Pathophysiology Study of Filler-Induced Blindness

SAVE OFFLINE

Cosmetic Medicine

Pathophysiology Study of Filler-Induced Blindness

Aesthetic Surgery Journal 2019, Vol 39(1) 96–106 © 2018 The American Society for Aesthetic Plastic Surgery, Inc. Reprints and permission: [email protected] DOI: 10.1093/asj/sjy141 www.aestheticsurgeryjournal.com

Abstract Background:  A number of authors have proposed retrograde arterial embolism as the responsible mechanism for filler-induced blindness. However, no previous human study has substantiated this proposed mechanism. Objectives:  The aim of this study was to investigate the pathophysiology of filler-induced blindness using a fresh cadaver perfusion technique. Methods:  A fresh cadaver head perfusion model that simulates both physiologic blood pressure and flow rate of the carotid artery, ophthalmic artery, and supratrochlear artery was used. The common carotid artery was cannulated and the internal jugular vein exposed for open venous drainage. A plasma-based perfusate was circulated through the cadaver head, which was connected to a perfusion system consisting of a roller pump, preload reservoir, and pressure monitor. The hyaluronic acid filler mixed with methylene blue was injected into the cannulated superficial branch of the supratrochlear artery. Cadaver dissection, angiographic study, and histology were used to investigate filler-induced blindness. Results:  Cannulation of the superficial branch of the supratrochlear artery was successful in all six cadavers. Emboli to the ophthalmic artery was successfully demonstrated in the three out of 6 fresh cadaver heads. The C-arm angiogram documented a cut-off sign in the ophthalmic artery due to hyaluronic acid filler emboli. An average intravascular volume of the intraorbital part of the supratrochlear artery was 50.0 µL. The average depth of location of the superficial branch of the supratrochlear artery from the epidermal surface was 1.5 mm. Conclusions:  Our cadaveric study demonstrated that retrograde hyaluronic acid filler emboli to the ophthalmic artery could be produced by the cannulation of the supratrochlear artery. The superficial location of the supratrochlear artery, the rich vasculature surrounding it, and the variability in the anatomy make this possible.

Editorial Decision date: May 23, 2018; online publish-ahead-of-print June 5, 2018.

Although rare, blindness following soft tissue filler injection can and does occur.1-4 With the remarkable growth in the popularity of the filler treatments, the reported incidence of this complication has correspondingly increased.4 However, given the perceived low risk of this procedure, the physician may well be ill prepared to treat such visual changes. If treatment is to be successful, it needs to be initiated within hours.5 A number of authors have proposed retrograde arterial embolism as the responsible mechanism for filler-induced blindness.1,3,6 However, no previous human study has substantiated this proposed mechanism. Further, the volume and injection pressure needed to create hyaluronic acid (HA) filler emboli is unknown.

Previous researchers have utilized perfused fresh cadavers to simulate vascular, neurosurgical, and microvascular interventions in an attempt duplicate real-life hemodynamics. 7-9 In this report, we describe From the Department of Plastic Surgery and the Cerebrovascular Center Neurological Institute, Cleveland Clinic, Cleveland, OH. Dr Zins is the Facial Surgery Section Editor for the Aesthetic Surgery Journal. Corresponding Author: Dr James E. Zins, Department of Plastic Surgery, Cleveland Clinic, 9500 Euclid Ave, Desk A60, Cleveland, OH 44195, USA. E-mail: [email protected]

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

Ki-Hyun Cho, MD, MSc; Edoardo Dalla Pozza, MD; Gabor Toth, MD, FAHA; Bahar Bassiri Gharb, MD, PhD; and James E. Zins, MD

Cho et al97

A

B

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

C

Figure 1.  Perfusion system consisting of (A) preload reservoir, (B) roller pump, and pressure monitor is connected to (C) the fresh cadaver head perfusion model of a 76-year-old female that simulates both physiologic blood pressure and flow rate of the carotid artery, ophthalmic artery, supratrochlear artery, and supraorbital artery.

a fresh cadaveric head perfusion model that simulates both physiologic blood pressure and flow rates in the carotid artery, ophthalmic artery, supratrochlear artery, and supraorbital artery and investigate the proposed mechanism of filler-induced blindness (Figure 1).

