diagnosis and management of ureteral injuries. World J Urol 1999; 17(2):78–83. 49. Mulligan JM, Cagiannos I, Collins JP, Millward SF. Ureteropelvic junction disruption secondary to blunt trauma: excretory phase imaging (delayed films) should help prevent a missed diagnosis. J Urol 1998; 159(1):67–70. 50. Medina D, Lavery R, Ross SE, Livingston DH. Ureteral trauma: preoperative studies neither predict injury nor prevent missed injuries. JACS 1998; 186(6):641–644. 51. Thal, Weigelt, Carrico. Operative Trauma Management: An Atlas. 2nd ed. McGraw Hill, 2002.

623

52. Azimuddin K, Ivatury R, Porter J, Allman DB. Damage control in a trauma patient with ureteric injury. J Trauma 1997; 43(6):977–979. 53. Brandes SB, McAninch JW. Reconstructive surgery for trauma of the upper urinary tract. Urol Clin N Am 1999; 26(1):183–199. 54. Meng M, Friese CE, Stroller ML. Extended experience with laparoscopic nephrectomy and autotransplantation for severe proximal ureter loss. J Urol 2003; 169(4):1363–1367. 55. Mathews R, Marshall FF. Versatility of the adult psoas hitch ureteral implantation. J Urol 1997; 158(6): 2078–2082. 56. Stief CG, Jonas U, Petry KU, et al. Ureteric reconstruction. BJU Int 2003; 91(2):138–142. 57. Ahn M, Loughlin KR. Psoas hitch ureteral reimplantation in adults—analysis of a modified technique and timing of repair. Urology 2001; 58:184–187. 58. Richter F, Irwin RJ, Watson RA, Lang EK. Endourologic management of benign ureteral strictures with and without compromised vascular supply. Urology 2000; 55(5):652–657. 59. Cass AS. Ureteral contusion with gunshot wounds. J Trauma 1984; 24(1):59–60. 60. Brandes SB, McAninch JW. Complications of genitourinary trauma. In: Taneja SS, Smith RB, Ehrlich RM, eds. Complications of Urologic Surgery: Prevention and Management. : WB Saunders, 2001:205–225. 61. Selzman AA, Spirnak JP, Kursh ED. The changing management of ureterovaginal fistulas. J Urol 1995; 153(3):626–628. 62. Corriere JN Jr., Sandler CM. Bladder rupture from external trauma: diagnosis and management. World J Urol 1999; 17(2):84–89. 63. Sandler CM, Goldman SM, Kawashima A. Lower urinary tract trauma. World J Urol 1998; 16:69–75. 64. Dreitlein DA, Suner S, Basler J. Genitourinary trauma. Emerg Med Clin N Am 2001; 19(3):569–590. 65. Corriere JN Jr., Sandler CM. Mechanisms of injury, patterns of extravasation and management of extraperitoneal bladder rupture due to blunt trauma. J Urol 1988; 139(1):43–44. 66. Peters PC. Intraperitoneal rupture of the bladder. Urol Clin N Am 1989; 16(2):279–282. 67. Iverson AJ, Morey AF. Radiographic evaluation of suspected bladder rupture following blunt trauma: critical review. World J Surg 2001; 25(12):1588–1591. 68. Brandes S, Borrelli J Jr.. Pelvic fracture and associated urologic injuries. World J Surg 2001; 25(12):1578–1587. 69. Morey AF, Iverson AJ, Swan A, et al. Bladder rupture after blunt trauma: guidelines for diagnostic imaging. J Trauma 2001; 51(4):683–686. 70. Carroll PR, McAninch JW. Major bladder trauma: mechanisms of injury and unified method of diagnosis and repair. J Urol 1984; 132:254–257. 71. Vaccaro JP, Brody JM. CT cystography in the evaluation of major bladder trauma. Radiographics 2000; 20(5):1373–1381. 72. Bigongiari LR, et al. Trauma to the bladder and urethra. ACR Appropriateness Criteria. American College of Radiology, 1998:733–740. 73. Corriere JN Jr., Sandler CM. Management of the ruptured bladder: seven years of experience with 111 cases. J Trauma 1986; 28(9):830–833.

624

Sandman and Lopez

74. Corriere JN Jr., Sandler CM. Management of extraperitoneal bladder rupture. Urol Clin N Am 1989; 16(2): 275–277. 75. Kotkin L, Koch MO. Morbidity associated with nonoperative management of extraperitoneal bladder injuries. J Trauma 1995; 38(6):895–898. 76. Libertino JA, ed. Reconstructive Urologic Surgery. 3rd ed. Mosby, 1998. 77. Volpe MA, Pachter EM, Scalea TM, et al. Is there a difference in outcome when treating traumatic intraperitoneal bladder rupture with or without a suprapubic tube? J Urol 1999; 161:1103–1105. 78. Parry NG, Rozycki GS, Feliciano DV, et al. Traumatic rupture of the urinary bladder: is the suprapubic tube necessary? J Trauma 2003; 54:431–436. 79. Evans LA, Ferguson KH, Foley JP, Rozanski TA, Morey AF. Fibrin sealant for the management of genitourinary injuries, fistulas and surgical complications. J Urol 2003; 169(4):1360–1362. 80. Gibbs MA, Schneider R. Genitourinary tract and renovascular trauma. In: Ferrera P, et al., eds. Trauma Management: An Emergency Medicine Approach. Vol 1. Mosby Year Book, 2000:317–329. 81. Hernandez J, Morey AF. Anterior urethral injury. World J Urol 1999; 17(2):96–100. 82. Dobrowolski ZF, Weglarz W, Jakubik P, Lipczynski W, Dobrowolska B. Treatment of posterior and anterior urethral trauma. BJU Int 2002; 89(7):752–754. 83. Koraitim MM. Pelvic fracture urethral injuries: the unresolved controversy. J Urol 1999; 161(5):1433–1441. 84. Mundy AR. Pelvic fracture injuries of the posterior urethra. World J Urol 1999; 17:90–95. 85. Morey AF, Hernandez J, McAninch JW. Reconstructive surgery for trauma of the lower urinary tract. Urol Clin N Am 1999; 26(1):49–60. 86. Hall SJ, Wagner JR, Edelstein RA, Carpinito GA. Management of gunshot injuries to the penis and anterior urethra. J Trauma 1995; 38(3):439–443. 87. Koraitim MM. Pelvic fracture urethral injuries: evaluation of various methods of management. J Urol 1996; 156(4):1288–1291. 88. Moudouni SM, Patard JJ, Manunta A, Guiraud P, Lobel B, Guille F. Early endoscopic realignment of post-traumatic posterior urethral disruption. Urology 2001; 57(4):628–632. 89. Jepson BR, Boullier JA, Moore RG, Parra RO. Traumatic posterior urethral injury and early primary endoscopic realignment: evaluation of long-term follow-up. Urology 1999; 53(6):1205–1210. 90. Webster GD, Guralnick ML. Reconstruction of posterior urethral disruption. Urol Clin N Am 2002; 29(2): 429–441, viii.

91. Gheiler EL, Frontera JR. Immediate primary realignment of prostatomembranous urethral disruptions using endourologic techniques. Urology 1997; 49(4):596–599. 92. Kielb SJ, Voeltz ZL, Wolf JS. Evaluation and management of traumatic posterior urethral disruption with flexible cystourethroscopy. J Trauma 2001; 50(1):36–40. 93. Hemal AK, Dorairajan LN, Gupta NP. Posttraumatic complete and partial loss of urethra with pelvic fracture in girls: an appraisal of management. J Urol 2000; 163(1):282–287. 94. Asci R, Sarikaya S, Buyukalpelli R, Saylik A, Yilmaz AF, Yildiz S. Voiding and sexual dysfunctions after pelvic fracture urethral injuries treated with either initial cystostomy and delayed urethroplasty or immediate primary urethral realignment. Scand J Urol Nephrol 1999; 33(4):228–233. 95. Jenkins BJ, Badenoch DF, Fowler CG, Slandy JP. Longterm results of treatment of urethral injuries in males caused by external trauma. Br J Urol 1992; 70(1):73–75. 96. Armenakas NA, McAninch JW, Lue TF, Dixon CM, Hricak H. Posttraumatic impotence: magnetic resonance imaging and duplex ultrasound in diagnosis and management. J Urol 1993; 149(5 Pt 2):1272–1275. 97. Machtens S, Gansslen A, Pohlemann T, Stief CG. Erectile dysfunction in relation to traumatic pelvic injuries or pelvic fractures. BJU Int 2001; 87(5):441–448. 98. Shenfeld OZ, Kiselgorf D, Gofrit ON, et al. The incidence and causes of erectile dysfunction after pelvic fractures associated with posterior urethral disruption. J Urol 2003; 169(6):2173–2176. 99. Gallentine ML, Morey AF. Imaging of the male urethra for stricture disease. Urol Clin N Am 2002; 29(2):361–372. 100. Levine J, Wessells H. Comparison of open and endoscopic treatment of posttraumatic posterior urethral strictures. World J Surg 2001; 25(12):1597–1601. 101. Goel MC, Kumar M, Kapoor R. Endoscopic management of traumatic posterior urethral stricture: early results and followup. J Urol 1997; 157(1):95–97. 102. Morey AF, McAninch JW. Reconstruction of posterior urethral disruption injuries: outcome analysis in 82 patients. J Urol 1997; 157(2):506–510. 103. Iselin CE, Webster GD. The significance of the open bladder neck associated with pelvic fracture urethral distraction defects. J Urol 1999; 162(2):347–351. 104. Turner-Warwick R. Prevention of complications resulting from pelvic fracture urethral injuries and from their surgical management. Urol Clin N Am 1989; 16(2):335–358. 105. Jordan GH. Lower Genitourinary Tract Trauma and Male External Genitalia Trauma Parts 2–3. Lessons 11 and 12. Vol. 19. AUA Update Series, 2000.

50 Surgical Complications of Kidney–Pancreas Transplantation Gaetano Ciancio, Joshua Miller, and George W. Burke Division of Transplantation, The DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A. Patricia M. Byers Division of Trauma, Burns, and Critical Care, The DeWitt Daughtry Family Department of Surgery, University of Miami, Miller School of Medicine, Miami, Florida, U.S.A.

Simultaneous pancreas–kidney (SPK) transplantation is becoming one of the standard treatment options for patients with type 1 diabetes and end-stage renal disease. More potent immunosuppression protocols, improvements in surgical techniques, and a better understanding of postoperative complications have made SPK transplantation a successful procedure of choice (1–3). Currently, SPK transplantation is offered to patients with type 1 diabetes and end-stage renal disease if there are no absolute contraindications to the procedure. For safety reasons, at our center, we primarily drain the exocrine pancreas and duodenal segment into the bladder. This technique also allows us to measure urinary amylase activity so that we can monitor any changes in the function of the pancreas graft (4). The 10-year survival rates at our center—84% for patients and 76% for pancreas grafts— are among the best-reported long-term survival rates (4).

TRANSPLANTATION Donor Operation The standard donor operation generally includes procurement of the liver, both kidneys, and the whole pancreas with the duodenum. The nasogastric tube is advanced into the duodenum of the donor and is irrigated with 25 mL of 1% povidone-iodine solution, followed by 50 mL of cold saline solution. The duodenum is irrigated with an antibiotic solution that includes amphotericin B. The nasogastric tube is then pulled back into the stomach, the duodenal contents are gently milked out of the duodenal segments, and the proximal and distal segments are staple divided. Care is taken to ensure that there is no distention. A Y vascular graft, composed of the donor’s common, external, and internal iliac arteries or the brachiocephalic trunk (Fig. 1), is stored in a separate

container of University of Wisconsin (UW) solution on the back table (5). The composite pancreas graft is then taken to the back table for further dissection. However, staple lines are not opened. At this time, splenectomy is performed and the duodenal segment is shortened; only the second portion of the duodenum is kept. Both the proximal end and the distal end are divided with staples and oversewn with interrupted 4–0 silk or Prolene1 sutures (Ethicon Suture Co., Sommerville, New Jersey, U.S.A.). Finally, the Y graft is fashioned, the internal iliac artery is anastomosed to the splenic artery of the pancreas, and the external iliac artery is anastomosed to the superior mesenteric artery of the pancreas with 6–0 Prolene sutures (Fig. 2). The portal vein is then dissected free from the pancreatic bed.

Recipient Operation The recipient operation (Fig. 3) begins with a midline incision. The left kidney is revascularized by anastomosing the renal vein to the common or external iliac vein and the renal artery to the common or external iliac artery. A 4-cm opening is made in the dome of the bladder, where the ureteroneocystostomy anastomosis is performed with a short submucosal tunnel. Next, the pancreas–duodenal allograft is revascularized on the right side. The portal vein is anastomosed to the common iliac vein with a 5–0 Prolene suture, and the common iliac artery of the Y graft is anastomosed to the recipient’s external or common iliac artery with a 6–0 Prolene suture. The duodenum is opened to allow drainage and prevent distention; a pancreatic duodenocystostomy (PDC) is then performed with a two-layer hand-sewn technique. Closure of the inner layer is performed with a running 4–0 Vicryl1 suture (Ethicon Suture Co., Sommerville, New Jersey, U.S.A.). The outer layer is then closed with a running 4–0 Prolene suture. No external drains are used, and a Foley catheter is left in place for 7 to 10 days.

626

Ciancio et al.

Surgical complications are more common after pancreas transplantation than after kidney transplantation. Nonimmunological complications of pancreas transplantation account for graft losses in 5% to 10% of cases. These complications usually occur within six months after the transplantation; their impact on loss of the pancreas graft is the same as that of acute rejection (6).

causes of technical failure, graft thrombosis is the most common; its incidence in association with SPK transplantation is 5.5% (7). It may be difficult to determine precisely which risk factor is implicated in the pathogenesis of graft thrombosis because many causal factors have been described. Graft thrombosis may occur for many reasons, such as preterminal donor hypoperfusion, poor preservation, technical or mechanical issues, immunologic problems, sepsis, or hypercoagulable states. In addition, graft thrombosis has been attributed to the hemodynamic changes in blood flow from a high-flow to a low-flow state after ligation of the distal splenic vessels, the superior mesenteric vessels, and all nonpancreatic branches. Graft salvage is unlikely after thrombosis and can occur only when parenchymal damage is minimal. For this reason, some surgeons have attempted to develop surgical techniques for preventing vascular thrombosis. Some have suggested creating a distal arteriovenous fistula (AVF) so that splenic artery flow is increased (8–11). Another alternative is transplanting the pancreas with the spleen so that physiological hemodynamic flow can be maintained (12). The disadvantage of this procedure is the risk of the potentially lethal complication of graft-versus-host disease (13). None of these techniques have consistently prevented early graft thrombosis (14), perhaps because graft perfusion probably remains unchanged despite attempts to increase flow through the larger vessels (15,16).

Thrombosis

Partial Venous Thrombosis

Vascular thrombosis is a very early complication that typically occurs no later than 48 hours and usually within 24 hours after transplantation. Of all potential

We retrospectively reviewed our experience with the outcome and treatment options associated with partial venous thrombosis of pancreas allografts. From July 1994 to April 1997, 66 patients scheduled for SPK transplantation underwent antilymphocyte induction therapy with a monoclonal anti-CD3 preparation (OKT3) and oral or intravenous tacrolimus in the operating room. None of these patients experienced partial venous thrombosis. In contrast, from May 1997 to June 1999, 48 patients underwent induction therapy with intravenous tacrolimus alone or in conjunction with humanized monoclonal antibody to the interleukin (IL)-2 receptor (IL2-rmAb; Daclizumab1) (17). Of these 48 patients, 14 (29%) experienced partial venous thrombosis, which was detected during routine color Doppler ultrasonography. Twelve of these patients had thrombosis of the splenic vein, one had thrombosis of the superior mesenteric vein (Fig. 4), and one had partial thrombosis of the splenic and superior mesenteric veins. We administered tacrolimus intravenously so that we could ensure sufficient concentrations early in the posttransplantation period to avoid acute rejection (18–21). Microvascular changes associated with pancreatic transplantation may also predispose patients to venous thrombosis. Prostacyclin (PGI2) and thromboxane A2 (TXA2), prostanoid derivatives of eicosapolyenoic fatty

Figure 1 Brachiocephalic trunk with an aortic patch (left) or an iliac Y graft (right).

SURGICAL COMPLICATIONS AFTER PANCREAS TRANSPLANTATION

Figure 2 The composite pancreas and duodenal graft, demonstrating the Y graft. The portal vein has been dissected free of the pancreatic bed. Key: Y, brachiocephalic trunk anastomosed to the splenic artery and the superior mesenteric artery of the pancreaticoduodenal graft; P, pancreas; D, duodenum; SA, splenic artery; and SMA, superior mesenteric artery.

Chapter 50: Surgical Complications of Kidney–Pancreas Transplantation

627

Figure 3 Surgical technique of simultaneous pancreas–kidney transplantation with bladder drainage through a midline intraperitoneal approach. The donor kidney is revascularized on the left side by anastomosing the renal vein to the external iliac vein and the renal artery to the external iliac artery. The dome of the bladder is shown with the ureteroneocystostomy and pancreaticoduodenocystostomy anastomoses. The pancreas– duodenal allograft is revascularized on the right side by anastomosing the portal vein to the common iliac vein and the common iliac artery of the Y graft to the recipient common iliac artery.

acids with opposing actions on vascular smooth muscle tone and platelet aggregation, provide a homeostatic mechanism for maintaining the integrity of the circulation (22–24). TXA2 promotes platelet aggregation

Figure 4 Partial thrombosis (arrow) of the superior mesenteric vein detected by routine color Doppler ultrasonography.

and causes vasospasm, whereas PGI2 opposes these actions, inhibiting platelet aggregation and causing vasodilation. Experimental studies noted an increase in the production of TXA2 by the pancreas and a decrease in the ratio of PGI2 to TXA2 after cold ischemia and reperfusion of the pancreas (25,26). Others have noted that the change in the ratio of these prostaglandins may result from the stasis of the blood flow in the splenic vessels (27). Tacrolimus may contribute to this problem by inducing vasospasm and causing microvascular injury (28). Alternatively, it may cause endothelial injury and thrombosis because of alterations in the ratio of TXA2 to prostaglandin PGI2 or because of the release of endothelin (29–31). The potential for increased adhesion of T cells that express IL-2 receptors bound by daclizumab, combined with the known low-flow state of the splenic vein, may provide the setting for venous thrombosis when the endothelial toxic effects of tacrolimus are added to the mix. Studies have reported microvascular changes late in the course of oral tacrolimus-based immunosuppression therapy (32–35). Corry et al. (36) reported that 9 of their 123 patients experienced pancreas thrombosis while receiving intravenous tacrolimus, but the authors attributed the thrombosis to ischemia and reperfusion injury rather than to the use of intravenous tacrolimus. We have also described microangiopathy among recipients of kidney transplants or SPK transplants, who were treated with tacrolimus within two to four weeks after transplantation (31). In each case, the intravenous administration of tacrolimus was discontinued, and immunosuppression

628

Ciancio et al.

was maintained by oral tacrolimus, steroids, and mycophenolate mofetil. Currently, there is limited information about the use of both tacrolimus and daclizumab for patients undergoing SPK transplantation. A multicenter retrospective analysis found only one case of pancreas thrombosis, which was probably not related to the use of tacrolimus and daclizumab as induction therapy (37). No clinical symptoms of thrombosis, such as graft tenderness, decreased or absent urinary amylase activity, sudden-onset hyperglycemia, hematuria, thrombocytopenia, leukocytosis, or intra-abdominal bleeding, were seen. None of the patients experienced any change in clinical parameters. Superior mesenteric or splenic vein thrombosis was detected incidentally during routine Doppler ultrasonography. Sonography, computed tomography (CT), magnetic resonance imaging, and nuclear medicine scanning have been used to diagnose pancreatic graft thrombosis (38–42). We have found color Doppler ultrasonography to be helpful and cost-effective in diagnosing and assessing venous thrombosis of the pancreatic allograft. More recently, six patients with partial thrombosis of the splenic vein were treated with aspirin and followed up with serial Doppler ultrasonography. None of the partial thromboses progressed to complete splenic vein or pancreatic graft thrombosis. It is possible that partial splenic vein thrombosis is a selflimited process and does not require heparinization. The administration of aspirin may be sufficient (43).

Complete Venous Thrombosis When partial thrombosis has progressed to complete venous thrombosis, attempts at salvaging the thrombosed pancreas graft have been disappointing (44). Immediate reoperation has been advised so that lifethreatening sequelae can be avoided (44,45). Thrombectomy can be performed through a previous portal anastomosis or through a longitudinal or transverse venotomy in the portal vein (44,46,47). Another approach is resection of the thrombosed segment of the pancreas graft (48). Transplant pancreatectomy with immediate transplantation of a new donor pancreas has also been described in cases of graft thrombosis (47,49). This option requires the absence of infection and the availability of a suitable donor within 24 to 48 hours. Recently, we reported our experience with three recipients of SPK transplants who experienced complete venous thrombosis of the pancreas (Fig. 5). All three underwent surgical thrombectomy followed by immediate systemic anticoagulation with heparin. The splenic vein was opened at the tail of the pancreas, and the superior mesenteric vein was opened at the level of the mesentery or the head of the pancreas. This procedure resulted in the successful salvage of the pancreas allografts (50). After resolution of the clot was demonstrated by Doppler ultrasonography, heparinization was converted to warfarin therapy for three months; after that, aspirin therapy was used. All of

Figure 5 Doppler ultrasonogram showing complete venous thrombosis of the pancreas allograft, including the splenic vein (a) and the superior mesenteric vein (b).

these patients were monitored closely with Doppler ultrasonography and checks of serum and urinary amylase activity. One patient experienced recurrent thrombosis. The allograft was salvaged by percutaneous thrombectomy and urokinase infusion, followed by systemic anticoagulation.

Arterial Thrombosis Arterial thrombosis is less common than venous thrombosis and is usually associated with anastomoses of atherosclerotic vessels. In our series, one patient experienced thrombosis of the superior mesenteric artery. Diagnosis was made by routine Doppler ultrasonography of the pancreas (Fig. 6). Surgical thrombectomy was performed successfully through the head of the pancreas allograft. We placed loops around the recipient iliac artery and the Y arterial graft, but did not have any need to occlude inflow; thus, we were able to avoid further ischemic injury to the pancreatic graft. Another patient experienced late thrombosis of the Y graft, although the function of the pancreas allograft persisted. It is possible that pancreatic function was maintained by collateral flow or neovascularization between the donor graft and the recipient vessels. This patient required no treatment (51).

Prophylaxis of Graft Thrombosis Thrombocytosis has been seen after pancreas transplantation without obvious pathophysiological explanations. Because multiple factors predispose the

Chapter 50: Surgical Complications of Kidney–Pancreas Transplantation

629

pancreatic endocrine insufficiency, hematuria, or a bruit over the graft. The causes of this abnormality are congenital malformation, needle injury during procurement, injury during the back-table preparation, and injury during reperfusion hemostasis in the recipient. Doppler ultrasonography is indicated if endocrine function suddenly deteriorates or if the patient experiences hematuria and pain over the graft. The diagnosis is confirmed by abnormal flow in the pancreatic head (Fig. 7). In our series of patients, four have experienced AVF. Two were treated with surgical correction; the other two were treated with angiography and embolization. Pancreatic endocrine function has been preserved in all four patients (51,60).

Transplant Pancreatitis

Figure 6 Doppler ultrasonogram of the pancreas allograft, showing thrombosis of the superior mesenteric artery (arrow).

vessels of the pancreas graft to thrombosis, platelet inhibitors should be administered during the first two postoperative months (52). Another option for preventing graft thrombosis is the use of prophylactic anticoagulant therapy. Various transplant centers use different therapeutic protocols in an attempt to prevent thrombosis. For example, one group advocates the use of lowmolecular-weight dextran, followed by intravenous heparin and antithrombin III supplementation in combination with long-term administration of acetylsalicylic acid (53,54). Another group uses dextran followed by low-dose aspirin, or a combination of aspirin and dipyridamole (55,56). The routine use of systemic anticoagulation has been controversial. Systemic anticoagulation is accepted therapy when splenic vein thrombosis has been documented, when SPK grafts have come from live donors, and when only the pancreas has been transplanted (43,57,58). However, Sollinger has suggested that the use of systemic anticoagulation does not reduce the incidence of vascular graft thrombosis and may increase the likelihood of postoperative bleeding, which may in turn cause venous compression, thereby actually increasing the risk of graft thrombosis (59).

Arteriovenous Fistula AVF has only recently been recognized as a complication of pancreas transplantation. AVF may cause

Graft thrombosis may occur after the development of reperfusion-induced graft pancreatitis, which is caused when pancreatic blood flow is reduced to a critical level (61). Multiple factors may be related to graft pancreatitis, and Troppmann has described these in detail (62). Donor risk factors include hemodynamic instability, brain injury, and vasopressor administration; procurement injury may be due to excessive intraoperative manipulation. Perfusion injury may also occur when excessive flush volumes or perfusion pressures are used. Also, total cold and warm ischemia times may have an effect on preservation and the occurrence of reperfusion injury. Grewal et al. (63) demonstrated that postoperative treatment of the recipient with calcium-channel blockers, combined with the administration of steroids to the donor at the time of procurement, protects against the development of pancreatitis. Animal models of pancreatitis demonstrate that the microcirculation is impaired as pancreatitis

Figure 7 Doppler ultrasonogram of the pancreas allograft, showing increased pulsatile flow within the area of the body of the pancreas (arrow), a finding diagnostic of a arteriovenous fistula.

630

Ciancio et al.

progresses, and this impairment leads to necrosis and thrombosis of the pancreas (64–66). Although these findings do not necessarily reflect allograft pancreatitis as associated with transplantation, they could explain some of the events that occur in association with pancreatic graft edema. With the introduction of UW (or Belzer) solution as a perfusate for pancreatic transplantation, pancreatic graft edema after reperfusion is less pronounced (67,68). The introduction of this solution was a pivotal contribution to the reduction of graft loss due to thrombosis during the early postoperative period (69). Reduced preservation time may also help reduce the incidence and the severity of reperfusion injury and edema of the pancreas allograft; however, other large studies have not found that cold ischemia time affects graft survival (62,70). The incidence of allograft pancreatitis during the postoperative period has decreased because of improvements in procurement techniques that use the spleen and duodenum as handles that avoid pancreatic manipulation, the introduction of Belzer preservation solution, decompression of the portal system during in situ flushing, and intraperitoneal transplantation of the whole organ with exocrine drainage. However, allograft pancreatitis is still an important cause of morbidity. When pancreatitis persists after transplantation, a thorough evaluation is necessary so that its cause can be determined (71,72). Reflux pancreatitis, high postvoid residual volumes, peripancreatic fluid collections, infectious pancreatitis, and leaks from the duodenal segment may cause hyperamylasemia. A Foley catheter should be placed so that postvoid residual volumes can be checked, and urodynamic studies should be performed so that high postvoid residual volumes and pressures can be detected (73). In addition, abdominal ultrasonography, CT of the abdomen, and cystography should be performed so that the cause of the pancreatitis can be determined. If the results of the work-up are nondiagnostic, a biopsy of the pancreas should be performed so that rejection can be ruled out as a cause of hyperamylasemia.

bladder mucosa. Although the urinary pH is generally alkalotic and will maintain proenzymes in an inactive state, a urinary tract infection may reduce the pH enough to activate these digestive enzymes. In addition, enterokinase in the brush border of the duodenal mucosa may activate the proenzyme trypsinogen and thereby initiate the pancreatic enzyme– activation cascade. Other proteases, such as plasmin, thrombin, and fibrolysin, as well as bacterial enzymes, may also activate the conversion of trypsinogen to trypsin. The severe burning and dysuria caused by the resultant urethritis are attributed to urethral autodigestion by the activated pancreatic enzymes trypsinogen and chymotrypsinogen. If untreated, these symptoms may progress to urethral disruption or stricture. Treatment of urethral complications requires both enteric conversion and urological expertise (75). Fortunately, this complication has become less common during the last 15 years. Metabolic acidosis is caused by the excretion from the bladder of large quantities of alkaline pancreatic secretions. Most patients require supplemental oral sodium bicarbonate once oral intake is tolerated; this treatment will minimize the degree of acidosis. With time, some of these patients may be able to decrease their need for oral sodium bicarbonate. Fluid management can become problematic for these patients because of the potential for relatively large volume losses. Patients are at risk of episodes of dehydration, which can be worsened by poor intake as the result of gastric-motility problems commonly associated with diabetes. The symptoms from dehydration can be further increased when patients with diabetes have preexisting orthostatic hypotension because of autonomic neuropathy. Fluid balance can be improved in some patients by the postoperative administration of fludrocortisone acetate for three to six months. Of our 200 patients who underwent

Complications Associated with Bladder-Drained Pancreas Transplantation Bladder-drained pancreas transplantation is associated with multiple urologic (73,74) and metabolic complications. Published reports have shown that 14% to 50% of patients require enteric conversion; nearly 8% of our 200 consecutive patients who underwent SPK transplantation required enteric conversion. Hematuria occurs very frequently and may be caused by the formation of a bladder stone on the staple or suture line. Approximately 30% of patients will require interventions such as Foley catheter placement, irrigation, and cystoscopy for evacuation of clots. Urinary tract infections are common; they occur in as many as half of all cases and are probably induced by the irritating effect of the exocrine secretion on the

Figure 8 Cystogram showing a late leak from the duodenum of the pancreas allograft. Key: L, leak; D, duodenum; P, perforation.

Chapter 50: Surgical Complications of Kidney–Pancreas Transplantation

631

CASE REPORT A 42-year-old woman with type 1 diabetes mellitus and renal failure underwent SPK transplantation with drainage into the urinary bladder. Three weeks postoperatively, a decrease in urinary amylase activity was noted. Doppler ultrasonography demonstrated thrombosis of the superior mesenteric vein. The superior mesenteric vein was opened at the level of the head of the pancreas, and surgical thrombectomy was performed. The patient was treated with heparin after the procedure until a therapeutic level was achieved and then switch to Coumadin1. The patient was then kept on Coumadin for three months. The patient is now taking aspirin daily and the graft is functioning well.

SPK transplantation with bladder drainage, 30% to 40% were readmitted within the first year after transplantation for correction of acidosis and dehydration. Their serum creatinine concentrations usually returned to baseline after the administration of intravenous fluids with bicarbonate. Occasionally, patients will experience rejection exacerbated by episodes of dehydration with consequent increases in the serum concentrations of creatinine. These patients will require conversion to enteric drainage of the pancreatic secretions. Urine leaks due to breakdown of the duodenal segment may occur years after transplantation, but this complication is usually encountered within the first two or three postoperative months. The causes of early urine leaks are technical in nature and usually require surgical correction with prolonged Foley catheter drainage. Late leaks (Fig. 8) can be caused by high pressure in the duodenum during urination. The onset of abdominal pain with elevated serum amylase activity, which can mimic reflux pancreatitis or acute rejection, is a typical presentation. Supporting imaging studies using cystography or CT may be necessary for confirming the diagnosis. Operative intervention may be required and includes reanastomosis to the bladder. Late leaks may develop as the result of rejection and can be treated successfully with Foley catheter drainage. In our series of 200 patients, two patients experienced early postoperative leaks. One patient experienced a disruption of the PDC during cystoscopy for hematuria two weeks postoperatively. Another patient experienced a leak as the result of an episode of biopsy-proven acute rejection six weeks after transplantation. Both patients required operative treatment that included a revision of the PDC. Another of our patients experienced a late leak during an episode of rejection; this leak was successfully treated with Foley catheter drainage. Despite these complications, bladder drainage of the pancreatic graft has many advantages. Early and late complications may cause morbidity; however, these complications are rarely lethal because enteroenterostomy can be avoided. Another advantage of bladder drainage is the ability to monitor the patient for graft rejection.

The technique also allows cystoscopic access for biopsy of the duodenal or pancreatic graft and easy access to pancreatic fluid. An immediate decrease in urine amylase activity after pancreas transplantation signals early acute rejection. Six months after transplantation, a decrease in urinary amylase activity may signal late acute rejection. The decrease in urinary amylase activity may be the only clinical indication of a problem, with no change in the serum concentrations of creatinine or glucose or in the activity of serum amylase or lipase. A biopsy of the pancreas should be performed for confirming the diagnosis of rejection. Using this algorithm at our center, we have not yet lost a pancreas graft to rejection. After the administration of rejection therapy with steroids, the need for repeat pancreatic biopsy can be determined by measuring urine amylase activity. If low urine amylase activity persists after rejection therapy, pancreatic biopsy is indicated. In contrast, if urine amylase activity is normal and the blood-glucose concentration remains high after therapy, the causative factor is steroid therapy rather than rejection, and pancreatic biopsy is not indicated.

Complications Associated with Enteric-Drained Pancreas Transplantation When pancreas transplantation was first performed in the early 1970s, the results of enteric-drainage methods were poor. The small-bowel drainage procedure fell into disfavor because anastomotic leaks with abscess formation and sepsis caused high rates of morbidity and mortality. Recently, more centers are experiencing success with enteric drainage (Fig. 9) because of improvements in donor management, optimized surgical techniques during organ procurement, better preservation solutions, advances in the implantation procedure, and new immunosuppressive drugs (76–81). Enteric-drainage techniques vary in bowel arrangement, the level of anastomosis, the site of the recipient small bowel, and the choice of either a stapled or a hand-sewn anastomosis (79). The most serious complication of enteric-drained pancreas transplantation is a leak at the anastomotic site. This serious problem occurs one to six months

632

Ciancio et al.

transplanting an intact pancreas allograft. Many of the technical problems associated with this procedure have been solved, the incidence of associated thrombosis has diminished, and the management of exocrine secretions with bladder and enteric drainage now provides good results. Immunosuppression therapy aimed at preventing rejection has also improved with the advent of tacrolimus, mycophenolate mofetil, daclizumab, and thymoglobulin (21,84,85). These drugs, when used in combination, have been highly effective in decreasing the incidence of rejection among patients who undergo pancreas transplantation. In a recent study in which 30 patients who underwent SPK transplantation were treated with daclizumab and thymoglobulin as induction therapy in combination with tacrolimus and mycophenolate mofetil or rapamycin, no instances of rejection occurred.

REFERENCES

Figure 9 Surgical technique of simultaneous pancreas–kidney transplantation with enteric drainage through a midline intraperitoneal approach.

after transplantation and causes fever, abdominal discomfort, and leukocytosis. CT scans are helpful in diagnosing the problem. The mandatory treatment is surgical exploration and repair of the enteric leak. Gastrointestinal bleeding may occur at the duodenoenteric suture line as a result of perioperative anticoagulation and inadequate hemostasis. Conservative management may not suffice, and reoperation is usually required (82,83). Enteric drainage has advantages that balance the risk of serious complications associated with this procedure. First because metabolic acidosis and dehydration do not occur, bicarbonate supplementation is not needed. Second, this procedure is obviously not associated with urological complications such as urinary infections, hematuria, bladder stones, and urinary leaks. Third, fewer laboratory tests are required because there is no reason to monitor urinary activity. However, rejection episodes may progress undiagnosed before treatment is started, and this delay increases the possibility of allograft loss.

SUMMARY The primary goal of therapy for type 1 diabetes mellitus is the optimal dosing of insulin and the restoration of normal metabolism. An ideal therapeutic choice for accomplishing this goal is the transplantation of endocrine pancreatic tissue, which can be achieved by

1. Burke G, Ciancio G, Alejandro R, et al. Cholesterol control: long-term benefit of pancreas-kidney transplantation with FK506 immunosuppression. Transplant Proc 1998; 30:513–514. 2. Burke GW, Ciancio G. The renal and pancreatic allograft recipient. In: Kirby RR, Taylor RW, Civetta JM, eds. Handbook of Critical Care. 3rd ed. Philadelphia: Lippincott, 1997:1311–1315. 3. Gruessner AC, Sutherland DE. Report for the International Pancreas Transplant registry-2000. Transplant Proc 2001; 33:1643–1646. 4. Burke GW, Ciancio G, Olson L, Roth D, Miller J. Ten-year survival after simultaneous pancreas/kidney transplantation with bladder drainage and tacrolimus-based immunosuppression. Transplant Proc 2001; 33:1681–1683. 5. Ciancio G, Olson L, Burke GW. The use of the brachiocephalic trunk for arterial reconstruction of the whole pancreas allograft for transplantation. J Am Coll Surg 1995; 181:79–80. 6. Ciancio G, Burke GW, Viciana AL, et al. Destructive allograft fungal arteritis following simultaneous pancreaskidney transplantation. Transplantation 1996; 61: 1172–1175. 7. Gruessner RW, Burke GW, Stratta R, et al. A multicenter analysis of the first experience with FK506 for induction and rescue therapy after pancreas transplantation. Transplantation 1996; 61:261–273. 8. Calne RY, McMaster P, Rolles K, Duffy TJ. Technical observations in segmental pancreas allografting: observations on pancreatic blood flow. Transplant Proc 1980; 12:51–57. 9. Du Toit DF, Reece-Smith H, McShane R, Denton T, Morris PJ. A successful technique of segmental pancreatic autotransplantation in the dog. Transplantation 1981; 31:395–396. 10. Du Toit DF, Heydenrych JJ, Louw G, et al. Intraperitoneal transplantation of vascularized segmental pancreatic autografts without duct ligation in the primate. Surgery 1983; 94:471–477. 11. Duron JJ, Roux JM, Imbaud P, et al. The arteriovenous fistula in segmental pancreatic transplantation in dogs—a hemodynamic study. Transplantation 1987; 44:600–601.

Chapter 50: Surgical Complications of Kidney–Pancreas Transplantation

12. Starzl TE, Iwatsuki S, Shaw BW Jr., et al. Pancreaticoduodenal transplantation in humans. Surg Gynecol Obstet 1984; 159:265–272. 13. Deierhoi MH, Sollinger HW, Bozdech MJ, Belzer FO. Lethal graft-versus-host disease in a recipient of a pancreas-spleen transplant. Transplantation 1986; 41: 544–546. 14. Booster MH, Wijnen RM, van Hooff JP, et al. The role of the spleen in pancreas transplantation. Transplantation 1993; 56:1098–1102. 15. Gooszen HG, van Schilfgaarde R, Terpstra JL. Arterial blood supply of the left lobe of the canine pancreas. II. Electromagnetic flow measurements. Surgery 1983; 93:549–553. 16. Yun M, Inoue K, Kaji H, et al. The hemodynamic time course of the pancreas after segmental autotransplantation in dogs. Transplant Proc 1991; 23:1648–1650. 17. Ciancio G, Cespedes M, Olson L, Miller J, Burke GW. Partial venous thrombosis of the pancreatic allograft after simultaneous pancreas-kidney transplantation. Clin Transplant 2000; 14:464–471. 18. Ciancio G, Lo Monte A, Buscemi G, Miller J, Burke GW. Use of tacrolimus and mycophenolate mofetil as induction and maintenance in simultaneous pancreas-kidney transplantation. Transpl Int 2000; 13:S191–S194. 19. Ciancio G, Burke G, Viciana A, et al. Use of intravenous tacrolimus to reverse vascular rejection in kidney and simultaneous kidney-pancreas transplantation. Transplant Proc 1998; 30:1536–1537. 20. Ciancio G, Burke GW, Roth D, Miller J. Use of intravenous FK506 to treat acute rejection in simultaneous pancreas-kidney transplant recipients on maintenance oral FK506. Transplantation 1997; 63:785–788. 21. Burke GW, Ciancio G, Alejandro R, et al. Use of tacrolimus and mycophenolate mofetil for pancreas-kidney transplantation with or without OKT3 induction. Transplant Proc 1998; 30:1544–1545. 22. Moncada S, Gryglewski R, Bunting S, Vane JR. An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature 1976; 263:663–665. 23. Hamberg M, Svensson J, Samuelsson B. Thromboxanes: a new group of biologically active compounds derived from prostaglandin endoperoxides. Proc Natl Acad Sci USA 1975; 72:2994–2998. 24. Moncada S, Vane JR. Arachidonic acid metabolites and interactions between platelets and blood-vessel walls. N Engl J Med 1979; 300:1142–1147. 25. Johnson BF, Thomas G, Wiley KN, et al. Thromboxane and prostacyclin synthesis in experimental pancreas transplantation. Changes in parenchymal and vascular prostanoids. Transplantation 1993; 56:1447–1453. 26. Kin S, Tamura K, Nagami H, Nakase A. Effect of preservation on blood flow and production of prostacyclin and thromboxane A2 in canine segmental pancreatic autografts. Transplant Proc 1991; 23:1651–1653. 27. Kawai T, Teraoka S, Hayashi T, et al. The changes in prostaglandins after segmental pancreatic transplantation. Transplant Proc 1991; 23:1645–1647. 28. Lieberman KV, Lin WG, Reisman L. FK506 is a direct glomeruloconstrictor as determined by electrical resistance pulse sizing (ERPS). Transplant Proc 1991; 23:3119–3120. 29. Peters DH, Fitton A, Plosker GL, Faulds DT. A review of its pharmacology and therapeutic potential in hepatic and renal transplantation. Drugs 1993; 46:746–794.