Finally, the volume and injection pressure needed to create a filler embolus by injecting hyaluronic acid in the cannulated supratrochlear and supraorbital arteries using this fresh cadaveric head perfusion model is described.

98

Aesthetic Surgery Journal Vol 39(1)

METHODS

Figure 2.  The carotid artery was cannulated for a 64-yearold male cadaver, and a roller pump was run intermittently to help dissect the supratrochlear artery. The pulsatile flow enabled easy identification of this small superficial vessel. The superficial branch of the supratrochlear artery was successfully cannulated in all 7 cadavers. HA filler was then injected into the supratrochlear artery (dashed arrow) using a 27 G blunt-tipped cannula.

result in the fluctuation of the injection pressure due to the kinesthetic and postural instability of the fingers pushing the plunger, we documented the pressure as a range. Filler injection was continued until there was no flow in the ophthalmic artery detected by color Doppler imaging. The volume of the injected filler was then measured. A C-arm (Siremobil Iso-C 3D; Siemens Medical Solutions, Erlangen, Germany) was introduced, and angiography of the ophthalmic artery was performed via the internal carotid artery to verify the emboli to the ophthalmic artery. The distribution of the embolus in the orbital vessels was studied using micro-catheter (Echelon 10, EV3 Endovascular Inc., Irvine, CA) injection of a contrast agent

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

Fresh cadaveric dissections were performed at the Cleveland Clinic Simulation & Advanced Skills Center–Surgical Skills Laboratory. Cadavers were obtained in accordance with the Declaration of Helsinki. This study was exempt from institutional review board approval. Six fresh cadaveric heads with no previous history of pathology of the head and neck regions were used for injection studies, and 1 cadaver head was used for histology. The study was carried out from January 2017 to December 2017. For preparation of the cadaveric head perfusion model, the common carotid artery was cannulated, and the internal jugular vein was exposed for open venous drainage. A Jelco® Cathlon ® I.V. Catheter, Clear, 14G × 2.0” was inserted into the common carotid and tied to the vessel wall. To remove the clots from the vessels, 2-meter tubing was connected to the faucet to infuse water, and the arterial and venous systems were pressurized and depressurized for 30-second intervals. After conditioning of the cadaver head, a perfusion system consisting of a roller pump, preload reservoir (Terumo Cardiovascular Systems Corp, Ashland, MA, USA), and pressure monitor (jewel-movement sphygmomanometer, Tycos, Arden, NC, USA) was connected to the catheter (Figure 1). The pressure monitor was connected to the Luer-lock connectors. Then, 5 liters of plasma-based perfusate, which is a mixture of Flexbumin (human albumin 25%, Baxter, Westlake Village, CA, USA), swine blood, pigmented dye, saline, and sodium heparin, were circulated through the cadaver head with open venous drainage. The flow rate of perfusion pump was controlled between 0.4 and 1.0 l/min to achieve systolic pressure less than 120 mmHg. With the cadaver placed in supine position, a SonoSite Edge II (Sonosite, Bothell, Washington, USA) L25x transducer was applied gently to the closed upper eyelid without pressure with coupling gel. Blood flow through the ophthalmic artery was confirmed by color Doppler imaging using the method previously described by Greenfield et al.10 Additionally, the sonographic investigation was performed in the orbits of the perfused cadaver to detect supratrochlear arteries pulsating in both orbits and forehead. The supraorbital region was subcutaneously dissected followed by cannulation of the superficial branch of the supratrochlear artery using a 27-gauge × 1”, 25 mm, blunt-tipped cannula (Figure 2).11,12 0.29 mL of hyaluronic acid injectable gel (Juvéderm Ultra, Allergan, Inc, Irvine, CA, USA) was thoroughly mixed with 0.01 mL of methylene blue and placed in a 1-mL syringe. The HA filler was injected against the antegrade flow of the perfusate in the supratrochlear artery. Injection pressure was measured with the disposable blood pressure transducers connected to the pressure monitor. As a manually driven syringe may