633

30. Goodall T, Kind CN, Hammond TG. FK506-induced endothelin release by cultured rat mesangial cells. J Cardiovasc Pharmacol 1995; 26(suppl 3):S482–S485. 31. Burke GW, Ciancio G, Cirocco R, et al. Microangiopathy in kidney and simultaneous pancreas/kidney recipients treated with tacrolimus: evidence of endothelin and cytokine involvement. Transplantation 1999; 68:1336–1342. 32. Randhawa PS, Tsamandas AC, Magnone M, et al. Microvascular changes in renal allografts associated with FK506 (Tacrolimus) therapy. Am J Surg Pathol 1996; 20:306–312. 33. Antoine C, Thakur S, Daugas E, et al. Vascular microthrombosis in renal transplant recipients treated with tacrolimus. Transplant Proc 1998; 30:2813–2814. 34. Morphopathological findings of renal allografts under FK 506 therapy. Japanese FK 506 Study Group. Transplant Proc 1994; 26:1933–1936. 35. Randhawa PS, Shapiro R, Jordan ML, Starzl TE, Demetris AJ. The histopathological changes associated with allograft rejection and drug toxicity in renal transplant recipients maintained on FK506. Clinical significance and comparison with cyclosporine. Am J Surg Pathol 1993; 17:60–68. 36. Corry RJ, Egidi MF, Shapiro R, et al. Tacrolimus without antilymphocyte induction therapy prevents pancreas loss from rejection in 123 consecutive patients. Transplant Proc 1998; 30:521. 37. Bruce DS, Sollinger HW, Humar A, et al. Multicenter survey of daclizumab induction in simultaneous kidney-pancreas transplant recipients. Transplantation 2001; 72:1637–1643. 38. Yang HC, Neumyer MM, Thiele BL, Gifford RR. Evaluation of pancreatic allograft circulation using color Doppler ultrasonography. Transplant Proc 1990; 22:609–611. 39. Snider JF, Hunter DW, Kuni CC, Castaneda-Zuniga WR, Letourneau JG. Pancreatic transplantation: radiologic evaluation of vascular complications. Radiology 1991; 178:749–753. 40. Krebs TL, Daly B, Wong JJ, Chow CC, Bartlett ST. Vascular complications of pancreatic transplantation: MR evaluation. Radiology 1995; 196:793–798. 41. Sebastian A, Cuenca A, Li SF, et al. Pancreas transplant graft evaluation using MIBI scan—a useful tool. Transplant Proc 1998; 30:257–260. 42. Patel B, Markivee CR, Mahanta B, Vas W, George E, Garvin P. Pancreatic transplantation: scintigraphy, US, and CT. Radiology 1988; 167:685–687. 43. Kuo PC, Wong J, Schweitzer EJ, Johnson LB, Lim JW, Bartlett ST. Outcome after splenic vein thrombosis in the pancreas allograft. Transplantation 1997; 64:933–935. 44. Douzdjian V, Abecassis MM, Cooper JL, Argibay PF, Smith JL, Corry RJ. Pancreas transplant salvage after acute venous thrombosis. Transplantation 1993; 56:222–223. 45. Douzdjian V, Abecassis MM, Cooper JL, Smith JL, Corry RJ. Incidence, management and significance of surgical complications after pancreatic transplantation. Surg Gynecol Obstet 1993; 177:451–456. 46. Nghiem DD. Pancreatic allograft thrombosis: diagnostic and therapeutic importance of splenic venous flow velocity. Clin Transplant 1995; 9:390–395. 47. Gilabert R, Fernandez-Cruz L, Real MI, Ricart MJ, Astudillo E, Montana X. Treatment and outcomes of pancreatic venous graft thrombosis after kidneypancreas transplantation. Br J Surg 2002; 89:355–360.

634

Ciancio et al.

48. Fisher RA, Munda R, Madden R. Pancreas transplant functional salvage after segmental vascular thrombosis. Transplant Proc 1993; 25:2138–2140. 49. Ciancio G, Julian JF, Fernandez L, Miller J, Burke GW. Successful surgical salvage of pancreas allografts after complete venous thrombosis. Transplantation 2000; 70:126–131. 50. Boudreaux JP, Corry RJ, Dickerman R, Sutherland DE. Combined experience with immediate pancreas retransplantation. Transplant Proc 1991; 23:1628–1629. 51. Ciancio G, Lo Monte A, Julian JF, Romano M, Miller J, Burke GW. Vascular complications following bladder drained simultaneous pancreas-kidney transplantation; the University of Miami experience. Transpl Int 2000; 13(suppl 1):S187–S190. 52. Hunziker D. Thrombozytose nach Pankreastransplantation—eine retrospektive klinische Studie. Schweiz Rundsch Med Prax 1989; 78:191–196. 53. Hopt UT, Bu¨sing M, Schareck W, et al. Prevention of early postoperative graft thrombosis in pancreatic transplantation. Transplant Proc 1993; 25:2607–2608. 54. Hopt UT, Bu¨sing M, Schareck WD, Becker HD. The bladder drainage technique in pancreas transplantation—the Tu¨bingen experience. Diabetologia 1991; 34:S24–S27. 55. Tibell A, Brattstro¨m, Kozlowski T, Tyde´n G, Groth CG. Management after clinical pancreatic transplantation with enteric exocrine drainage. Transplant Proc 1994; 26:1797–1798. 56. Bynon JS, Stratta RJ, Taylor RJ, Lowell JA, Cattral M. Vascular reconstruction in 105 consecutive pancreas transplants. Transplant Proc 1993; 25:3288–3289. 57. Gruessner RW, Kendall DM, Drangstveit MB, Gruessner AC, Sutherland DE. Simultaneous pancreas-kidney transplantation from live donors. Ann Surg 1997; 226:471–482. 58. Bartlett ST, Kuo PC, Johnson LB, Lim JW, Scheitzer EJ. Pancreas transplantation at the University of Maryland. In: Cecka JM, Teraski PI, eds. Clinical Transplants, 1996. Los Angeles: UCLA Tissue Typing Laboratory, 1997. 59. Sollinger HW. Pancreatic transplantation and vascular graft thrombosis. J Am Coll Surg 1996; 182:362–363. 60. Khan TF, Ciancio G, Burke GW III, Sfakianakis GN, Miller J. Pseudoaneurysm of the superior mesenteric artery with an arteriovenous fistula after simultaneous kidney-pancreas transplantation. Clin Transplant 1999; 13:277–279. 61. Schaapherder AF, van Oosterhout EC, Bode PJ, van der Woude FJ, Lemkes HH, Gooszen HG. Pancreatic graft survival after arterial thrombosis in simultaneous renalpancreatic transplantation. Clin Transplant 1993; 7:37–42. 62. Troppmann C, Gruessner AC, Benedetti E, et al. Vascular graft thrombosis after pancreatic transplantation: univariate and multivariate operative and operative risk factor analysis. J Am Coll Surg 1996; 182:285–316. 63. Grewal HP, Garland L, Novak K, Gaber L, Tolley EA, Gaber AO. Risk factors for postimplantation pancreatitis and pancreatic thrombosis in pancreas transplant recipients. Transplantation 1993; 56:609–612. 64. Bassi D, Kollias N, Fernandez-del Castillo C, Foitzik T, Warshaw AL, Rattner DW. Impairment of pancreatic microcirculation correlates with the severity of acute experimental pancreatitis. J Am Coll Surg 1994; 179:257–263. 65. Fernandez-del Castillo C, Schmidt J, Warshaw AL, Rattner DW. Interstitial protease activation is the central event in progression to necrotizing pancreatitis. Surgery 1994; 116:497–504.

66. Klar E, Messmer K, Warshaw AL, Herfarth C. Pancreatic ischaemia in experimental acute pancreatitis: mechanism, significance and therapy. Br J Surg 1990; 77:1205–1210. 67. Belzer FO, Southard JH. Principles of solid-organ preservation by cold storage. Transplantation 1988; 45:673–676. 68. Wahlberg JA, Love R, Landegaard L, Southard JH, Belzer FO. 72-hour preservation of the canine pancreas. Transplantation 1987; 43:5–8. 69. D’Alessandro AM, Kalayoglu M, Sollinger HW, Pirsch JD, Southard JH, Belzer FO. Current status of organ preservation with University of Wisconsin solution. Arch Pathol Lab Med 1991; 115:306–310. 70. Morel P, Gillingham KJ, Moundry-Munns KC, Dunn DL, Najarian JS, Sutherland DE. Factors influencing pancreas transplant outcome: Cox proportional hazard regression analysis of a single institution’s experience with 357 cases. Transplant Proc 1991; 23: 1630–1633. 71. Ciancio G, Burke GW, Roth D, Luque CD, Coker D, Miller J. Reflux pancreatitis after simultaneous pancreaskidney transplantation treated by a-1 blocker. Transplantation 1995; 60:760–761. 72. Ciancio G, Montalvo B, Roth D, Miller J, Burke GW. Allograft pancreatic duct dilatation following bladder drained simultaneous pancreas-kidney transplantation: clinical significance. JOP 2000; 1:4–12. 73. Ciancio G, Burke G, Lynne C, et al. Urodynamic findings following bladder-drained simultaneous pancreas-kidney transplantation. Transplant Proc 1997; 29:2912–2913. 74. Ciancio G, Burke GW, Nery J, et al. Urologic complications following simultaneous pancreas-kidney transplantation. Transplant Proc 1995; 27:3125–3126. 75. Ciancio G, Burke GW, Nery JR, Coker D, Miller J. Urethritis/dysuria after simultaneous pancreas-kidney transplantation. Clin Transplant 1996; 10:67–70. 76. Ciancio G, Burke G, Roth D, Miller J. Tacrolimus. Curr Opin Transplant 1997; 2:62–67. 77. Zucker K, Rosen A, Tsaroucha A, et al. Unexpected augmentation of mycophenolic acid pharmacokinetics in renal transplant patients receiving tacrolimus and mycophenolate mofetil in combination therapy, and analogous in vitro findings. Transpl Immunol 1997; 5:225–232. 78. Ciancio G, Burke GW, Miller J. Current treatment practices in immunosuppression. Expert Opin Pharcother 2000; 1:1307–1330. 79. Di Carlo V, Castoldi R, Cristallo M, et al. Techniques of pancreas transplantation through the world: an IPITA Center survey. Transplant Proc 1998; 30:231–241. 80. Ciancio G, Burke GW, Roth D, et al. Tacrolimus and Mycophenolate Mofetil as primary immunosuppression for renal allograft recipients. In: Racusen LC, Solez K, Burdick JF, eds. Kidney Transplant Rejection: Diagnosis and Treatment. New York: Marcel Dekker, 1998:519–529. 81. Stratta RJ, Gaber AO, Shokouh-Amiri MH, et al. A prospective comparison of systemic-bladder versus portal-enteric drainage in vascularized pancreas transplantation. Surgery 2000; 127:217–226. 82. Reddy KS, Stratta RJ, Shokouh-Amiri MH, Alloway R, Egidi MF, Gaber AO. Surgical complications after pancreas transplantation with portal-enteric drainage. J Am Coll Surg 1999; 189:305–313.

Chapter 50: Surgical Complications of Kidney–Pancreas Transplantation

83. Ciancio G, Burke G, Roth D, Tzakis AG, Miller J. Update in transplantation 1997. In: Cecka JM, Terasaki PI, eds. Clinical Transplants, 1997. Los Angeles: UCLA Tissue Typing Laboratory, 1998:241–264. 84. Ciancio G, Miller J, Burke GW. The use of intravenous tacrolimus and mycophenolate mofetil as induction and maintenance immunosuppression in simultaneous

635

pancreas-kidney recipients with previous transplants. Clin Transplantation 2001; 15:142–145. 85. Ciancio G, Miller A, Burke GW, et al. Dacliximab induction for primary kidney transplant recipients using tacrolimus, mycophenolate mofetil, and steroids as maintenance immunosuppression. Transplant Proc 2001; 33:1013–1014.

51 Complications of General Surgery During Pregnancy Raymond P. Compton Paris Surgical Specialists, Paris, Tennessee, U.S.A.

Although pregnancy is described as the most natural thing that can occur to a woman, it is unfortunately often complicated by events out of her control. Approximately 1 in 500 women will, during pregnancy, suffer some type of complication that must be addressed by a general surgeon (1). Such complications and their treatment are often affected by the altered physiology of pregnancy. The surgeon must now consider two lives instead of one. The well being of the mother and the treatment method that will least affect the fetus are of prime importance. The surgeon must always remember, however, that it is impossible to bring a fetus to term without a living mother. The conditions affecting pregnant women that may require general surgery are, for the most part, the routine day-to-day conditions faced in the general surgery practice. These conditions include appendicitis, cholecystitis, bowel obstruction, pancreatitis, thyroid disorders, ulcer disease, breast cancer, melanoma, colon cancers, and anorectal disorders that are particular to the complications of delivery. For the most part, the best outcome can be expected when prudent treatment is administered in a fashion similar to that offered to women who are not pregnant.

GENERAL PRINCIPLES OF MANAGEMENT Laparoscopy or Open Procedures When general surgeons began performing laparoscopy, pregnancy was usually listed as an absolute contraindication for most procedures. Over the past 15 to 20 years, it has become clear that pregnant women can safely undergo laparoscopic procedures. Gouldman et al. (2) reported three spontaneous abortions, one occurring two months after surgery, and only one episode of preterm labor in association with 107 cholecystectomies performed on pregnant women. Their summary encompassed 30 published reports with 1 to 20 patients per study. Graham et al. (3) performed a review and reported similar outcomes. The preterm labor rate associated with open cholecystectomy may be as high as 40% (4). Barone et al. (5) reported eight episodes of premature labor associated with 26 open cholecystectomies and only one episode associated

with 20 laparoscopic procedures. In a more recent publication, Affleck et al. (6) reported a similar rate of preterm delivery in association with laparoscopic (10%) and open (12%) cholecystectomy. All of these reports stress that second-trimester procedures carry the lowest risk of preterm delivery or problems associated with processes such as symptomatic cholelithiasis. Often, surgery can be delayed until the second trimester. First-trimester laparoscopy has been established as quite safe, having been performed for many years to rule out ectopic pregnancies (7). Unfortunately, the same results cannot be reported for appendicitis or other inflammatory conditions of the abdomen. Rates of preterm labor approach 20% in association with either open or laparoscopic appendectomy (8). The real risk of an open or a laparoscopic procedure during pregnancy seems to relate more to the disease process than to the actual technique involved. Certain inherent risks are, however, specific to laparoscopy, including the risk of uterine injury during placement of a trocar or a Veress needle, changes in end-tidal CO2 measurements, and changes in maternal–fetal hemodynamics. Hunter et al. (9) have described the effects of pneumoperitoneum on fetal dynamics in the model of a pregnant ewe. In this study, pneumoperitoneum induced by CO2 caused increases in maternal CO2 that led to fetal hypercapnia, tachycardia, and hypertension. End-tidal CO2 was not found to correlate with measured arterial CO2. The Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) has published a list of guidelines for laparoscopic surgery during pregnancy (10). These guidelines are listed in Table 1. The three complications most frequently seen after nonobstetric operations during pregnancy are preterm labor, premature delivery, and spontaneous abortion or fetal demise. Clearly, the last of these complications is by far the worst. The incidence of these problems varies directly with the severity of the underlying disease process. This fact is well illustrated by the spectrum of appendicitis: rates of fetal complications extend from less than 1% in association with early appendicitis to well above 50% in association with frank rupture and peritonitis (11). The same increase

638

Compton

Pneumatic compression devices Lead shield with selective fluoroscopy Dependent positioning Obstetric consultation Intraoperative fetal and uterine monitoring Serial arterial blood gas analyses with entitled CO2 monitoring Second-trimester deferment, if possible Minimal pneumoperitoneum (8–12 mmHg) Open (Hasson) technique

abnormalities. The administration of chloramphenicol to pregnant women is particularly worrisome; because the fetus can only poorly detoxify this antibiotic, high serum concentrations can result. These serum concentrations cause the gray baby syndrome, which often leads to fetal or neonatal demise. Other medications routinely used during the perioperative period may also carry risks to the fetus. Table 2 lists medications routinely used during the perioperative period, classified by risk stratification.

Abbreviation: SAGES, Society of American Gastrointestinal and Endoscopic Surgeons.

Imaging

Table 1 SAGES Recommendations for Laparoscopic Surgery During Pregnancy

in fetal complications is seen when the outcomes of semielective procedures, such as laparoscopic cholecystectomy, are compared with those of emergent procedures such as bowel obstruction. In short, the worse the intra-abdominal inflammatory response, the higher the chance of a serious pregnancy-related complication. General surgeons must keep this underlying principle in mind when counseling patients and family members, both preoperatively and postoperatively. Laparoscopy is safe during pregnancy, but the inherent risks of the disease process remain ever present.

Prophylactic Tocolysis The use of prophylactic tocolytic agents, such as progesterone, magnesium sulfate, and terbutaline, has decreased in the last 10 years. A study of 78 women at the University of North Carolina, Chapel Hill (11), found no measurable benefit in the use of prophylactic tocolytic agents. This study found that uterine contractions occurred twice as often when prophylactic perioperative tocolytic agents were used (84%) than when they were not (47%). These findings were unchanged when reviewed with respect to disease process. Likewise, Hill et al. (12) reported that their 17-year experience at the Mayo Clinic found no benefit in the use of prophylactic perioperative progesterone (12). Tocolytic agents should be used only at the discretion of the obstetrician, usually in response to actual contractions, and should not be used prophylactically.

Medications When choosing medications for the pregnant patient, general surgeons must take into consideration not only the illness to be treated but also the effects on the fetus. Many antibiotics, including most penicillins, cephalosporins, and monobactam antibiotics, are not associated with teratogenic effects, nor are they associated with substantial risk to the developing fetus. Aminoglycosides, tetracycline, and chloramphenicol should be used with caution or not at all during pregnancy. Aminoglycosides are to be avoided because of the potential of ototoxic effects to the fetus. Tetracyclines rapidly cross the placenta and bind to the fetus’s developing teeth and bones, causing discoloration of enamel, enamel hypoplasia, and bony

Diagnostic imaging of pregnant patients can present some difficulty for the surgeon. Clearly, the imaging method of choice is ultrasound. Unfortunately, ultrasound cannot provide the general surgeon with all of the information that may be needed to make a diagnosis. Ultrasound is well accepted for diagnosing biliary and pancreatic pathology. It has also proved to be useful in diagnosing appendicitis. The safety of ultrasound for the pregnant patient and for the fetus is unquestioned. Radiographic imaging must be chosen with somewhat more care. The National Counsel on Radiation Protection has set radiation exposure for the gestating fetus at an absolute limit of 0.5 rem, with a monthly exposure limit of no more than 0.05 rem. Most clinicians are more familiar with measuring X-ray dosage in rad. The term ‘‘rem’’ refers to the biologic affect of a radiation dose. It is the absorbed radiation dose (in rad) times a quality factor based on the biologic nature of the tissue exposure multiplied by a factor ‘‘n’’ that represents the type of radiation delivered. In terms of soft-tissue X-ray exposure, the rem and the rad are essentially equal (13,14). Table 3 lists common radiographs and the organ dose that can be expected for the uterus, fetus, or both (15,16). This situation illustrates an important point of judgment. Although physicians are often reluctant to obtain radiologic studies during pregnancy, it must be remembered that the risk of radiation damage to the fetus is greatest during the first trimester when organogenesis occurs (17). Whenever shielding is possible, it should be used. But a radiographic study, even one that delivers more than the monthly limit of radiation, may be necessary for a decision about the absolute need for surgical exploration or intervention. A cholecystogram (with shielding) would yield a low dose of radiation, and, surprisingly, the helical computed tomography (CT) scan for appendicitis also yields a relatively low dose. The helical CT scan delivers an amount of radiation that is well above the recommended monthly exposure limit but lower than that recommended for the total gestation period. The radiation dose to an embryo for a radiopharmaceutical, such as radiolabeled iodine or technetium, must be calculated on the basis of millicuries administered. These imaging technologies should be used only after consultation with a radiologist or a nuclear medicine physician.