Cho et al99

A

B

(Omnipaque 300 mg/mL, GE Healthcare, Princeton, NJ, USA) as described by Kim et al and Wei et al.13,14 The perfusion circuit was disconnected, and osteotomies were made to remove the orbital part of the frontal bone and orbital roof allowing visualization of the orbital structures. Frontal lobectomy was performed, and the bony walls of the optic canal were removed with a high-speed drill to observe the branching pattern of the orbital vessels and the distribution of the filler emboli stained with methylene blue. Using a microcaliper, precise measurement was made of the distance from the midline to the point where the supratrochlear artery crosses the supraorbital rim, the depth of the superficial branch of the supratrochlear artery from epidermal surface, the length of the supratrochlear artery from the point where it crosses the supraorbital rim to the branching point of the ophthalmic artery, and the diameter of the supratrochlear artery at the point where it crossed the superior orbital rim. Statistical analysis included mean and standard deviations using Stata (College Station, TX, software). A histologic study was performed to microscopically locate the depth of the superficial branch of the supratrochlear artery and calculate the vessel diameter, thus confirming earlier gross measurements. A 20 × 10 mm rectangle with its longer sides in the horizontal direction was drawn on the skin surface approximately 20 mm above the superior margin of the right or left eyebrow along the

midsagittal line. A 20 × 10 mm block of cadaveric tissue, which included the superficial branch of the supratrochlear artery, was then obtained. The block was then embedded in paraffin and cut into 6 μm sections in the horizontal plane. The sections were stained with hematoxylin and eosin (H&E). Images were captured using a light microscope (model BX50; Olympus Corp, Tokyo, Japan) and an image analyzer (Image Pro Plus, version 5; Media Cybernetics, Bethesda, Maryland).

RESULTS Cannulation of the superficial branch of the supratrochlear artery was successful in all 6 cadavers. There were 3 male cadavers and 3 female cadavers. The mean age was 71.7 years (range, 65-80). After the retrograde injection of HA filler, cadaver dissection verified the HA/methylene blue emboli in the ophthalmic artery in 3 of 6 cadaver heads (Figures 3, 4B, and 5B). The supratrochlear artery was found to branch off proximal to the corrugator supercilii muscle to give rise to the superficial branch and the deep periosteal branch (Figure 5B).11,12 The deep branch of the supratrochlear artery then penetrated the corrugator supercilii muscle before it reached the pericranium. The superficial branch initially coursed cephalad between the corrugator and orbicularis oculi muscles and then ran superficial to

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

Figure 3.  Presented here is the C-arm angiographic findings of a 67-year-old female cadaver—Lt. orbit, lateral view. (A) Prior to the filler/methylene blue injection, contrast agent was injected into the superficial branch of the supratrochlear artery as a baseline angiogram to confirm the patency of the supratrochlear artery (yellow arrow). (B) After the filler/methylene blue injection, the cut-off sign at the branching point of the ophthalmic artery is shown (green arrow).

100

A

Aesthetic Surgery Journal Vol 39(1)

B

C

the frontalis muscle to travel in the subcutaneous plane from 15 mm to 25 mm above the orbital rim (Figure 6). Anastomoses of the supratrochlear artery, the dorsal nasal artery, with the angular artery were below the eyebrow along the side of the radix (Figure 5B). The average distance from the midline to the point where the supratrochlear artery crosses the supraorbital rim was 18.1 mm (SD, 2.1; range, 15.2-21.3 mm) (Table 1). The average depth of the superficial branch of the supratrochlear artery from the epidermal surface was 1.5 mm (SD, 0.2; range, 1.3-1.8 mm) (Figure 6). The average diameter of the supratrochlear artery at the point where it crosses the superior orbital rim was 1.13 mm (SD, 0.34; range, 0.84-1.56 mm) (Figure 5A). The diameter of the supratrochlear artery in those cadavers that were successfully embolized was 1.42 mm (range, 1.231.56 mm), whereas the average diameter was 0.84 mm (range, 0.84-0.93 mm), significantly smaller for the unsuccessful ones. The average injection pressure needed to embolize the ophthalmic artery was 166.7 mmHg (range, 160-180 mmHg). The average length of the supratrochlear artery from the point where it crosses the supraorbital rim to the branching point of the ophthalmic artery was 51.1 mm (SD, 11.25; range, 36.6-67.4 mm) (Figure 5A). The calculated intraorbital volume of the supratrochlear artery after branching off from the ophthalmic artery ranged from 30.1 µL to 80.2 µL. The average vessel volume was 50.0 µL. The filler/methylene blue emboli were noted not only in the orbital arteries but also in the dorsal nasal artery and anastomotic channels between the supratrochlear artery and the supraorbital artery (Figure 5B).

Histologic Studies Microscopic observations of the forehead skin specimens from the seventh cadaver revealed the following findings. Depth of the superficial branch of the supratrochlear artery was 1.628 mm from the epidermal surface (Figure 7). The circumference of the artery was 4.222 mm. Considering it as a cylindrical structure, calculated luminal diameter of the artery was 1.34 mm.

DISCUSSION Experimental Study Design Human cadaver circulation models have been used to help in the design of a variety of neurovascular and endovascular devices because they so closely simulate the clinical situation.15,16 We have borrowed from those designs in order to investigate the pathophysiology of filler-induced blindness. In a review of the literature detailing filler-induced visual changes, the glabella was the most common injection site implicated accounting for 38 of 98 (38.8%) cases in the largest reported study.4 Therefore, the supratrochlear artery was the injection site chosen in this study.

Anatomic Findings of the Branches of the Ophthalmic Artery The supratrochlear artery, the terminal branch of the ophthalmic artery, has been described as most commonly piercing the orbital septum to enter the forehead and scalp. But it may also arise anterior to the septum anastomosing

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

Figure 4.  Lt. orbit, lateral view of the C-arm angiogram of a 71-year-old male cadaver showing (A) cut-off sign at the branching point of the ophthalmic artery (arrow). (B, C) Axial cut of the 72-year-old female cadaver head demonstrating HA/ methylene blue embolism in the supratrochlear artery (yellow arrow). Skin layer (blue arrow) has been reflected to expose the axial section of the ethmoid bone (●), globe (*), and the optic nerve (black arrow).

Cho et al101

A

B

Figure 5.  (A) The average distance from the midline to the point where the supratrochlear artery crossed the supraorbital rim was 18.1 mm (range, 15.2-21.3 mm). The average length of the supratrochlear artery from the point where it crossed the supraorbital rim to the branching point of the ophthalmic artery was 51.1 mm (range, 36.6-67.4 mm). The average diameter of the supratrochlear artery where it crossed the supraorbital rim was 1.13 mm (range, 0.84-1.56 mm). (B, C) A 67-year-old female cadaver. The supratrochlear artery (blue arrow) was the terminal branch of the ophthalmic artery. 0.3 mL of HA filler resulted in vascular occlusion in not only the supratrochlear artery but also dorsal nasal (yellow arrow) and angular artery (black arrow), which are anastomosed near the radix (*).

with the surrounding arteries (Figure 5B).17,18 In all 7 of our cadavers, the supratrochlear artery pierced through the orbital septum and formed a network of anastomoses, consistent with previous reports.19 The vascular relationship between the angular artery, the supratrochlear artery, and dorsal nasal artery, however, was not always consistent with previous reports.20 The dorsal nasal artery pierced the orbital septum above the medial palpebral ligament to run on top of the nose (Figure 5B). It represents the other terminal branch of the ophthalmic artery, and it establishes an important anastomosis with the angular branch of the facial artery.17,18

Anatomic Findings of the Supratrochlear Artery Both our study and the work of others highlight why the supratrochlear vessels are so frequently implicated in filler-induced blindness. We found that the supratrochlear artery is surprisingly superficial, lying 1.5 mm (SD, 0.2; range, 1.3-1.8 mm) deep to the epidermal surface. This makes inadvertent cannulation or injection very possible. This finding was contrary to a previous report suggesting superficial injection in the glabellar region is relatively safe.21,22

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

C

102

Aesthetic Surgery Journal Vol 39(1)

B

C

D

Figure 6.  Cadaver dissection revealing the depth of the superficial branch of the supratrochlear artery for a 72-year-old male cadaver. (A, C) HA filler/methylene blue was injected into the superficial plane to demonstrate the clinical scenario. Leakage of the blood from the superficial branch of the supratrochlear artery can be seen (yellow arrow). Radix (*) (B, D) Left lateral view of the cadaver head of a 67-year-old female. The black circle shows the supratrochlear artery crossing the orbital roof. The average depth of location of the superficial branch of the supratrochlear artery from the epidermal surface was 1.5 mm. Radix (*), angular artery (black arrow), dorsal nasal artery (orange arrow), superficial branch of the supratrochlear artery (blue arrow).