Chapter 51:

Complications of General Surgery During Pregnancy

639

Table 2 Medications Stratified by Class and the Published Risk of Fetal Complications

Table 3 Commonly Performed Radiographs and the Organ Dose that They Deliver to the Uterus, Fetus, or Both

Drug class or druga

Radiograph

mrad

Chest Cervical spinea Thoracic spinea Cholecystogram Lumbar spine KUB Pelvis Helical-appendiceal CT Computed tomographic pelvimetry

0.06 0.01 0.6 5 408 264 194 300 250

Penicillins Cephalosporins Quinolones Aminoglycosides Macrolides Azithromycin, erythromycin Clarithromycin Other antibiotics Aztreonam Clindamycin Metronidazole Sulfamethoxazole/trimethoprim Vancomycin Doxycycline, tetracycline, minocycline Antifungal agents Amphotericin B Amphotericin B lipid complex Fluconazole, itraconazole, ketoconazole H2 blockers Proton pump inhibitors Lansoprazole Omeprazole Anastrozole Anesthetic agents Lidocaine Atracurium, pancuronium Cisatracurium, rocuronium Desflurane, enflurane, sevoflurane Ondansetron Etomidate Ephedrine Anticoagulant or vascular agents Clopidogrel Enoxaparin Warfarin Heparin Other Levothyroxine Glycopyrrolate Albuterol Pain medications Codeine, morphine, fentanyl Acetaminophen Ketorolac Sedatives Propofol Midazolam Temazepam

Riskb B B C D B C B B B C C D

B C C B B C D B C B B B C X B B X C A B C C C B B D X

a

This is not meant to be an exhaustive list but rather to give an idea of classes that are available. b Categories A through D: progressively increasing risk with progressively less knowledge about the drug–pregnancy interaction. Category X: unsafe for use by pregnant patients, should be completely avoided.

Physiologic Changes The accurate diagnosis of intra-abdominal problems is made more difficult by pregnancy because many of the signs and symptoms of these pathologic states mimic findings that are normal during pregnancy, especially during the first trimester. Anatomic and

a

The screening radiographs usually performed after trauma (cervical spine, chest, and pelvis radiographs with shielding) deliver less than the total exposure recommended for the gestational period (0.5 rem) but more than the recommended monthly exposure (0.05 rem). Abbreviations: KUB, kidney, ureter, and bladder; CT, computed tomography. Source: From Refs. 14–16.

physiologic changes during pregnancy may alter the usual presentation of these conditions and can affect the reliability of laboratory studies frequently used to help establish a diagnosis (18). It has long been believed that the anatomic changes that affect the presentation of intraabdominal problems usually occur because of displacement of the abdominal organs by the enlarging uterus. The enlarging uterus is definitely a factor in the ability to accurately diagnose intra-abdominal problems by physical examination. Baer et al. (19), in 1932, originally described the changes in the position of the normal appendix that were assumed to take place during pregnancy (19). Recently, Mourad et al. (20) reviewed 67,000 deliveries of which 67 were complicated by appendicitis. In this series, pain in the right lower quadrant was the most common presenting symptom despite gestational age (first trimester, 86%; second trimester, 83%; and third trimester, 78%). Nevertheless, the enlarging uterus must be kept in mind during physical examination of a pregnant woman. Physiologic changes that are usually present during pregnancy include an expanded intravascular volume, a relative anemia with a decreased heart rate, leukocytosis, and a mild increase in alkaline phosphatase and transaminase activity (18). Despite these changes, a high index of suspicion on the part of the surgeon and open communication between specialists will surely lead to the best outcome. In general, the health of the mother should be the first priority in the treatment of surgical diseases during pregnancy (18).

SPECIFIC DISORDERS IN PREGNANT WOMEN Appendicitis Appendicitis is the most common nonobstetrical surgical diagnosis during pregnancy. It occurs in approximately 1 in 1500 pregnancies (21). Although the incidence of appendicitis among pregnant women

640

Compton

mirrors that among age-matched control subjects, perforated appendicitis is reported more frequently than would be expected (a rate of 43–55%) (22,23). This increased rate is believed to be primarily due to a relative delay in diagnosis. This delay is often attributed to the physiological changes that accompany pregnancy, such as relative leukocytosis, a high incidence of paroxysmal abdominal pain, and other changes occurring in the intra-abdominal anatomy (22). When diffuse peritonitis occurs with appendicitis, it is believed to be facilitated by the inability of the omentum to wall off the infection because of a relative obstruction by the gravid uterus (23). Clearly, the physical finding most frequently associated with appendicitis and pregnancy is pain in the right side of the abdomen. Some controversy exists between classic teaching and current reviews regarding the location of the physical findings in acute appendicitis in pregnancy. It has now been established that the finding of rebound tenderness and abdominal tenderness, whether localized to the right lower quadrant or higher, suggests the diagnosis of appendicitis. The operative treatment may involve either laparoscopy or an open approach. During the third trimester, laparoscopy may be extremely difficult because of the enlarged uterus; the altered port placements may make laparoscopic exposure very difficult. Nonetheless, the number of reports of the use of laparoscopy for appendectomy during the last trimester is increasing (6). If any open procedure is contemplated, the incision is usually placed over the area of maximal tenderness. Maternal death is quite uncommon when appendectomy is performed during the first trimester, but increases in incidence with gestational age and is highly associated with delay in diagnosis and appendiceal perforation. Overall mortality rates should be less than 1% when appendicitis is found and treated in a timely fashion (24,25). Postoperative complications associated with appendectomy during pregnancy include maternal morbidity and mortality, premature labor, and fetal loss. Factors most associated with an increased risk of these complications include the presence of symptoms for more than 24 hours, marked elevation of white blood cell count with significant left shift, and appendiceal perforation at the time of surgery (13). The likelihood of onset of premature labor is similar for both negative laparotomy and appendectomy for early acute appendicitis. When perforation has occurred, fetal loss may occur in as many as 20% of cases or more (13,20).

Cholelithiasis and Cholecystitis Biliary tract disease is the second most common nonobstetrical surgical diagnosis during pregnancy. Neither a clear cause for the frequency of the disease nor a relationship between pregnancy and gall stone formation has been established. It appears that

patients with gall stones who become pregnant are predisposed to symptomatic biliary disease (13). Pregnancy results in physiologic and hormonal changes that lead to an increase in bile stasis and decreased gall bladder contraction (18). Fortunately, the symptoms and physical findings of biliary tract disease are relatively similar for pregnant and nonpregnant women. As noted previously, changes in serum biochemistry that are inherent in pregnancy may make the diagnosis of biliary tract disease more difficult. Ultrasound is the mainstay of imaging for biliary tract disease during pregnancy. Views of the pancreas are often limited because of overlying bowel gas and the relative decrease in the gastrointestinal tract motility that accompanies pregnancy. Although gall stone pancreatitis is not a frequent finding, it does occur and it carries a high rate of complications for both mother and fetus. If patients exhibit the signs and symptoms of cholecystitis, if an ultrasonogram shows stones, and if the biochemical findings are consistent with a diagnosis of pancreatitis, it is reasonable to treat pregnant women for biliary pancreatitis (13). Typically, a patient with symptomatic biliary tract disease can be treated supportively until the second trimester of pregnancy. Barone et al. (5) retrospectively compared the outcomes of open and laparoscopic cholecystectomies performed on pregnant women over a five-year period. Complications were found among patients who underwent either procedure; these complications ranged from spontaneous hepatic rupture to premature contractions. The most frequently occurring complication was premature contraction; there was no statistically significant difference in the frequency of this complication between the open and laparoscopic groups (5). Similar findings have been published by Epstein (24). Barone’s report (5) summarized and compared the results of reports of laparoscopic and open cholecystectomies during all trimesters of pregnancy that were published through March 1999. The rates of spontaneous abortion and premature labor associated with each procedure were similar, as would be expected. It should be stressed that when laparoscopy is performed for any reason during pregnancy, the open Hasson technique should be favored for access, and a low-pressure pneumoperitoneum (12 mmHg or less) should be used. In short, the maternal mortality rates are low, and the rate of complications associated with emergent surgery for biliary tract disease is not substantially higher among pregnant women than among nonpregnant women (13). The overall management of biliary tract disease among pregnant women parallels the management of the disease among patients without symptoms. Indications for early operation include a lack of response to antibiotics, progression of the disease, and other complications of the disease process. Endoscopic retrograde cholangiopancreatography (ERCP) is a valuable imaging tool even for pregnant

Chapter 51:

women. ERCP is a useful adjunct to the treatment of cholecystitis and biliary tract disease, and the procedure can be undertaken without fluoroscopic exposure (26,27). There have even been reports describing the use of ultra-short fluoroscopy for imaging studies of pregnant patients (28). These techniques greatly increase the surgeon’s ability to delay operative treatment until the second trimester. The clinician must be alert to other conditions that may mimic cholecystitis, such as viral or alcoholic hepatitis and the HELLP (hemolysis, elevated liver function, and low platelet count) syndrome (29). A short review and case report by Watson et al. (29) has established the similarity of these conditions to cholecystitis (29). The findings associated with the HELLP syndrome may precede the features of preeclampsia.

Pancreatitis Pancreatitis occurs in approximately 1 in 1500 pregnancies. It is more likely to occur during the third trimester and carries a rate of fetal loss of between 10% and 20% (30). The causes of acute pancreatitis during pregnancy are similar to its causes among the population as a whole, although alcohol is probably a less common cause among pregnant women than among the general population (30). The diagnosis of pancreatitis can usually be made by the biochemical measurement of amylase and lipase activity. The amylase-to-creatinine clearance ratio is also a useful marker among pregnant women (31). During pregnancy, pancreatitis should be managed supportively. Special attention should be paid to patients who experience pancreatitis in relation to hypertriglyceridemia. Pregnancy is known to relatively increase the incidence of hypertriglyceridemia as a cause of pancreatitis. This condition may make it very difficult to provide nutritional support for these patients. If total parenteral nutrition is needed, it may be necessary to delete intralipid from the regimen, using only carbohydrate and protein support for a short period of time until the triglyceride concentration falls to a more acceptable level (32).

Hepatic Rupture If hepatitis is the most common liver disease during pregnancy, spontaneous hepatic rupture is the worst hepatic complication that can occur. It is often seen in association with a high-risk pregnancy and poor prenatal care. It is associated with eclampsia and preeclamptic syndromes (18,33). Hepatic rupture usually manifests itself during the third trimester of pregnancy. Hypertensive patients who complain of symptoms such as headache, third-trimester nausea and vomiting, and mild to moderate epigastric discomfort for several days to weeks are at risk. Mild jaundice and increased alkaline phosphatase concentrations may be seen. An ultrasound will often show a subcapsular hematoma or frank rupture.

Complications of General Surgery During Pregnancy

641

The severity of these symptoms seems to be directly related to the severity of the hypertension (13). Clearly, the management of spontaneous hepatic rupture is similar to that of a traumatic liver injury. This treatment often involves intra-abdominal packing and damage-control laparotomy. If the fetus is at a viable stage of development and has not succumbed to the events surrounding the hepatic rupture, then a Cesarean section at the time of damage-control laparotomy must be considered.

Intestinal Obstruction Intestinal obstruction occurs infrequently, at a rate of 1 in 1500 to 1 in 60,000 pregnancies (18). The most common cause is adhesion. Interestingly, volvulus and intussusception cause intestinal obstruction much more frequently (5–25% of cases) among pregnant women than among nonpregnant patients (18). The surgical approach that provides the best exposure of an intestinal obstruction involves a midline laparotomy incision. Bowel strangulation is a common finding at the time of exploration (34). Studies have found a maternal mortality rate of 5% to 6% and a fetal mortality rate of 25% to 30% (34,35).

Breast Cancer The incidence of breast cancer among pregnant women ranges from 1 in 3000 to 1 in 3500 (36). Many myths surround the treatment of breast cancer during pregnancy, including the value of negative mammographic results, the need for a therapeutic abortion or oophorectomy, and the relatively poorer prognosis associated with pregnancy at the time of diagnosis or subsequent to diagnosis. Negative results from mammography are not necessarily reassuring. A study performed at Memorial Sloan-Kettering Cancer Center found that 22% of women with pregnancy-associated breast cancers had mammographic results that did not demonstrate any radiologic signs of cancer (37). An older series found that six of eight pregnant patients with subsequent biopsy-proven cancer had negative mammographic results (38). Routine therapeutic abortion should not be recommended because no published series has demonstrated any additional benefit associated with this procedure (39). Women should undergo therapeutic abortion only if the issue is fetal damage due to proposed chemotherapy or radiation treatments. Women must make an informed decision about their pregnancies when possible. Oophorectomy cannot be uniformly recommended for the pregnant patient because this procedure is not associated with any survival benefit (40). It has long been held that breast cancer associated with pregnancy has a dismal prognosis and is often untreatable. Three frequently quoted early studies reported these findings. In 1929, Kilgore (41) found a five-year survival rate of only 17% for pregnancy-associated breast cancer. In 1937, Harrington (42) from the Mayo Clinic reported

642

Compton

a five-year survival rate of only 5.7%; he was the first to begin to realize the marked difference in survival associated with nodal status (a five-year survival rate as high as 61% in patients without nodal involvement). In 1943, Haagensen and Stout (43) went so far as to suggest that the survival rate of pregnancyassociated breast cancer was so poor that mastectomy should not be considered. Unfortunately, these reviews were flawed by the late presentation of the disease and its comparison with the breast cancer population as a whole, among other factors. Finally, Petrek et al. (44) was able to demonstrate that the breast cancer survival rates were equivalent among pregnant women and premenopausal patients. In short, nodal status, not pregnancy, is the most significant predicator of 5- and 10-year survival rates (44). In addition, pregnancy after a diagnosis of breast cancer does not detrimentally affect long-term prognosis (45). Most pregnancy-associated breast cancers are first detected as a painless mass. The physician may watch these masses to assess their clinical behavior, but only for a month or two. If the masses do not resolve spontaneously, biopsy is indicated; fineneedle aspiration, core biopsy, or open biopsy may be used. The treatment of choice for the pregnant woman with breast cancer is a modified radical mastectomy. Breast-conserving therapy with ionizing radiation is associated with a much higher risk to the fetus and cannot be considered safe. Most breast cancers that occur during pregnancy are hormonereceptor negative (46). The treatment of pregnant women with breast cancer mirrors the treatment of breast cancer patients who are not pregnant. During the childbearing years, there is no statistically significant difference in survival, stage for stage, between women who are pregnant and those who are not (47). The current recommendation that breast cancer among pregnant women should be treated in the same way as breast cancer among nonpregnant women also holds for the treatment of melanoma and colon cancer among these populations. This treatment recommendation is well established but must incorporate a multidisciplinary approach (48–50).

Ulcer Disease The true incidence of ulcers during pregnancy is unknown. The symptoms of this disease among pregnant women are similar to those among nonpregnant women. Surgical intervention is really reserved for the complications associated with the disease. Bleeding gastric or duodenal ulcers must be considered a surgical emergency. Given the increased intravascular volume of the pregnant patient and the relative increase in hematocrit, stress to the fetus may occur before hemodynamic changes are apparent in the mother. For several reasons, ulcers are less common among pregnant women than among other populations (51).

First, the diagnosis of ulcer is often not definitive because pregnant women are fairly reluctant to undergo diagnostic studies to confirm the diagnosis. Next, there is an increase in plasma histaminase levels during pregnancy, but whether this increase is associated with a decreased incidence of ulcer disease is questionable (52). Finally, female sex hormones, namely progesterone, seem to increase gastric and duodenal mucosal protection against stomach acid by stimulating mucin production (51). However, it is important to stress that if a pregnant patient exhibits signs and symptoms of peptic ulcer disease, such as severe pain, vomiting, or bleeding, diagnostic (and possibly therapeutic) endoscopy should be undertaken. The treatment of ulcer disease in a pregnant woman does present some problems. Pregnant women commonly take antacids because of gastroesophageal reflux, and these drugs seem to be safe. The use of sucralfate also seems to be relatively safe; this drug is listed as a category B drug for pregnant patients (Table 3). H2-receptor blockers are associated with variable risks, and proton pump inhibitors should be avoided during pregnancy (51). It is rare, though, that the pregnant patient will require surgery solely because of intractable symptoms. It is much more likely that surgery will be required because of perforation or hemorrhage. When perforation occurs, the risk to the unborn fetus is high. Some series report a fetal mortality rate of 21%, but this rate is much better than that associated with medical treatment of perforated ulcer, which approaches 65% (51). Likewise, the fetal mortality rate is much lower when pregnant women are treated surgically for bleeding peptic ulcers (14%) than when they are treated medically (44%) (51). Surgical options include exploration and endoscopic methods of control.

Anorectal Complications of Delivery Five percent of women require third- or fourth-degree episiotomy during normal vaginal deliveries. These episiotomies are routinely repaired at the time of delivery, but 10% of these repairs will break down. The most common finding associated with such breakdown is complete disruption of the sphincter or the formation of a rectovaginal fistula. The treatment of sphincter disruption involves sphincteroplasty with reconstruction of the perineal body. The treatment of rectovaginal fistula usually involves an endorectal advancement flap of muscle tissue; success rates are good. When the patient is a good surgical risk, colostomy is frequently not needed. Most abscesses that occur will respond to local treatment, including incision and drainage. A fair number will drain spontaneously. Interestingly enough, as many as 30% of sphincter disruptions will heal with medical treatment along with satisfactory return of content rectal function (53).

Chapter 51:

Hyperparathyroidism Hyperparathyroidism is a rare complication during pregnancy but can be catastrophic if not treated. The outcome of untreated maternal hyperparathyroidism is neonatal tetany in the infant because of the rapid placental transport of calcium. After birth, when maternal calcium is no longer available, the infant’s parathyroid cannot mobilize calcium stores and hypocalcemic tetany results. Surgical resection should be undertaken, preferably during the second trimester, to cure hyperparathyroidism among pregnant women (54). Fewer than 150 cases of maternal hyperparathyroidism have been reported. The fetal complications include stillbirth and spontaneous abortion (54).

SUMMARY The diagnosis and management of surgical complications in the pregnant patient challenge the general surgeon due to the altered physiology of pregnancy and the concern for the outcome of both the mother and the fetus. The concern for the fetus will sometimes necessitate a slightly altered approach in the selection of diagnostic modalities and pharmacologic agents. However, in general, the best outcome for the mother will ensure the best outcome for the fetus. A prudent therapeutic approach that would usually be selected for the woman who is not pregnant will result in the best outcome.

REFERENCES 1. Griffen WO Jr, Dilts PV Jr, Roddick JW Jr. Non-obstetric surgery during pregnancy. Curr Probl Surg 1969; Nov:1–56. 2. Gouldman JW, Sticca RP, Rippon MB, McAlhany JC Jr. Laparoscopic cholecystectomy in pregnancy. Am Surg 1998; 64:93–98. 3. Graham G, Baxi L, Tharakan T. Laparoscopic cholecystectomy during pregnancy: a case series and review of literature. Obstet Gynecol Surv 1998; 53:566–574. 4. McKellar DP, Anderson CT, Boynton CJ, Peoples JB. Cholecystectomy during pregnancy without fetal loss. Surg Gynecol Obstet 1992; 174:465–468. 5. Barone JE, Bears S, Chen S, Tsai J, Russell JC. Outcome study of cholecystectomy during pregnancy. Am J Surg 1999; 177:232–236. 6. Affleck DG, Handrahan DL, Egger MJ, Price RR. The laparoscopic management of appendicitis and cholelithiasis during pregnancy. Am J Surg 1999; 178: 523–529. 7. Soper NJ, Hunter JG, Petrie RH. Laparoscopic cholecystectomy during pregnancy. Surg Endocs 1992; 6: 115–117. 8. Al-Mulhim AA. Acute appendicitis in pregnancy. A review of 52 cases. Int Surg 1996; 81:295–297. 9. Hunter JG, Swanstrom L, Thornburg K. Carbon dioxide pneumoperitoneum induces fetal acidosis in a pregnant ewe model. Surg Endosc 1995; 9:272–279.