The distance of the supratrochlear artery from the midline has been studied by others.19,23 However, we found the location of the supratrochlear artery far more lateral than previous reports. Our study found the vessel to be 18.1 ± 2.1 mm (range, 15.2-21.3 mm) from the midline, whereas Tansatit et al reported an average distance of 11.3 ± 1.9 mm (range, 9-19 mm) and Schwenn et al a distance of 16.4 ± 2.2 mm (range, 14-21 mm) from the midline (Table 1). In addition, we found the vascular relationships among the angular artery, the supratrochlear artery, and dorsal nasal artery form both a very rich and inconsistent anatomic relationship. This rich vascularity and variable anatomy enhance the likelihood of needle induced misadventure.

The outer diameter of the supratrochlear artery was measured both grossly and histologically. Gross measurement of the supratrochlear artery demonstrated an outer diameter of 1.13 ± 0.34 mm (range, 0.84-1.56 mm). Microscopic finding demonstrated a calculated luminal diameter of 1.34 mm, which is larger than an 18 gauge needle (1.27 mm).24 Others have found the diameter to be somewhat smaller.23,25,26 In any case, as the diameter of the vessel may well be larger than the injection needle used, cannulation or penetration of the lumen is readily possible. Those cadavers that were unsuccessfully embolized had on average a smaller diameter of the supratrochlear artery. In order to estimate the volume of filler needed to create a filler embolism, the intravascular volume of the

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

A

Cho et al103 Table 1.  Supratrochlear Artery Distance From Midline, Depth, Length, Diameter, and Volume Measurements With Means and Standard Deviations Cadaver 2

Cadaver 3

Cadaver 4

Cadaver 5

Cadaver 6

Female

Male

Female

Male

Female

Female

Age (years)

67

71

76

64

72

80

Eye laterality

OS

OS

OS

OS

OD

OD

Distance from the midline to the point where the supratrochlear artery crosses the supraorbital rim (mm)

17.8

21.3

15.2

18.5

19.4

Depth of superficial branch of supratrochlear artery from epidermal surface (mm)

1.4

1.7

1.3

1.8

Length of the supratrochlear artery from the point where it crosses the supraorbital rim to the branching point of the ophthalmic artery (mm)

42.3

36.6

54.3

Diameter of supratrochlear artery at the point where it crosses the superior orbital rim (mm)

1.23

1.56

Intra-orbital volume of supratrochlear artery after branching off from the ophthalmic artery* (µl)

50.2

Emboli in the ophthalmic artery after retrograde injection of HA filler

Gender

Location of the filler embolism

Mean

SD

71.7

5.8

16.6

18.1

2.1

1.4

1.4

1.5

0.2

58.5

47.3

67.4

51.1

11.25

0.84

0.93

1.47

0.75

1.13

0.34

69.9

30.1

39.7

80.2

29.7

50.0

21.1

Success

Success

Fail

Fail

Success

Fail

Ophthalmic a.

Ophthalmic a.

Ophthalmic a.

SD, standard deviation; OD, oculus dexter (right eye); OS, oculus sinister (left eye), HA, hyaluronic acid. * Calculated volume based on axial length (l), diameter, and radius (r) of the supratrochlear. (V = πr2l). It is estimated volume, and may not accurately reflect intravascular volume.

supratrochlear artery and ophthalmic artery from the glabella to the orbital apex was calculated as 50 µL (SD, 21.1; range, 30.1-80.2 µL). This is similar to the findings of Schwenn et al and Khan et al.23,25 In any case, the volume needed to induce an embolus in both models is exceedingly small.