Complications of General Surgery During Pregnancy

643

10. Board of Governors of the Society of American Gastrointestinal Endoscopic Surgeons (SAGES). Guidelines for laparoscopic surgery during pregnancy. Surg Endosc 1998; 12:189–190. 11. Kort B, Katz L, Watson WJ. The effect of nonobstetric operation during pregnancy. Surg Gynecol Obstet 1993; 177:371–376. 12. Hill LM, Johnson CE, Lee RA. Prophylactic use of hydroxyprogesterone caproate in abdominal surgery during pregnancy. A restrospective evaluation. Obstet Gynecol 1975; 46:287–290. 13. Fallon WF Jr, Newman JS, Fallon GL, Malangoni MA. The surgical management of intra-abdominal inflammatory conditions during pregnancy. Surg Clin North Am 1995; 75:15–31. 14. National Council on Radiation Protection and Measurements. Limitation of exposure to ionizing radiation: recommendations of the National Council on Radiation Protection and Measurements. Bethesda, MD: The Council. 1993; Report 116:9–21. 15. Dietrich MF, Miller KL, King SH. Determination of potential uterine (conceptus) doses from axial and helical CT scans. Health Phys 2005; 88:S10–S13. 16. Ames Castro M, Shipp TD, Castro EE, Ouzounian J, Rao P. The use of helical computed tomography in pregnancy for the diagnosis of acute appendicitis. Am J Obstet Gynecol 2001; 184:954–957. 17. Brent RL. The effect of embryonic and fetal exposure to x-rays, microwaves, and ultrasound: counseling the pregnant and nonpregnant patient about these risks. Semin Oncol 1989; 16:347–368. 18. Firstenberg MS, Malangoni MA. Gastrointestinal surgery during pregnancy. Gastroenterol Clin North Am 1998; 27:73–88. 19. Baer JL, Reis RA, Araens RA. Appendicitis in pregnancy with changes in position and axis of the normal appendix in pregnancy. JAMA 1932; 98:1359–1364. 20. Mourad J, Elliott JP, Erickson L, Lisboa L. Appendicitis in pregnancy: new information that contradicts longheld clinical beliefs. Am J Obstet Gynecol 2000; 182: 1027–1029. 21. Babaknia A, Parsa H, Woodruff JD. Appendicitis during pregnancy. Obstet Gynecol 1977; 50:40–44. 22. Tracey M, Fletcher HS. Appendicitis in pregnancy. Am Surg 2000; 66:555–560. 23. Tamir IL, Bongard FS, Klein SR. Acute appendicitis in the pregnant patient. Am J Surg 1990; 160:571–576. 24. Epstein FB. Acute abdominal pain in pregnancy. Emerg Med Clin North Am 1994; 12:151–165. 25. Mahmoodian S. Appendicitis complicating pregnancy. South Med J 1992; 85:19–24. 26. Zagoni T, Tulassay Z. Endoscopic sphincterotomy without fluoroscopic control in pregnancy. Am J Gastroenterol 1995; 90:1028. 27. Berger Z. Endoscopic papillotomy without fluoroscopy in pregnancy. Endoscopy 1998; 30:313. 28. al Karawi M, Mohamed SA. Therapeutic endoscopic retrograde cholangiopancreatography with ultra-short fluoroscopy: report of two cases. Endoscopy 1997; 29:S31. 29. Watson CJ, Thompson HJ, Calne R. HELLP–It’s not cholecystitis. Br J Surg 1990; 77:539–540. 30. Kline KB. Pancreatitis in pregnancy. In: Rustgi VK, Cooper JN, eds. Gastrointestinal and Hepatic Complications in Pregnancy. New York: John Wiley & Sons, 1986:138.

644

Compton

31. Ordorica SA, Frieden FJ, Marks F, Hoskins IA, Young BK. Pancreatic enzyme activity in pregnancy. J Reprod Med 1991; 36:359–362. 32. Scott LD. Gallstone disease and pancreatitis in pregnancy. Gastroenterol Clin North Am 1992; 21:803–815. 33. Cunningham FG, Clark SL, Gant NF, Leveno KJ, Hauth JC. Williams Obstetrics. 21st ed. New York: McGraw-Hill, 2001. 34. Perdue PW, Johnson HW Jr, Stafford PW. Intestinal obstruction complicating pregnancy. Am J Surg 1992; 164:384–388. 35. Watanabe S, Otsubo Y, Shinagawa T, Araki T. Small bowel obstruction in early pregnancy treated by jejunotomy and total parenteral nutrition. Obstet Gynecol 2000; 96:812–813. 36. Moore JL Jr, Martin JN Jr. Cancer and pregnancy. Obstet Gynecol Clin North Am 1992; 19:815–827. 37. Liberman L, Giess CS, Dershaw DD, Deutch BM, Petrek JA. Imaging of pregnancy-associated breast cancer. Radiology 1994; 191:245–248. 38. Max MH, Klamer TW. Pregnancy and breast cancer. South Med J 1983; 76:1088–1090. 39. Gemignani ML, Petrek JA, Borgen PI. Breast cancer and pregnancy. Surg Clin North Am 1999; 79:1157–1169. 40. Hoover HC Jr. Breast cancer during pregnancy and lactation. Surg Clin North Am 1990; 70:1151–1163. 41. Kilgore AR. Tumors and tumor-like lesions of the breast in association with pregnancy and lactation. Arch Surg 1929; 18:2079–2098. 42. Harrington SW. Carcinoma of the breast: results of surgical treatment when the carcinoma occurred in the course of pregnancy or lactation and when pregnancy

43. 44. 45. 46.

47. 48. 49. 50. 51. 52. 53. 54.

occurred subsequent to operation (1910–1913). Ann Surg 1937; 106:690–700. Haagensen CD, Stout AP. Carcinoma of the breast. Ann Surg 1943; 118:859–870. Petrek JA, Dukoff R, Rogatko A. Prognosis of pregnancyassociated breast cancer. Cancer 1991; 67:869–872. Donegan WL. Breast cancer and pregnancy. Obstet Gynecol 1977; 50:244–252. Elledge RM, Ciocca DR, Langone G, McGuire WL. Estrogen receptor, progesterone receptor, and HER-2/ neu protein in breast cancers from pregnant patients. Cancer 1993; 71:2499–2506. DiFronzo LA, O’Connell TX. Breast cancer in pregnancy and lactation. Surg Clin North Am 1996; 76: 267–278. Walsh C, Fazio VW. Cancer of the colon, rectum, and anus during pregnancy. The surgeon’s perspective. Gastroenterol Clin North Am 1998; 27:257–267. Gallenberg MM, Loprinzi CL. Breast cancer and pregnancy. Semin Oncol 1989; 16:369–376. Dillman RO, Vandermolen LA, Barth NM, Bransford KJ. Malignant melanoma and pregnancy: ten questions. West J Med 1996; 164:156–161. Cappell MS, Garcia A. Gastric and duodenal ulcers during pregnancy. Gastroenterol Clin North Am 1998; 27:169–195. Barnes LW. Serum histaminase during pregnancy. Obstet Gynecol 1957; 9:730–732. Kort KC, Schiller HJ, Numann PJ. Hyperparathyroidism and pregnancy. Am J Surg 1999; 177:66–68. Mestman JH. Parathyroid disorders of pregnancy. Semin Perinatol 1998; 22:485–496.

Index

Abdominal compartment syndrome (ACS), 62–63, 78 Abdominal gynecologic surgery, 553 Abdominal organs, displacement of, 639 Abdominal trauma, 399–401 Abdominal wall defects, 70, 78–79 Abdominal wall integrity, 78 Abdominal wall surgery, complications of, 77 Abdominal wound closure, complications of, 77 complications of, 78–79 Abdominoperineal resection (APR), 102 Abelcet1, 182, 375 ABIOMED BVS 50001, 367–368 ABIOMED VAD, 368 Abortion spontaneous. See Fetal demise therapeutic, 641 Abscess, 145 formation, 138 Absorbable meshes, 79 Acetaminophen (IV) derivatives, 469 Acidosis, 47 metabolic, 630 ACS. See Abdominal compartment syndrome Acticoat1, 534 Activated clotting times (ACT), 370–371 Acute cellular rejection (ACR), 179, 187 diagnosis, 185 Acute lung injury (ALI), 271 Acute pancreatitis, complications, 148 Acute rejection accelerated, 360 classification of, 292 Acute stress disorder (ASD), 516 Acyclovir crystals, 36 Addison’s disease, 209 Adenosine triphosphate (ATP), 19 Adhesiolysis, 96 Adrenal autotransplantation, 209 Adrenal incidentalomas, 207 Adrenal metastases, 208 Adrenal tumors, 212 Adrenalectomy, 174 complications of, 208–211 indications for, 207 laparoscopic, 212–213 Adrenocortical adenocarcinoma, 207 Adrenocortical adenomas, 207 Adrenocortical insufficiency, 63 Adrenocorticotrophic hormone (ACTH), 208 Adult respiratory distress syndrome (ARDS), 179, 271, 275 fluid management for, 280 identification, 276 immunomodulation in, 280 inhaled nitric oxide in the management of, 278

[Adult respiratory distress syndrome (ARDS)] steroids in, 279–280 treatment of, 276 Aerobic–anaerobic infections, mixed, 31 Aeromonas infection, 531 Air embolism coronary, 329 sources of, 328 systemic, 328–329 Air leaks, categories of, 234 Air lock, venous, 328 Airway, breathing, circulation (ABC) algorithm, 299 Airway management, complications of, 2–4 Airway pressures, limitation of, 277 Airway trauma, 3 Albumin solutions, 25 Aldomet1, 49 Aldosteronomas, 207 Alkaline reflux gastritis, 119–120 diagnosis of, 120 Allen’s test, 430–431, 435, 549 Allis clamps, 98 AlloDerm1, 524 Allograft, 524 rejection, 360 Alpha-adrenergic stimulation, 335 Alveolar–pleural fistula, 234 Alveolar overdistention, 274 Alveoli, flooding of, 271 Ambisome1, 182 American Society of Anesthesiologists (ASA) algorithm, 2 American Spinal Injury Association (ASIA), 493 Amikacin therapy, 34 Aminoglycosides, 33–34 Aminopenicillin, 32 Aminophylline, 359 Amoxicillin–clavulanate, 32 Amphotericin B, 32, 36 Ampicillin–sulbactam, 33 Ampulla, 143, 145 of Vater, 90, 148 Amputation(s) complications, 423–427 determination, 424 fingertip, 437, 442 considerations general, 423 intraoperative, 424–425 pain syndromes after, 426 surgical technique rules, 425 traumatic, 441–442 Amrinone, 335 Amylase-to-creatinine clearance ratio, 641 Anaerobiosis, 23 Anal surgery, complications of, 104 Analgesia, 466 multimodal, 472

Analgesics effects, 469, 472 oral, 467 parenteral, 468 Anaphylactoid, 43 Anaphylaxis, 29 Anastomosis, 628 atrial, 360 bicaval, 357 Billroth I and II, 115, 118 leaks in, 99, 262 Roux-en-Y, 115 strictures of, 103 Anemia, hemolytic, 32 Anesthesia awareness, 9 complications, 436–437 epidural, 5 Anesthesia-related nerve injuries, 507 Aneurysm abdominal, 316 anastomotic, 309, 316 aortic, 318 juxta-anastomotic, 316 ruptures, 317, 319 Angiogenesis, 522 Angiographic embolization, 146 Angiography, 136 Animal bite injuries, 439 Anorectal abscesses and fistulas, complications of, 104 Antacids, 642 Antagonism, definition of, 30 Anterograde amnesia, 513 Antibiotics, 29, 62 Antibiotic synergy, 30 broad-spectrum, 91 multiple, 31 Anticoagulants, 526 therapy, prohylactic, 629 Anticoagulation, devices used for, 371 Anticonvulsants, 476 tricyclic, 475–476 Antidiuretic hormone (ADH), 488 Antifolate agents, 34–36 Antifungal agents, 36 Antigen-presenting cells (APCs), 185 Antihypertensive agents, 210 Antimicrobial agent combinations of, reasons, 31 complications of, 29 induction of resistance to, 30 Antimicrobial chemotherapy, 29 Antimicrobial therapy, 30, 541 Antimicrobials, topical, complications of, 533–534 Antithrombin III, 379 Antiviral drugs, 36 Anxiety disorders, 514 Aorta, complications of surgery to, 315

646

Index

Aortic aneurysm repair, open, 315 surgery, 306 Aortic astheroma, 337 Aortic cross clamping, 301–302, 306 Aortic curtain, 342 Aortic dissection, 328 Aortic injury, diagnosis of, 402 Aortic valve, 342 rupture, 343 Aorto-cardiac fistulae, 348 Aortoduodenal fistula, 318 Aortofemoral bypass, 315 Aortotomy, 342 Apathy, symptoms of, 514 Apligraft1, 524 Appendectomy, 173 complications of, 98 Appendicitis, 98 complications of, postoperative, 98–99 diagnosis of, during pregnancy, 640 Aprotinin, 50 a serine protease inhibitor, 371 ARDS. See Adult respiratory distress syndrome Arglaes. See Silver-impregnated dressing Argon beam coagulator, 234 Arrhythmia, 264, 359 intraoperative, 6 treatment of, 240 Arterial base deficit, definition, 56 Arterial blood gas analysis, 383 Arterial esophageal fistula, 72 Arterial infarcts, 487 Arterial insufficiency, acute, 320–321 Arterial vasculopathy, 188 Arteriogram, 307 Arteriography, diagnostic, 314 Arteriovenous fistula (AVF), 347–348, 626, 629 lack of maturation of, 321 Arteriovenous graft (AVG), 321 Artery injury brachial, 419 popliteal, 419, 420 Artificial urinary sphincter (AUS), complications of, 594 Ascites, 181 Asherman syndrome, 560 Aspartate aminotransferase (AST), 33 Aspergillus, 293 Aspiration pneumonia, 70–71 Aspiration pneumonitis, 3 Atelectasis, 5 Atheroembolism, 306, 308 Atraumatic clamps, 307 Atrial cannulation, 329–330 Atrial fibrillation and flutter, 333 Atrial free wall, right, 368 Atrial natriuretic protein (ANP), 488 Atrial quick connectors, 370 Autograft, 523 Avascular necrosis (AVN), 407 clinical presentation, 414 diagnosis, 414 dislocation, after, 420 incidence, 414 treatment, 415 Avitene1, 100 Axial compression, 492 Axial flow impeller pumps, 375 Axillary lymphadenectomy, 435–436

Axillary nerve, 420 Axonal injury, diffuse, 512 Azine dye, 125 Azithromycin, 34 Azoles, 36 Aztreonam. See Monobactams Bacterial infections, 361 Balloon dilation, endoscopic, 112 Balloon embolectomy, complications of, 320–321 Bandaging, complications, 433 Bankart lesion, 421 Barotrauma, 275 B-cell lymphoma, 186 Bennett’s fracture, reversed, 451 Beta-adrenergic blockade with drugs, 210 Beta-adrenergic stimulation, 335 Beta-lactam ring, 30 Beta-lactamase production, 30 Biaxin tablets. See Clarithromycin Bicitra1, 3 Bilateral lung transplantation (BLT), 288–289 Bile duct, 148 injuries, 171 steps in preventing, 88 Bile peritonitis, 91 Bile-stained fluid, 91 Biliary complications, diagnosis of, 88–89 Biliary fistula, 138 Biliary leaks, 88 detector of. See Hepatobiliary scintigraphy Biliary microorganisms, 91 Biliary operations, 87 complication management in, 90 Biliary stasis, 91 Biliary stricture, 90–91, 140, 184, 188 Biliary surgery, elective, 87 Biliary tract disease, 640 lacerations, 91 Biliary trauma diagnosis of, 91–92 introduction, 91 treatment to, 92 Biliary tree, 89 Bilious emesis, 121 Billroth Iand II anastomosis, 112–113, 115, 118 Biloma, 138 drainage of, 89 Biobrane1, 534 Biofeedback therapy, 105 Biomedicus pump, 347 Biotrauma, 275 Bispectral index1 (BIS1), 9 Bite injuries, types, 439–440 Bladder augmentation, 579–581 Bladder cancer, 573 Bladder catheter, 78 Bladder-drained pancreas transplantation, 630 Bladder injuries, epidemiology and diagnosis of, 612–615 Bladder pressure, 78 Bladder rupture, 612 diagnosis, 613 Bladder stones, 581 Bladder trauma, 585 Bladder tumor, 573 Blalock’s description of shock, 55

Bleeding, 137–138 diathesis, 376 early, 371 late, 371 postoperative, 331 substantial, 233 thumbtack for presacral, 567 Blind loop syndrome, 97, 121 Blood donation, autologous, advantages and disadvantages of, 49 intraoperative recovery of, 50 oxygen-carrying capacity of, 41 autologous, 49 bacterial contamination of, 46–47 circulatory overload during, 48 complications of, 42, 45, 46, 49 diagnosis of, 48–49 hemolytic reactions of, 44 immunologic complications of, 42 nonhemolytic reactions of, 42 pathophysiology of, 42–43 risks of, 46 role of laboratory tests in, 48–49 treatment axioms of, 51 products administration of, 370 indications and contraindications for, 42 transfusion of, 60 transfusion, 41, 371 Blue dye absorption, 71 Blunt aortic injury (BAI), 344–345 Blunt finger dissection, 257 Blunt trauma, 347 Body fluid compartments, membrane characteristics of, 18–19 Bone complications, 292 disease, 187 formation, 421 and joint assessment, 430–432 Both left and right ventricular assist devices (BiVAD), 371 Bowel injuries, 82, 558 inadvertent, 96 Bowel obstruction, 638 Bowel resection, 96 Bowstringing, 450 Brachial artery injury, 419 Brachial plexus injuries, 11, 338, 504–505 Brachial plexus neuropathy, 475 Bradyarrhythmias, 332 Brain abscess, 487 edema, 485 injury, 403–404 acute phase of, 513 management of, 401 neuropathology of, 512 pathophysiology of, 512 trauma, 399–402 Breast biopsy, complications of, 193, 195 Breast cancer, 193 pregnancy associated, 641, 642 Breast conservation therapy, 193, 195 Breast reconstruction complications associated with, 195–196 criticism of, 198 Breast tumor, 194 Brescia–Cimino fistulas, 322 Brescia–Cimino shunt, 322

Index Bridge-to-recovery therapy, 368 Bridge-to-transplantation therapy, 369 Broad-spectrum antibiotics, 91 Bronchial anastomotic complications, 288 Bronchial stump, 237–238 Bronchiolitis obliterans, 292 Bronchiolitis obliterans syndrome (BOS), 272 Bronchogenic carcinoma, 229, 231 Bronchomalacia, 236 Bronchopleural fistula, 237 diagnosis and treatment of, 238 Bronchoscopy, 71, 231, 234 Bronchospasm, 4 Brown-Sequard syndrome, 493 Buck fascia, 593, 621 Budd–Chiari syndrome, 184 Bupivacaine, 6 Burn(s), 390 categories of, 531 chemical, 531 contractures, 543 immersion, 531 types, 440 wounds, dimensions of, 531–532 Burn injuries complications of, 531 infectious, 532 psychological, 535 pulmonary, 534 reconstruction, 535 renal complications of, 535 thermal, 443 Bypass surgery, 447 Calcification, 414 Camino fiberoptic monitors, 74 Cancer bladder, 573 complications of surgery for pancreatic biopsy, 154 proximal resection, 155 pancreatic, distal resections for, 155–157 Candida, 533 Cannulation, 368 Capillary endothelium, 18–19 Capillary leak syndrome, 60 Capsule contracture, 196 Carbapenems, 33 Carbon monoxide diffusion capacity (DLCO), 229–230 Carcinogenesis, 194 Carcinoid tumors of appendix, 224 of colon and rectum, 224 of duodenum, 223 history and epidemiology of, 219 of small intestine, 221 of stomach, 220 Carcinoids, goblet cell, 224 Carcinoma, 561 bronchogenic, 229, 231 gastric, 113, 125 Cardiac allograft vasculopathy (CAV), 362 causes, 363 Cardiac arrest causes of, 299 factors of, 299 management of, 300–301 normothermic, 299 pathophysiology of, 299 traumatic, 299

Cardiac assist devices, 367 Cardiac complications, 311 Cardiac drugs, effects, 359 Cardiac herniation, 241 Cardiac index, 333 Cardiac massage manual, 301 minimally invasive directed, 301 open, technique of, 301 Cardiac morbidity and mortality, 195 Cardiac operation, complications of repeat, 330 Cardiac risk factors, 253–254 Cardiac tamponade, 68–69 Cardiac toxicity, 6 Cardiac transplantation allograft vasculopathy in, 362–363 complication, 362 dysfunction, electrophysiologic, 359–360 gastrointestinal complications, 363 infectious complications bacterial infections, 361 viral infections, 361 intraoperative complications graft failure, 358 hyperacute rejection, 358 operative techniques heterotopic, 357–358 orthotopic, 357 postoperative complications mediastinal hemorrhage, 358 ventricular dysfunction, 358–359 posttransplantation malignancy of, 361–362 Cardiac transplant recipients, 357 acute renal failure in, 360 immunosuppressive regimen agents, 361 antimetabolite, 363 calcineurin inhibitor, 363 corticosteroids, 363 Pneumocystis carinii pneumonia in, 361 stroke risk for, 360 Toxoplasma gondii infection in, 361 Cardiac trauma, 346 Cardioplegia, 328–334 blood, 332 Cardiopulmonary bypass (CPB), 288, 327, 339, 347, 358, 360 central nervous system injury during, 336–337 Cardiopulmonary resuscitation (CPR), 299–300 cardiac pump model of, 300 complications of, 302 management, 303 thoracic pump model of, 300 Cardiopulmonary tests, 229 Cardiovascular instability, 403 Cardiovascular pressures, normal, 56 Cardiovascular system, complications of, 5–6 Cardiowest total artificial heart (TAH), 370, 372–373 Cardizem1, 363 Carlen’s tube. See Double-lumen endotracheal tubes Carotid artery puncture, 68 Carotid endarterectomy (CEA), 504 complications, 313 Carotid–subclavian bypass, 345 Carpal tunnel syndrome, 473

647

Catastrophic intra-abdominal complications, 362 Catecholamines produces, 216 Catheter colonization, 69 drainage, 615 dwelling time, 61 lumens, 61 Catheter-related bloodstream infection (CRBSI), 69 organisms involved in, 69 pathogenesis of, 69 Catheter-related infection, 61 Cauda equina syndrome, 493 CBD. See Common bile duct Cefazolin. See Cephalosporins therapy, 33 Cefuroxime. See Cephalosporins Cell death, 492 Cell-saver blood, 308 Cell-saver transfusion, 335 Cellular swelling, 22 Cellulitis, 341 See Lymphangitis Cell-washing device, 50 CelsiorTM, 289 Central nervous injury during cardiopulmonary bypass, 336–337 during hypothermic circulatory arrest, 337 Central venous lines, 67 Central venous pressure (CVP), 17, 335 Cephalosporins, 30, 33 Cerebellar infarction, 487 Cerebral edema, 180 Cerebral perfusion pressure (CPP), 400 Cerebral salt wasting (CSW), 488 Cervical dysplasia, 561 Cervical hematoma, postoperative, 203 Cervical nerves, 504 Cervical spine, 495 Cervix, conization of, 561 Chemical burns, 531 Chemical neurolysis, 479 Chemotaxis, 32 Chemotherapy antimicrobial, 29 neoadjuvant, 232 Chest injury, 404 Chest radiograph (CXR), 292, 383, 402 Chest tube insertion, 233 placement, 73 Chest wall implantation, 245 reconstruction, 243 resection, 229 trauma, 281 Chloramphenicol, 35 Cholangiography, 171 advantage of, 88, 151 Cholecystectomy, 148, 171, 222 Cholecystitis, 340, 640 Cholelithiasis, 87, 118 Chronic adrenal insufficiency, 209 Chronic atrophic gastritis type A (A-CAG), 220 Chronic cholestatic syndromes, 187 Chronic gastric atony, 122 Chronic ischemia, 309 Chronic nonhealing wounds, 523

648

Index

Chronic obstructive pulmonary disease (COPD), 234, 240, 253–254, 272 Chronic pain, 426, 473 syndrome, 82 Chronic pancreatitis, complications of surgery for, 148–149 Chronic rejection, 187, 292 Chronic subdural hematoma, 484 Chylothorax, 239–240 diagnosis and treatment of, 240 Chylus drainage, 239 Ciprofloxacin, 34 Cirrhosis, 80, 88, 137, 180 Citrate toxicity, 47 Clagett procedure, 239 Clamp-andsew technique, 345 Clarithromycin, 34 Clindamycin therapy, 35 Clostridia, 533 Clostridium difficile titer, 35, 77, 119, 291, 317 Clotting factors, 41 CMV. See Cytomegalovirus Coagulation factors, 180 Coagulopathy, 180, 307 Cobble-stoning, 348 Cognitive-behavioral therapy (CBT), 516 Cognitive dysfunction, postoperative, 8 Cognitive rehabilitation, 513 Colchicine, 526 Collis gastroplasty, 127 Colorectal injury, 105 Colorectal surgery, complications of, 99 Colorectal trauma, complications of, 105 Colostomy, 101 Colporrhaphy, 562–564 Combitube1, 2 Common bile duct (CBD), 87 injuries to, 90 Common bile duct explorations (CBDEs), 87 Common duct stones, retained, 91 Common hepatic duct (CHD), 88 Compartment syndrome, 10, 438–439, 456, 503 causes of, 410 clinical evaluation, 411 complications, 443 pathogenesis, 410–411 tissue-pressure measurements in, 411 Complex regional pain syndrome, 433, 437–438 Component therapy, role of, 42 Computed axial tomography (CT), 89 Congestive heart failure (CHF), 253, 306 Congo red. See Azine dye Conus medullaris syndrome, 493 Coomb’s test, 33, 45, 48 COPD. See Chronic obstructive pulmonary disease Cord syndrome, 493 Core needle biopsy (CNB), 193 Coronary artery bypass grafting (CABG), 504 Coronary artery disease (CAD), 5, 305 Coronary artery fistulae, 348 Coronary artery spasm, 330–331 Coronary atherosclerosis, 195 Coronary bypass procedures, complications after, 327 Coronary ischemia, 308, 311–312 Coronary sinus, rupture of, 328 Coronary syndromes, definition, 6 Coronary thrombosis, 347

Corticosteroids, 526 side effects, 363 Cortisol-producing adenomas, 207 Cosmesis, 194 Cosmetic deformity, 437 Coumadin1, 70, 413, 526 CPR. See Cardiopulmonary resuscitation Cranial nerves, 504 Craniotomy, 400 Creatinine phosphokinase (CPK), 331 Cricoid pressure, 3 Cricoidthyroid membrane, 3 Cricothyroidotomy, 2 Crohn’s disease, 96 Crohn’s ileitis, 97 Cryosurgical ablation (CSA), 225 Crystalloid solutions, 57–58 Cuff erosion, symptoms of, 594 Cushing’s disease, 207, 209 Cutaneous and musculoskeletal complications, 496 Cutaneous flaps, 547 Cutaneous nerve injury, 435 Cut-down technique, 367 Cyclooxygenase-2 (COX-2) inhibitors, 468 Cyclosporine, 361 applications, 186 effects, 363 Cyproheptadine, 223 Cystduodenostomy, 151 Cystectomy, partial, 579 Cystgastrostomy, 150–151 Cystic duct leaks, 89 diagnosis and treatment of, 90 Cystic duct, ligature of, 90 Cystjejunostomy, 150–151 Cystolitholapaxy, transurethral, 582 Cystoprostatectomy, radical, general steps in, 575 Cytogam1, 182 Cytomegalovirus (CMV), 182, 291–293, 361–363 diagnosis of, 293 Cytovene. See Ganciclovir Dacron1, 369, 375 Dakin’s solution, 533 D-amino-D-arginine vasopressin (DDAVP), 331 D-dimer assays, 380, 383 Deep inferior epigastric artery perforator (DIEP), 198 Deep vein thrombosis (DVT), 313, 379, 413 complication of, 382 diagnosis of, 379 treatment, 382 Deformity, cosmetic, 437 Dehydration, chronic, 101 Delayed gastric emptying (DGE), 156, 157 Delayed-type hypersensitivity (DTH), 31 Delivery, anorectal complications of, 642 Denitrogenating, 2 Depression, 514 risk factors for, 513 Device infection, prevention of, 374 Dexon1. See Polyglactic acid Diabetes mellitus, 525 Diagnostic peritoneal lavage (DPL), 400, 408 Diarrhea, postvagotomy, 119 Diastolic dysfunction, 334 Diffuse axonal injury, 512 Dilation and curettage (D&C), 559–560 Dipyridamole thallium imaging, 306

Dislocation after, avascular necrosis, 420 elbow, 421 glenohumeral, 421 heterotopic bone formation, 421 hip and ankle, 421 instability, 421 joint stiffness, 422 knee, 419, 420 musculotendinous injury, 421 neural injuries associated with, 420 posttraumatic arthritis, 421 traumatic, 421–422 Disseminated intravascular coagulation (DIC), 45, 47 DLCO. See Carbon monoxide diffusion capacity Dobutamine, 62 Dopamine (DA ), 7, 335, 495 Dopexamine, 7 Doppler ultrasonography, 183, 188, 412 Double-lumen endotracheal tubes, 237 Driveline infections, 373 Drug fever, 30 Duct injury, thoracic, 264 Dumping syndrome, 117, 265 Duodenal stump blowout, 116 Duodenal ulcer, 112 Duplex ultrasonography (US), diagnostic principle of, 379–380 DVT. See Deep venous thrombosis D-xylose breath test, 98 Dysesthesia, 442, 450 Dysfunction electrophysiologic, 359 sinoatrial node, 359 ventricular, 359 Dysphagia, 128 Dysrhythmias, postoperative, 332 Eastern Association for the Surgery of Trauma (EAST) guidelines, 404 Echocardiography, 384 ECMO. See Extracorporeal membrane oxygenation Ectopic bone formation, 421 Ectopic calcification, 414 Egyptian mummies, 581 Ehlers–Danlos syndrome, 526 Ejection fraction (EF), 333, 374 Elbow capsular release, 422 dislocations, 421 injuries, 502 Electrical injuries, 443 complications, 535–536 Electrocardiography (ECG), 383 Eloesser drainage, 239 Emboli, source of, 336 Embolism, 241 Embolization, angiographic in, 146 Emergency medical technicians (EMTs), 392 Emergency Medical Technician-Ambulance (EMT-A), 392 Emphysema, 244 Emphysematous crepitus, 256 Empyema, 72, 73, 239 subdural in, 487 En bloc splenectomy, 113 End-diastolic volume (EDV), 56 End–expiratory pressure, 277 Endocarditis, 341

Index Endoleak, types of, 319 Endoluminal ultrasonography (EUS), 224 Endomyocardial biopsy, 360 Endoscopic dilatational techniques, 91 Endoscopic retrograde cholangiopancreatography (ERCP), 89, 183, 640 Endoscopic ultrasound (EUS), 217 Endotracheal intubation, 2, 274 Endovascular aneurysm repair (EAR), complications of, 318 Enteric drainage techniques, 631 Enterobacter cloacae, 533 Enterococcus faecalis, 35 Enterococcus faecium, 35 Epididymis, 599 Epidural analgesia, continuous in, 470 Epidural hematoma, diagnosis of, 10 Epidural opioids, infusions of, 470 Epinephrine, 6, 10, 501, 335–336, 436 Epithelial cells, 539 Epithelialization, 522 Epsilon-aminocaproic acid (EACA), 332 Epstein–Barr virus (EBV), 182, 361–362 ERCP. See Endoscopic retrograde cholangiopancreatography Eruptions, cutaneous (dermal), 30 Erythromycin, 34 Escherichia coli. See Biliary microorganisms Esophageal anastomotic leaks, 263 Esophageal exclusion, 258 Esophageal injuries, 261 management of, 262 Esophageal intubation, 2 Esophageal perforation, 71 causes of, 256 cervical, 257 management of, 256–257 principles of, 257 symptoms of, 256 Esophageal resection, complications of, 263, 264 Esophageal surgery, 251 Esophageal trauma, 261 Esophageal wall, layers of, 258, 264 Esophagogastroduodenoscopy (EGD), 111, 318 Esophagoscopy, 257 Esophagraphy, 258 Esophagus anatomy and physiology of, 251–254 injuries to abdominal, 261 cervical, 262 thoracic in, 262 Essex-Lopresti lesion, 457 Expanded polytetrafluoroethylene (ePTFE), 79–80. See also nonabsorbable meshes Extracellular fluid (ECF), 24 Extracellular fluid volume (ECV), 22 Extracorporeal life support (ECLS), 280 Extracorporeal membrane oxygenation (ECMO), 367, 371, 372 application of, 290 Extracorporeal membrane oxygenationassociated blood dyscrasia, 370 Falls, 389–390 Fasciotomy, 409 complications of, 321

Fat embolism (FE) syndrome clinical symptoms, 409 pathogenesis, 409–410 treatment, 410 Febrile reactions, 43 Femoral nerve, 506 Femur fractures, fixation of, 404 Fenoldopam, 8 Fentanyl, 468, 470 Fetal demise, 637 Fetus medications effect on, 639 radiation damage to, 638 Fiberoptic bronchoscopy (FOB), 3, 290, 534 Fibrin glue, 235 Fibrinogen, radioactively labeled, 380 Fibrogenesis, 522 Fibrinolysis, 332 Fibroblast growth factor (FGF), 522 Fibular free flap, 550 Fill-to-empty mode, 370 Fine-needle aspiration (FNA), 154 Fine-needle aspiration cytology (FNAC), 193 Fingertip amputations, 437, 442 injuries, 442 Finney pyloroplasty, 116 First-line screening test, 380 Fistula, 147 colocutaneous in, 100 formation of, 582 Rectovaginal in, 642 Flaps cutaneous, 547 free microvascular, 549 muscle, Mathes–Nahai classification, 547–548 pedicled, 548–549 Flexor tendon, reconstruction of, 450 Fluconazole, 36 Fluid collection, 144–145 distribution theoretical basis, 17 maintenance and regulation, 18 replacement solutions, complications of, 25 resuscitation, 60 delay, 58 sequestration, 60 history, 17 therapy, 17 goals of, 23 theoretical models of, 19–21 Fluorodeoxyglucose positron emission tomography (FDG-PET), 229 Fluoroquinolones, 34 Folate absorption of, 126 deficiency of, 126 Foley catheter, 100, 308, 346 drainage, 631 Forced expiratory volume in one second (FEV1), 230 Forearm fractures, 455–456 Foreign bodies, 442–444 Fournier gangrene, 597 Fraction of inspired oxygen (FiO2), 271 Fracture complications of, 408–413 dislocations of, 452, 453–454 fixation, 403–404

649

[Fracture] forearm, 455–456 hemorrhage, 408 hook of hamate, 451, 454–455 and joint injuries complications, 451, 458 late complications of avascular necrosis, 414–415 heterotopic ossification, 415 nonunion, 413–414 malunion, 413 metacarpal, 452–453 open, 411 pelvic, 408 phalangeal, 452 radial head, 457–458 scaphoid, 451, 453 Fresh frozen plasma (FFP), 42, 47 Frostbite, 443 Fulguration, 88 Full-thickness wounds, 523 Functional residual capacity (FRC), 234 Fundoplication, 127, 172 Fungal infection, 181 Furosemide, 340 Galeazzi fracture-dislocation, 457 Gallbladder disease, diagnosis of, 89 Gallbladder injuries, 92 Gallstone pancreatitis, 640 Galveston orientation amnesia test, 513 Gamekeeper’s thumb, 451, 453 Ganciclovir, 36 Gangrene of lung, 236 Gas bloat syndrome, 128 Gastric carcinoid tumors, 220 Gastric carcinoma, 113 Gastric disease, diagnostic procedures of, 111 Gastric emptying, rapid, 118–119 Gastric intramucosal pH (pHi), 57 Gastric lymphoma, 114 Gastric motor physiology, 114 Gastric perforation, 116 Gastric peristalsis, 115 Gastric remnant carcinoma, 125 Gastric remnant syndrome, small, 123 Gastric resection, 115 Gastric surgery, 111 complications of, 115, 117 nutritional consequences of, 125 Gastric tonometry, 57 Gastric ulcer, 497 complications of, 113 treatment of, operative, 112 Gastrinoma, 216 complications of, 217 surgical management of, 217 Gastrocnemius, 549 Gastrografin1, 71, 257, 261 Gastroesophageal reflux disease (GERD), 127, 642 Gastrografin esophagography, 257 Gastrointestinal (GI) complications, 291, 317, 340, 362, 534 Gastrointestinal continuity, 112 Gastrointestinal injury, 558 Gastrointestinal hemorrhage, 497 Gastrointestinal motility, 114 Gastrojejunostomy, 112–113, 115 Gastroparesis, 116 Gastrostomy tubes, 70 G-cell hyperplasia, 123–124 Gelatin–resorcin–formalin glue, 328

650

Index

Genitofemoral neuralgia, 82 Gentamicin nephrotoxicity, 34 Glasgow coma scale (GCS), 399, 511 Glenohumeral dislocations, 421 Glomerular filtration rate (GFR), 6 Glucagon-like peptide-1 (GLP1), 118 Glucose-dependent insulinotropic peptide (GIP), 118 Gluteus myocutaneous flap, 103 Goblet cell carcinoids, 224 Goldman cardiac index, 272 Gore-Tex1, 243, 328 Gott shunts, 345 Gracilis muscle, 548 Graft dysfunction, 360 complications of, 289 Graft failure primary, 358 skin, 541 Graft infection, 312, 318 Graft limb thrombosis, 320 Graft loss, 541 Graft salvage, 626 Graft syndrome, melting of, 542 Graft thrombosis, 307, 312 Graft versus host disease (GVHD), 188 signs and symptoms of, 44 Graft-wound healing, 539 Gram-negative enteric bacilli, 234 Gray baby syndrome, 36, 638 Grooves, intercostal, 245 Guillotine amputation, 313 Gunshot wounds (GSWs), 390, 610 GVHD. See Graft versus host disease Haddon matrix, 396 Hagedorn insulin, 329 Hamman’s sign, 261 See Xiphisternal crepitus Hand injury, complications, 437–439 mangling injuries, 443–445 pathologic fractures of, 452 procedures, complications of, 435–436 surgery, complications of, 429, 432 Hasson open technique, 88 Head trauma, pathophysiology of, 512 Heart bypass, left, 329 HeartMate1, 369, 371 Heimlich valve, 235 Heineke–Mikulicz pyloroplasty, 112, 116 Helical computed tomography (HCT), 384 Helicobacter pylori, 111 HELLP (hemolysis, elevated liver function, low platelet count) syndrome, 641 Hematologic toxicity, 35 Hematoma, 193, 483–484 formation of, 203 Hematuria, 630 Hematuria–dysuria syndrome (HDS), 578 Hemisection cord syndrome. See BrownSequard syndrome Hemithorax, 232, 234, 236–238, 241 Hemobilia, 139–140 Hemodialysis, complications of surgery for, 321 Hemodynamic changes in, 210 effects, 168 monitoring, 56 parameters, 57

Hemodynamic-cardiac signs, 383 Hemoglobin-based oxygen carriers (HBOC), 58 Hemoglobin, oxygen saturation of, 42 Hemoglobinuria, 45, 49 Hemolytic reactions, 45 Hemorrhage, 68, 112, 316 cerebral, 314 control of, 98 fractures, 408 herald, 318 management, 358 massive, 308 mediastinal, 358 pancreatitis, 146 pelvic, 566 fractures, 408 perioperative, 307 shock, 408 Hemorrhoidectomy, complications of, 104 Hemostasis abnormality, 47 enhancement of, 42 meticulous, 309 Hemostatic techniques, 541 Hemosuccus pancreaticus, 146 Hemothorax, 68, 73 diagnosis of, 68 Hemovac system, 73 Henderson–Hasselbalch equation, 57 Heparin bonded circuits, 335 coated circuits, 371 low-dose, 381 Heparin–protamine complexes, 329 Heparinization, 413, 628 Hepatic artery stenosis (HAS), 181, 183 Hepatic artery thrombosis (HAT), 179, 183 Hepatic blood vessels, ligation of, 155 Hepatic graft, 182 Hepatic rupture, 641 Hepatic surgery complications of, 217 techniques for, 135–136 Hepatic veins, 184 Hepatitis B surface antigen (HbsAg) or viral DNA (HBV-DNA), 189 Hepatitis B virus (HBV), 46, 189 transmission of, 46 Hepatitis C virus (HCV) infection, 46, 189 Hepatobiliary iminodiacetic acid (HIDA), 116 Hepatobiliary scintigraphy, 89 Hepatocellular carcinoma (HCC), 184, 189 Hepatopulmonary syndrome, 180 Hepatorenal syndrome, 180 Herald hemorrhage, 318 Hernia incisional, 78, 318 parastomal, 102 prolapse and parastomal, 101 trocar-site, 170 repair inguinal, 81 ventral, 173 Herniorrhaphy, 170 complications of, 79–80, 83 Herpes simplex virus, 361 Herpetic whitlow, 458 Heterotopic bone formation after dislocation, 421 See Heterotopic classification Heterotopic ossification, 414

Hevea brasilienses, 12 Hiatal hernia, 127 Hibernation, 335 High-frequency oscillatory ventilation (HFOV), 278 Hip and ankle dislocations, 421 Histamine, production of, 220 Hollander test, 124 Holosystolic murmur, 347 Homograft, 524 valve, 342 Hook of hamate fractures, 451, 454–455 Hospital-acquired renal insufficiency (HARI), 256 HSV. See Highly selective vagotomy Huber–Weiss reaction, 45 Human bite injuries, 439 Human immunodeficiency virus (HIV), transmission of, 46 Human leukocyte antigen (HLA), 185, 288, 358 Human papillomavirus infection, 186 Human platelet antigen (HPA), 44 Human T-lymphotropic virus-I (HTLV-I), 46 Hunt-Lawrence pouch procedure, 123 Hydrocele, 82, 598 Hydrocephalus, 485–486 Hydrocodone, 467 Hydrogrip-type clamps, 310 Hydromorphone, 468 Hypaque1, 101 Hyperacute rejection, 185, 290, 358 Hyperaldosteronism, 209 Hyperalgesia, 474 Hyperamylasemia, 340, 630 Hyperbilirubinemia, 89, 137 Hypercoagulability, 379 Hypercortisolism, 208 Hypergastrinemia-associated tumors, 220 Hyperglycemia, 187 Hyperkalemia, 48 Hyperlipidemia, 187 Hypernatremia, 24–25, 488 Hyperparathyroidism, 205 Hyperperfusion syndrome, 315 Hyperplasia, neointimal, 309 Hypersensitivity reactions, 29 Hypertension, 210 postoperative, 315 venous, 322 Hypertonic and hemoglobin substitutes, use of, 281 Hypertonic saline (HTS), 21, 58, 488 Hypertonic saline and dextran (HSD), 395 Hypertriglyceridemia, cause of pancreatitis, 641 Hypertrophic scars, 528 Hypocalcemia, 204–205 Hypogastric artery, bilateral ligation of, 566 Hypokalemia, 209 Hypomagnesemia, 332, 576 Hyponatremia, 488 types of, 24 Hypotensive patients, 400 Hypothalamic-pituitary-adrenal (HPA), 63 Hypothermia, 48 Hypovolemia, 117, 308 Hypovolemic shock, 408 Hysterectomy, radical, 566 Hysteroscopy as a diagnostic and therapeutic method, 564

Index IABP. See Intra-aortic balloon pump Iatrogenic arterial dissection, 327 Iatrogenic hemolysis, 45 Iatrogenic injuries, 105, 610 incidence of, 558 Ibuprofen, 280 Idiopathic thrombocytopenic purpura (ITP), 161 Ileal pouch anal anastomosis (IPAA), 103 Ileostomy, 100 Ileus, postoperative, 95–96 Imipenem, 33 Immune response, effects of antimicrobial agents on, 31 Immunohistochemical analysis, 290 Immunosuppression therapy, tacrolimusbased, 627 chronic, 361 complications, 293, 363 Immunosuppressive agents, 186, 363 Immunosuppressive medications, 291 Impedance plethysmography (IPG), 380 Incisional wounds, 523 Incretin effect, 118 Induction therapy, 362 Infantile hypertrophic pyloric stenosis, 128–129 Infarction, myocardial, 242 Infection-related device malfunctions, 373 Infections approaches to reduce, 373 complications of, 293, 361, 457 driveline, 373 pump pocket, 373 Infectious complications, acute, 340 Inferior mesenteric artery (IMA), 317 Inferior vena cava (IVC), 184, 211 Inferior vena cava filter (IVCF), 382, 385 Inflammatory bowel disease (IBD), 186 Inhalation injury, 534 Injection injury, 440–441 Injuries carpal, 454 degloving, 443 electrical, 443 extravasation, 435 injection, 440–441 intentional, 389 ligament, 451, 454 unintentional, 389 burns, 390 drownings, 389 falls, 389–390 gunshot wounds, 390 motor vehicle crashes (MVCs), 389 mechanism of, 392 intestinal, 565–566 metacarpal joint, 453–454 ring avulsion, 447 Inotropes, 61 Insect bites, 439 Insulinoma, 215 localization of, 216 Intact parathyroid hormone (iPTH), 205 Integra1, 524 Intentional wounds. See Factitious Interatrial groove, 368 Interleukin (IL), 59 Intermittent pneumatic compression (IPC), 413 Internal mammary arteries (IMAs), 330, 504

International normalized ratio (INR), 382 Intestinal injuries, 565–566 Intestinal obstruction, 641 Intestinal operations, complications of specific, 96 Intestinal surgery, complications, 95, 98 Intra-abdominal abscess, 138–139 hypertension, 78 Intra-aortic balloon pump (IABP), 331, 347 Intra-articular fractures, 451–452 Intracranial pressure (ICP) monitoring, 73, 400, 404 hemorrhage associated with, 74 malfunctioning and malpositioning of, 74 Intraoperative ultrasonography (IOUS), 216 Intraperitoneal onlay method (IPOM), 83 Intrathecal pump, 479 Intravascular coagulation, disseminated, 358 Intravenous infusions, 17 Intravenous pyelography (IVP), 558 Ischemia and reperfusion, complications of, 321 sensitivity of right ventricle to, 333 upper extremity, 448 Ischemia–reperfusion injury, 289, 290, 335 Ischemia–reperfusion syndrome, 321 Ischemic colitis, 317 Ischemic optic neuropathy (ION), diagnosis of, 8 Ischemic steal, 321 Jackson-Pratt drain, 341 Jarisch–Herxheimer reaction, 32 Jarvik 2000, 375 Jarvik 7. See CardioWest total artificial heart Jaundice, 146 Jejunal conduit syndrome, 578 Jejunal free flap, 550 Jejunal Roux limb, 91 Joint stiffness after dislocation, 422 Kaposi’s sarcoma, 186 Keloid core excision, 245 Keloids, 244 Ketamine, 5 Ketoconazole, 280 Ketorolac, 468 Kidney injury, presenting signs, symptoms of, 605 Knee dislocations, 419, 420 Kocherization, 148, 153 Lacerations, 523 mural, 346 Lactated Ringer’s (LR) solution, 57 Lactic acidosis, 56 Laparoscopy, 565, 584 adrenalectomy, 212–213 complications related to, 84 inherent risks associated with, 637 in pregnant women, 637 Laparoscopic cholecystectomy (LC), 87 complications, after, 88 trocar placement during, 88 Laparoscopic placement of an adjustable gastric band (LAPBAND), 174 Laparoscopic splenectomy, advantages of, 162 Laparotomy, 400 Laryngeal injuries, 3 Laryngeal mask airway (LMA), 2

651

Laryngeal nerve injury, 203–205 recurrent, 246 Laryngospasm, 3 Lateral femoral cutaneous nerve (LFCN), 506 Latex allergy, 12–13 Latissimus dorsi dorsi muscle, 239 free flap, 550 Left anterior descending artery (LAD), 330 Left ventricular assist device (LVAD), 368, 371 Leukoagglutinins, 43 Leukocytosis, relative, 640 Leukodepletion, 50 Levator fascia, 595 Ligament injuries, 98, 451, 453 Limb amputation, complications, 313 ischemia, 317, 320 pathology, 425 Roux, 151 swelling, 312 tourniquets, 541 Linezolid, 35 Lithotripsy, shockwave, 87 Liver dysfunction, 137–138 failure, 188 parenchyma, 136 Liver transplantation, 223 complications of bilary, 184–185 cardiovascular, 186 endocrine, 180, 186 gastrointestinal, 180 graft-related, 182–183, 187 hematologic, 180 neurolgical, 180, 187 renal, 180, 186 respiratory, 179–180 vascular, 183–184 techniques of, 184 trauma, management of, 136–137 Living-donor lung transplantation, 294 Long-bone fractures, management of, 404 Longitudinal pancreaticojejunostomy (LPJ), 149 Loop syndromes, 121 Lower-extremity bypass, 309–313 Low-molecular-weight heparin (LMWH), 381, 413 Lumbar drain, 485 Lumbar sympathectomy, 427 Lung donor, criteria for, 289 hyperinflation, complications of native, 291 injury acute, 339 transfusion-related acute, 44 Lung transplantation complications of, 290–291 contraindications to, 288 donor selection for, 289 historical perspective of, 287 indications for, 288 living-donor, 294 recipient selection for, 288 risk factors of, 292 surgical techniques of, 289–290 Lung-protective ventilation strategies, 276 clinical studies using, 277

652

Index

Luschka ducts, 88 Lymph node primary gastrinoma, 217 Lymphadenectomy, axillary, 435–436 Lymphangitis, 193 Lymphatic channels, 193 Lymphedema, 193 Lymphocyte transformation, 31 Lympholytic agents, 363 Lympholytic therapy, 360 Lymphoma aggressive forms of, 184 gastric, 114 Lytic therapy, 242 Macrolides, 34 Magnetic resonance cholangiopancreatography (MRCP), 89, 146 Male reproductive organs, injuries to, 621 Malignant hyperthermia (MH), 11 pathogenesis of, 12 Mallet finger, 450 Maloney dilator, 265 Mammography, 197–198 Marlex1, 101 Mass suture, 567 Mastisol1, 434 Maternal–fetal hemodynamics, 637 Matrix metalloproteinases (MMPs), 520 Maxillofacial surgery, 550–551 Mean arterial pressure (MAP), 400 Mechanical airway obstruction, 237 Mechanical ventilation, monitoring during, 273 principles of, 273 Meckel’s diverticulitis, 97 Mediastinitis, 282, 340 Mediastinoscopy, 231 Medications classification of, 639 during pregnancy, 638 Meissner’s plexus, 251 Meloxicam, 468 Melting graft syndrome, 542 Membrane oxygenators, 337 Memory, types of, 513 Meningitis, 486 Meropenem, 33 Mersilene1. See Polyester Mesh repairs, tension-free, 79 Metabolic imbalances, 488 Metacarpophalangeal joint disarticulation, 442 Metastasis of carcinoid tumor to liver, 225 Metastatic liver disease, 216 Metastatic pancreatic endocrine tumors, 217 Methicillin-resistant staphylococcus aureus (MRSA), 30, 532, 533 Metoclopramide, 3 Metronidazole, 35 Microvascular flaps, 549 Minimum bactericidal concentration (MBC), 30 Minimum inhibitory concentration (MIC), 30 Mitral valve, 343, 347 Mitral valve–coronary bypass procedures, 343 Monobactams, 33 MonostrutTM mechanical disc valves, 368 Monteggia fracture-dislocation, 457 Morphine, epidural infusions of, 470 Motor vehicle accidents (MVAs), 515

Motor vehicle crashes (MVCs), 389 fatality rates, 390 Haddon matrix for, 396 Mucin production, 642 Multiglandular disease (MGD), 205 Multiple endocrine neoplasia (MEN), 124, 216 Multiple organ dysfunction syndrome (MODS), 56 Mu-opioid receptors, 95 Muscle flap closure, 341 transposition, 239 Mathes–Nahai classification of, 545–546 Musculotendinous injury after dislocation, 421 Mycophenolate mofetil (MMF), 292 Myocardial band (MB), 330 Myocardial edema, 334–335 Myocardial failure, definition, 333 Myocardial infarction (MI), 305, 315 and death, 242, 310, 347 postoperative, 330 Myocardial ischemia, 335 Myocardial preservation technique, 358 Myofibroblasts, 522 Myoglobinuria, 306, 321 Myopathy, critical illness, 282 Myositis ossificans, 414 n-Acetylcysteine, 280 Narcotics, 244 NASA/DeBakey VAD, 375 Nasogastric suction, 96 Nasogastric tubes (NGTs), 70 malposition of, 72 National Highway Transportation Safety Administration (NHTSA), 396 National Nosocomial Infections Surveillance (NNIS) System, 69 Natural killer (NK) cells, 32 Naturopathic agents, 526 Near-infrared spectroscopy (NIRS), 57 Necrosectomy, 148 Necrosis, 101 Needle or trocar, injuries caused by, 168–170 Needlestick injuries, 435 Nelson’s syndrome, 209 Neoadjuvant chemotherapy, 232 Neointima, fibrin-based, 372 Neoplasia, 525 Nephrectomy, 606 living-related donor, 174 potential complications of, 583 Nephrotoxicity, 33, 230, 360 Nerve blocks, complications of, 10–11 Nerve injuries, 245, 263, 315, 409 anatomy of, 501–502 classification of, 502 complications of, 450–451 cutaneous, 435 diagnosis of, 501 perioperative, 11 peripheral, 338 Nerve repair, complications of, 502 Nerves, sciatic, 554 intercostal, 245 phrenic, 245 Neural blockade, for acute pain management, 471 Neural injuries and dislocation, 420 Neuroaxial analgesic techniques, 470 Neurologic symptoms, 36

Neurologic system, complications of, 8–9, 336 Neurological sequelae, 535 Neurolysis, chemical, 479 Neuroma, 427 Neuromuscular acute respiratory failure, 281 Neuromuscular blocking agents, 5 Neuromyopathy, 282 Neuropathies brachial plexus, 475 femoral, 474 ilioinguinal, 475 pain diagnosis of, 475 identification of, 474–475 treatments for, 475 perioperative of, 473 Neuropeptides, 521 Neuropraxia, 501 Neurotoxicity, 34 Neurovascular injury, 554 Neutrophil counts, 373 NIPPV. See Noninvasive positive-pressure ventilation Nissan fundoplication, 127 Nitric oxide synthetases, 278 Noninvasive positive-pressure ventilation (NIPPV), 274 Nonsteroidal anti-inflammatory drugs (NSAIDs), 468 Nontraumatic cardiac arrest, 299 factors affecting, 300 Norepinephrine, 336 Normothermic cardiac arrest, 299 Normovolemic hemodilution, advantages of, 49–50 Novacor1 device, 369 implantation, 370 Novacor1 left ventricular assist system (LVAS), operating modes of, 369 Nuchal crepitus, 71 Obesity, 208 Octreotide, 147, 216 Ofloxacin, 34 Oliguria, cause of, 308 Omental patch closure, 112 Open cholecystectomy (OC), 87 Opioids, 478 Orchiectom, 598 Organ failure multisystem, 275 Organogenesis, 638 Osteoarthritis, 421 Osteomyelitis, 525 Osteonecrosis, 414 Osteoporosis, 187, 292 Overwhelming postsplenectomy infection (OPSI), 161–163 Oxycontin, 468 Oxygen affinity, changes in, 47 Oxygen-carrying capacity, 41 enhancement of, 42 Oxygen saturation of hemoglobin, 42 Pain management, 472 Pain therapies, advanced, 476 Pancreas transplantation complications associated with entericdrained, 631 early acute rejection, 631 postsurgical complications, 626

Index Pancreatectomy total, 156–157 Pancreatic abscesses, 145 Pancreatic anastomosis, leaks from, 155 Pancreatic cancer, distal resections for, 155 Pancreatic debridement, 148 Pancreatic endocrine tumors, metastatic, 217 Pancreatic fistula, 147 Pancreatic injuries, 153 delayed presentation of, 153 Pancreatic necrosis, 145–146 Pancreatic pseudocyst, complications of surgery for, 149 Pancreatic resection complication of, 156 Pancreatic surgery, complications associated with, 144 Pancreatic trauma, complications associated with , 152 Pancreaticoduodenectomy, 147 Pancreatitis, 91, 116, 497, 629, 641 acute complications of, 148 Pancytopenia, 188 Panel-reactive antibodies (PRA), 288, 358 Paradoxical wall motion, 243 Paralysis, 10, 345 Paraplegia, 317 Parathyroid gland, 204 Parathyroidectomy, 204 failure of, 205 Partial pressure of arterial oxygen (PaO2), 271 Partial pressure of carbon dioxide (PaCO2), 255 Partial thromboplastin time (PTT), 49 Partial-thickness wounds, 523 Patch disruption, 315 Patient-controlled analgesia (PCA), intravenous, 469 Patient-controlled epidural analgesia (PCEA), 470 advantages of, 470 Patient-related risk factors, 255 Pectoralis minor muscle, 419 Pediatric trauma, management of, 396 PEEP. See Positive end–expiratory pressure Pelvic diaphragm, 595 Pelvic fractures, 408 Pelvic hematoma, 614, 621 Pelvic hemorrhage, 566 Pelvic sepsis, 103 Penectomy, 600 Penicillins, 32 antipseudomonal, 33 with beta-lactamase inhibitor, 32 Penile amputation, 619 Penile anatomy, 599 Penile fracture, 621 Penile trauma, 599 Penis, 621 Peptic ulcer, 340 Peptic ulcer disease, 111 Percutaneous transhepatic cholangiography (PTC), 89–90 Percutaneous transluminal coronary angioplasty (PTCA), 330 PerfadexTM, 289 Pericardial effusions, 348 Pericardial friction rub, 348

Pericardial tamponade. See Tension pneumothorax Pericarditis, constrictive, 348 Perineal abscess, 103 Perineurium, 501 Peritoneotomy or pleurotomy, 212 Peroneal nerve, 506 injury, 476 Phalangeal fracture, complications of, 452 Phalangeal neck fractures, 451 Pheochromocytoma, 208, 210 Phlegmasia cerulea dolens, 382 Phrenic nerves, 245, 504 injury, 338 Piggyback technique, 184 Piperacillin–tazobactam, 33 Plain-film radiography, 68 Plasma histaminase levels, in pregnancy, 642 Plasminogen system, 521 Plateau pressure, 274 limiting, 275 Platelet aggregation, 627 Platelet-derived growth factor (PDGF)-BB, 522 Platelet dysfunction, 33 Platelet-related sepsis, 47 Platelet-specific antigen (PLA), 44 Pleurodesis, 73 Pneumatic compressive devices (PCD), 381 Pneumocephalus, 484 Pneumocystis carinii pneumonia, 361 Pneumonia, 233 aspiration, 70–71 infectious agents of, 342 nosocomial, 233–234 Pneumocystis carinii, 361 Pneumoperitoneum effects of, 637, 168 establishment of, 167 Pneumothorax, 67, 73, 232, 261 diagnosis of, 68 risk of, 68 Polycarbonate housing, 367 Polyenes, 36 Polyester, 79. See also Nonabsorbable meshes Polyglactic acid, 79. See also Absorbable meshes Polymerase chain reaction (PCR), 182, 186, 189 Polymorphonuclear leukocyte function, 32 Polymorphonuclear leukocytes (PMNs), 533 Polyneuropathy critical illness definition of, 282 Polypropylene, 79. See also Nonabsorbable meshes Polypropylene suture material, 311 Polyurethane bladder, 367 Polyurethane blood sac, 368 Popliteal artery injury, 419–420 Portal hypertension, complications of, 183 Portal vein, 183 Positive end–expiratory pressure (PEEP), 274, 331 optimal level of, 277 determination of, 278 Post phlebitic syndrome (PPS), 382 Post traumatic amnesia (PTA), 511, 512 Post–dural puncture headache (PDPH), pathophysiology of, 9

653

Postgastrectomy malabsorption, 125 Postgastrectomy syndromes, 114, 117 Postoperative pain management, anatomy and pathophysiology, 465 Postpericardiotomy syndrome, 236, 348 prevention, 236 Post-transplant lymphoproliferative disease (PTLD), 182, 186, 291, 361–363 Posttransplantation malignancy, 361 Posttraumatic arthritis, after dislocation, 421 Posttraumatic stress disorder (PTSD), 511, 535 diagnostic features of, 515–516 syndrome of, 515 risk factors for, 516 therapeutic approaches to, 516 Postvagotomy diarrhea, 119 Pouchitis, 104 Preeclampsia, 641 Pregnancy associated breast cancers, 642 nonobstetric operations, during pregnancy, 637 physiologic changes in, 639 Prehospital care education of personnel, 391, 394 Preoperative autologous donation, 49 Preterm labor, 637 Primary biliary cirrhosis (PBC), 187 symtoms, 189 Primary graft failure, 358 Primary sclerosing cholangitis (PSC), 189 Proctocolectomy, restorative, 103 Proctoscopy, 35 Prophylactic antibiotics, applications, 425 Prophylactic medications, 382 Prophylaxis, 379, 381 Prosopagnosia, 513 Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED), 383 Prosthesis socket, 426 Prosthetic materials, classification of, 79 Protamine reactions, 329 Protein C, anticoagulant role of, 62 Prothrombin time (PT), 49, 332 Proximal thrombi, 412 Pruritus, treatments for, 543 Pseudo–Zollinger–Ellison syndrome, 124 Pseudoarthrosis, 414 Pseudocyst, 144, 149 Pseudomembranous colitis (PMC), 29, 35 Pseudomonas aeruginosa, 533 Pseudoneointima, 369 Psoralen, 47, 50 Psychosis, symptoms of, 514 PTC. See Percutaneous transhepatic cholangiography PTLD. See Posttransplant lymphoproliferative disease PTSD. See Posttraumatic stress disorder Pubovaginal sling (PVS), 595 Pulmonary angiography, 384 Pulmonary artery catheters, 56, 276, 306, 308 Pulmonary artery occlusion pressure (PAOP), 56 Pulmonary complications, 254, 404, 497 Pulmonary contusion, 276, 281 Pulmonary dysfunction, 339 Pulmonary embolectomy, 386

654

Index

Pulmonary embolism (PE), 241, 306, 380, 382, 413 diagnosis of, 383 prophylaxis against, 385 symptoms of, 383 treatment for, 385 Pulmonary gangrene, 236 Pulmonary hypertension, 240, 371 diagnosis of, 241 Pulmonary injuries, 404 management of, 404 Pulmonary resection, 229 air leaks after, 234 airspace problems in, 238–239 Pulmonary risk factors, 254 Pulmonary torsion, 235 Pulmonary vascular resistance, 357, 359 Pulmonary venous anastomosis, 357 Pulse oximetry, 306 Pulseless electrical activity (PEA) causes of, 300 Pulseless electrical activity (PEA). See Ventricular asystole Pulseless ventricular tachycardia (VT). See Ventricular fibrillation (VF) Pump pocket infections, 373 Purpura, posttransfusion, 44 Pyloromyotomy, 265 Pyloroplasty, 115 Pylorospasm, 265 Quadrigia syndrome, 450 Quinpristin–dalfopristin, 35 Radiation therapy, 196 Radical hysterectomy, 566 Radical pelvic surgery, for gynecologic malignancy, 566 Radiofrequency ablation (RFA), 225 Radionuclide scanning, 91 Rattlesnake bites, 440 Recombinant human activated protein C (rHAPC) anticoagulant effect of, 62 Reconstructive surgery, complications of, 547 Rectus abdominis free flap, 550 Recurrent nerve injury, prevention of, 339 Red blood cell (RBC), 41 sepsis, 47 transfusions, 58, 317 Red cell substitute, use of, 50 Red man syndrome, 35 Re-endothelialization, 343 Re-expansion pulmonary edema (RPE), 73 Regional anesthesia, complications of, 9 Regurgitation, tricuspid, 347 Reimplantation response. See Ischemia reperfusion injury Renal disease, 306 Renal dysfunction, postoperative, 7 Renal failure, 339 acute, 308, 317, 360 nonoliguric, 308 Renal injuries, epidemiology and diagnosis of, 605–606 Renal system, complications of, 6–8 Renal toxicity of general anesthetics, 6 Renal trauma classification, 606 complications, 608 nonoperative management of, 606 Reperfusion injury, 338, 627

Resection bowel, 96 inadequate, 96 Respiratory complications, 339 Respiratory depression, 470 Respiratory failure, 308 causes of, 272 classifications of, 271 definition of, 271 determination of risk factors of, 272 epidemiology of, 271–272 hypercarbic, 271 hypoxemic, 271 management strategies of, 272 neuromuscular acute, 281 Respiratory pump, 271 Respiratory system, complications of, 4–5 Restenosis, 315 Resuscitation with colloid solutions, 20 with crystalloid solutions, 19–20 end points of, 23 with hypertonic saline solution, 21 massive volume, 22 Resuscitative fluid categories of, 57 technique of, 301 therapy, 17, 23 Retinaculum of Weitbrecht, 420 Retrograde amnesia, 513 Retrograde urethrography (RUG), 592 Rib fractures, 281 Right ventricular (RV) cannulation, 368 Right ventricular assist devices (RVADs), 374 Ring avulsion injuries, 447 Ringer’s lactate solution, 25 R-on-T phenomenon, 333 Roto-Rest1, 496 Roux limb, 151 Roux stasis syndrome, 121, 122 Roux-en-Y anastomosis, 115, 118 Roux-en-Y esophagojejunostomy, 113 Roux-en-Y gastrojejunostomy, 121 Roux-en-Y hepaticojejunostomy, 90, 181, 184 Sacrospinous colpopexy, 561–564, 514 Sacrouterine ligaments, 595 Salvage surgery, complications of, 195 Saphenous nerve, 506 Saphenous vein bypass technique, 309 Saphenous vein grafts, 310 Scalp deformity concrete, treatment of, 544 Scalp donor-site alopecia, 544 Scaphoid fractures, 451, 454 Scars, hypertrophic, 244 Schwann cells, 501 Sciatic nerve, 476 Sclerotherapy complications of, 320 Scrotal injury, 597 Scrotum, 597 Scrotum and testicle, 621 Seizures, 180 Seldinger method, 67, 72 Selective serotonin reuptake inhibitors (SSRIs), 514 Sepsis, 424 coagulation in, 62 Septic source control, 60 Serratus anterior muscle, 239 Serum fluoride, 7 Seton procedure, 104 Sevoflurane, 7

Sexual dysfunction, 103 Sheathed blade, 129 Shock, 55 Blalock’s description of, 55 cardiogenic, 55 distributive, 55 hypovolemic classification of volume loss in, 55–56 diagnosis of, 55 hypovolemic, 55 obstructive, 55 specific complications of, 62 treatment for, 57 septic crucial components of, 60 definition of, 59 pathophysiology of, 59–60 theory of, 17 treatment of, 60 Shockwave lithotripsy (SWL), 87, 582 Short-bowel syndrome, 97 Shunting intrapulmonary, 276 Siggard–Anderson nomogram, 56 Sigmoidoscopy, 317 Sigstad scoring system, 118 Silvadene1, 533 Silver-impregnated dressing, 373 Silver sulfadiazine (SSD), 533 Single lung transplantations (SLTs), 287 Sinoatrial node dysfunction, 359 Sinoscopy, 71 Sintered titanium microspheres, 372 Sinusitis, 71–72 Sirolimus (rapamycin), 186 Ski pole thumb, 453 Skin cancers, 186 Skin graft failure, 541 Skin grafts, 523 complications of, 540–541 freehand harvesting of, 540 phases of, 539 techniques of, 540 types of, 539 Small-bowel drainage procedure, 631 Small-bowel injury, 98 Small-bowel obstruction (SBO), 96 Small stomach syndrome, 126 Sniff test, 246 Somatostatin analogs, use of, 216 Somatostatin receptors (SSTRs), 219 Somatostatin receptor scintigraphy (SRS), 217 Sperm granuloma, 599 Spermatic cord, 81, 83 Spinal cord compression, delayed, 493 Spinal cord injury, 338, 497 in children, 497 pharmacologic therapy for, 496 Spinal cord injury without radiographic abnormality (SCIWORA), 497 Spinal cord ischemia, 338 Spinal cord trauma biomechanics of injury, 491–492 epidemiology of, 491–492 pathophysiology of, 492 Spinal hematoma, 10 Spinal opioid analgesia, 469 Splanchnic vein thrombosis, 162 Splenectomy, 161–163, 175, 625 Splenic injury, complications in treatment of, 162 elective, 161

Index Splenic rupture, 162 Splenic surgery, categories of, 161 Splenic tears, 115 Splenorrhaphy, 163 Sponge deformity, 542 Sporadic primary hyperparathyroidism (SPHPT), 205 Staphylococcus aureus, 30, 322, 341 Steatorrhea, 118 Stenosis, 101, 184 Stenosis and stricture, 102 Sternal infections, 340 Sternotomy, 340 Steven–Johnson syndrome, 29–30, 34 Stiff ventricle, 334–335 Stomas, complications of, 100 Stone removal, open, 582 Streptococcal cellulites, 532 Streptococci, 532, 542 Streptomyces clavuligerus, 32 Streptomyces erythraeus, 34 Streptomyces nodosus, 36 Strictures, 184, 265 anastomotic, 103 Stroke, 360 causes of, 314 perioperative, 8 Stunning, perioperative, 335 Subxiphoid pericardiocentesis, 69 Sucker bypass, 328 Sulfamylon1, 533 Sulfonamides, 34 Sunderland classification, nerve injuries, 502 Superior gluteal artery perforator (S-GAP) flap, 198 Superior mesenteric artery (SMA), 155 Superior mesenteric vein (SMV), 146, 155 Superior vena cava (SVC), 68 Surfactant replacement therapy, 280 Surgical residents, training of, 394 Surgical tissue injury, 526 Suture length (SL), 78 Suture ligation, 608 Swan–Ganz catheterization, 23 Symbion TAH. See CardioWest total artificial heart Syndrome of inappropriate antidiuretic hormone (SIADH) secretion, 486, 488 Synechia, 560 Synercid1, 181 Synergy and antagonism, categories of, 30 Synostosis, 456–457 Syringe distension technique, 310 Systemic and metabolic complications, 11 Systemic inflammatory response syndrome (SIRS), 57, 59 Systemic vascular resistance (SVR), 56 Systolic blood pressure (SBP), 399

Tachyarrhythmias, 308, 332 Tachycardia, 168 Tacrolimus, 626 applications, 186 effects, 187, 363 TAH. See Total artificial heart Talc slurry. See Sclerosing agent T-cell panel-reactive antibodies, 371 TEE. See Transesophageal echocardiography Temafloxacin syndrome, 34

Temporomandibular joint injuries, 4 Tendon injuries, complications of, 449 Tendon rupture, 455 Tension-free vaginal tape (TVT), 595 technique, 564 Tension pneumothorax, 299 Testicle, 598 Tetany, neonatal, 643 Tetracyclines, 34 Theophylline, 340 Thermal burn injuries, 443 Thoracentesis, 232 Thoracic aorta, descending, 338 Thoracic aorta injuries, 402 Thoracic compartment syndrome, 282 Thoracoabdominal aortic replacement, 338 Thoracoabdominal approach, 211 Thoracoplasty, 238–239 Thoracosternotomy incision, transverse, 289 Thoracostomy tubes, complications of, 72 Thoracotomy, 302 anterolateral, 346 posterolateral, 258 Thoratec BiVAD, 375 Thoratec device, 369 Thrombectomy, 311 Thrombin–thrombomodulin complex, 62 Thrombocytopenia, 44 Thromboelastogram, 331 Thromboembolic disease, treatment for, 412–413 Thromboembolism, 370–372, 496–497 complication, 413 diagnosis, 413 risk factors, 413 treatment, 413 venous, 313 Thrombolytic therapy, 241, 385 Thromboprophylaxis, strategies for, 382 Thrombosis, 69–70, 183, 317, 320 arterial, 628 complete venous, 628 early and late, 322 early graft, 307, 309 graft limb, 320 graft preventing, 629 partial venous, 626 therapeutic protocols, 629 Thrombus formation, 241 Thymus-derived lymphocytes (T-lymphocytes), 32 Thyroidectomy, 204 Ticarcillin–clavulanate, 32 Tissue atrophy, 594 Tissue culture assay, 35 Tissue flaps, 262 Tissue injury, surgical, 526 Tissue-pressure measurements, 411 Titanium-pedestal skull, 373 T-lymphocyte panel, 358 Tocolysis, prophylactic, 638 Total artificial heart (TAH), 367, 370–371, 375 Total extraperitoneal repair (TEP), 83–84 complications, 84 Total parenteral hyperalimentation (TPN), 154 Total parenteral nutrition (TPN), 147 Tourniquet palsy, 435 Toxic shock syndrome, 435 Toxoplasma gondii infection, 361 Tracheal injuries, 3

655

Tracheostomy, mini, 342 Transabdominal preperitoneal repair and total extraperitoneal repair, comparison of, 85 Transabdominal preperitoneal repair (TAPP), 83–84 complications related to, 84 Transabdominal properitoneal approach (TAPP), 172 Transesophageal echocardiographic grading, 337 Transesophageal echocardiography (TEE), 289, 337–347, 369, 375, 402–403 advantages of, 403 Transforming growth factor-a (TGF-a ), 522 Transfusion blood, 41 indications for, 41 Transfusion therapy, 41 physiology of, 41 Transfusion-related acute lung injury (TRALI), 282 Transthoracic echocardiography (TTE), 346–347 Transtracheal jet ventilation (TTJV), 2 Transurethral resection of bladder tumor (TURBT), 573 Transurethral resection, technique of, 573 Transverse rectus abdominis muscle (TRAM) flap, 197–198, 546, 550 Trash foot syndrome, 317 Trauma, 344 blunt, 347 chest wall, 281 colorectal, 105 complications of, 511 deaths, 390 pediatric, 396 prevention, 396–397 Trauma center volume, 393 Trauma centers, rural and urban, 395 Trauma liver injury classification, 136 Trauma surgery, techniques for, 135 Trauma systems, 390 Traumatic brain injury (TBI), 399, 511–512 severity and prognostic indicators of, 511 Traumatic coma databank (TCDB), 399 Traumatic dislocations, complications of, 421–422 Traumatic injury, community and risk factors after, 515 Treitz ligament, 119 Tricuspid regurgitation, 347 Tricuspid valve surgery, complications of, 344 Triiodothyronine, 336 Trileaflet polyurethane, 367 Trocar-site hernias, 170 Trocar technique, 233 Troponin, 347 TTE. See Transthoracic echocardiography T-tube fistula, 258 Tumor implantation, 245 Tumor necrosis factor (TNF), 59 Tumor necrosis factor-a (TNF-a), 280 Tumor, bladder, 573 Ulcer during pregnancy, 642 recurrent, 123 diagnostic tests of, 123 Ulnar nerve dysfunction, 504 Ulnar nerve palsy, 436

656

Index

Ulnar neuropathy, 474 characteristics of, 11 Unfractionated heparin (UFH), 381 Ureter, 557 Ureteral injuries complications, 612, 619 epidemiology and diagnosis, 610 management, 611 Ureteroceles, 582–583 Ureterosigmoidostomy, 576 Urethra anatomy of, 591 female reconstructive surgery of, 595–597 male reconstructive surgery of, 593–595 Urethral catheter drainage, 612 Urethral diverticulum, repair of, 596 Urethral injuries complications of, 617 epidemiology and diagnosis of, 614–615 management of, 616 posterior, 591 Urethral stricture, anterior, diagnosis of, 593 Urethrectomy, 575 Urethrovaginal fistula, repair of, 596 Urinary amylase activity, 628, 631 Urinary diversion, 576 metabolic complications of, 578 Urinary incontinence, 583–584, 593–594 sling procedures for, 564 stress retropubic repair of, 595 vaginal repair of, 595 Urinary stress incontinence, abdominal retropubic procedures for, 559 Urinary tract infection (UTI), 497, 575 Urinary tract injury, 555, 565 Urine leaks, 631 Urine, examination of, 49 Urinoma, incidence, 609 Urosepsis, 575, 577, 584 Urticaria, 45 Uterine distention, 565 Uterine injury, 637 Uterine perforation, 564–565 Uterus enlargement, 639 Vagal anatomy, 123 Vagal injury, 128 Vaginal hysterectomy, 561 Vaginal surgery, 557 Vagotomy, 114–115 and antrectomy, 112 highly selective (HSV), 112–115 and pyloroplasty, 112 truncal, 112, 114 VALI. See Ventilator-associated lung injury Valsalva maneuver, 67, 380 Valve-cutting technique, 310 Valve infection, acute, 341 Valve replacement or repair, acute complications of, 342 Vancomycin, 30, 34, 532 Vancomycin-resistant Enterococcus faecium (VRE), 181

Varicocele, 598–599 Varicose vein surgery, complications of, 320 Vas deferens, 599 Vascular cannulation injuries, 435 Vascular injuries, 168–169, 408–409, 565 associated with dislocation, 419–420 complications of, 447–449 Vascular laceration, 447 Vascular procedures, complications common to, 307 Vascular surgery, complications of, 305 Vascularity, 525 Vasculopathy, arterial, 188 Vasectomy, 599 Vasodilators, use of, 186 Vasomotor symptoms, 117 Vasopressors, 61, 280 Vein grafts, 307 Vein harvest, site of, 341 Vein-to-vein anastomosis, 17 Vena cava filter placement, complications of, 320 Venodilation, 346 Venography, 380 Venous air embolism, 67 Venous device and infection, central, 69 Venous infarcts, 487 Venous oximetry catheter, 24 Venous oxygen saturation (SvO2), 56 Venous thrombosis, 70, 163 deep, treatment for, 384 Venous–arterial (V–A) configuration, 367 Venous–venous (V–V) mode, 367 Ventilation high-frequency jet, 278 high-pressure, 275 liquid, 280 low tidal volume, 277 methods of, 273 characteristics of, 273 key features of, 273 Ventilation–perfusion, 279 mismatch, 281 Ventilation/perfusion (V/Q) scan, 383–384 Ventilator-associated lung injury (VALI), 275 Ventral hernia repair, 173 complications of, 79 Ventral hernia. See abdominal wall defects Ventricular assist devices (VADs), 367 complications of, 370–375 bleeding, 370–371 infection, 372–374 mechanical, 374 right heart failure, 374 Ventricular asystole, 299–300 Ventricular diastolic relaxation, 359 Ventricular dysfunction, 359 left, 334 Ventricular failure, 359 Ventricular fibrillation (VF), 299, 535 Ventricular function, 347 Ventricular–aortic discontinuity, 342 Ventriculitis, 486–487 Ventriculostomy, 486

Veress needle, 84 Vertigo, 34 Vesicoureteral reflux (VUR), 572, 582–583 Vesicular monoamine transporters (VMAT), 225 Vicryl1. See polyglycolic acid Video-assisted thoracic surgery (VATS), 244 Viral inactivation, 50 Viral infections, 365 Visceral swelling, 62 Visual loss, perioperative, 8 Vitamin K, deficiency in, 180 Voiding cystourethrography (VCUG), 591, 592, 596 Volkmann’s ischemic contracture, 503 Volutrauma, 275 Von Willebrand factor, 50 Von Willebrand’s disease, 307 Warfarin therapy, 413, 628 White blood cell counts, 373 Wood’s lamp, 96 World Health Organization (WHO), 219, 224 Wound care, complications of, 433–435 Wound closure and coverage, 523 poor timing of, 445–446 technical failure, 446 W-Plasty wound closure, 528 Wound complications, 192, 312, 425–426 examples of, 528 Wound failure, acute, 77 Wound healing complication, case report, 528 Wound infection, 205, 242, 309 classification of, 77 Wound length (WL), 78 Wound management, principles in, 528 Wound repair, abnormal, risk factors for, 524 Wound repair processes, 521 Wound tension, 78 Wounds complications of, 440 genetic conditions of, 526 types of, 523 Wrinkle test, 431 Wrist fractures, 502 Xenograft, 524 Xiphisternal crepitus, 71 Yersinia enterocolitica, 46 Yttrium-aluminum-garnet laser, 234 Y-V advancement flap, 105 Zollinger–Ellison syndrome (ZES), 123–124, 216–217 Zollinger–Ellison syndrome caused by multiple endocrine neoplasia 1 (ZE-MEN1), 220 Z-plasty wound closure, 528 Zyvox1, 181