Injection Pressure for Embolism of Filler The average injection pressure that was needed to embolize the ophthalmic artery was 166.7 mmHg (Table 2). Monitoring of the injection pressure showed that it is easy for the injector to exert the pressure well above 200 mmHg within 2 to 3 seconds. Also, it was not easy to keep the injection pressure below 80 mmHg, as the injector needs a certain amount of pressure on the syringe to push the material through a small needle. Whether the injector needs to overcome the diastolic blood pressure or systolic blood pressure to create an embolus in the ophthalmic artery is controversial. Paul et al claimed that corticosteroid embolization can occur when injection pressure exceeds the diastolic pressure of the ophthalmic artery.27 On the other hand, Li et al suggested that injection force needs to overcome the systolic arterial pressure in order to push the injected droplets of filler proximally along the ophthalmic artery.21 In our study, the injection pressure above the systolic arterial pressure was needed to transfer

the filler into the ophthalmic artery. This reaffirms the clinical recommendation that excessive syringe pressure should be avoided during the filler procedure. The pressure during the injection is likely to be dispersed to multiple vessels rather than transmitted to one artery. This is consistent with the present dissection findings, which showed that the filler was not only injected into the orbital vessels but also into the dorsal nasal artery and anastomotic channels between the supratrochlear artery and the supraorbital artery (Figures 4 and 5). This is consistent with clinical findings of skin necrosis occurring in the radix and nasojugal groove after filler injection to the glabella.13 Although our experiment did not measure the flow rate of the filler through the cannula, this is also another important factor that can affect the retrograde embolism. A slow injection rate affects the viscosity of HA filler, which demonstrates the typical behavior of non-Newtonian liquids.28 With increasing shear rate of HA filler during the injection, viscosity drops.29 In contrast, the viscosity of the HA filler increases at low flow rates. As a result, the resistance of the filler is greater during the initiation of the injection than when the flow is established. Therefore, the injector needs to apply a greater amount of pressure to initiate flow.30 This higher initial pressure is directly transmitted during the injection. Therefore a slow injection with a high initial injection pressure combined with a sufficient volume of HA filler can increase the possibility of retrograde embolism.

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

Cadaver 1

104

Aesthetic Surgery Journal Vol 39(1)

A

C

Figure 7.  Microscopic finding of the depth of cutaneous arteries from the epidermal surface and circumference on the forehead area of a 72-year-old male cadaver. (A, B, C) The depth was 1.628 mm, and the luminal diameter was 1.34 mm, which is larger than an outer diameter of the 20G needle.

Angiographic Findings in Retrograde Filler Injection Angiography has been used to study various orbital vasculature. Kim et al have used cerebral angiograms to localize filler-induced ophthalmic embolic obstruction sites and characterize the pattern of distal angiographic runoff.13 Wei et al described use of the C-arm

angiography as a reasonable and practical diagnostic tool to identify vascular lesions intraoperatively.14 We also used the C-arm for our study. It should, however, be noticed that the C-arm presents with relatively reduced image resolution and restricted imaging angle compared to a standard biplane digital subtraction angiography.14 A high-resolution angiogram would likely have shown a more precise level of obstruction.

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

B

Cho et al105 Table 2.  Outcome of Emboli in the Ophthalmic Artery, Supratrochlear Artery Diameter, and Injection Pressure With Means

Gender Age (years) Emboli in the ophthalmic artery after retrograde injection of HA filler Diameter of the supratrochlear artery at the point where it crosses the supraorbital rim (mm)

Cadaver 1

Cadaver 2

Cadaver 3

Cadaver 4

Cadaver 5

Cadaver 6

Female

Male

Female

Male

Female

Female

67

71

76

64

72

80

Success

Success

Fail

Fail

Success

Fail

1.23

1.56

1.47

0.84 160

160

Future Directions for Further Filler Research

1.42

0.75 180

200<

200<

71.7

0.84 166.7

200<

200<

CONCLUSIONS

The results of the study can well be extended. Possible avenues for further research include evaluation of various hyaluronidase dosage and treatment intervals on intravascular HA dissolution, the effect of previous surgery (ie, vascular anatomy alteration) on filler embolus propagation, and the size of vessels on embolus propagation.

In this study, we demonstrate that retrograde filler embolus to the ophthalmic artery can be reproduced by cannulation of the supratrochlear artery. This may be the pathophysiology, which results in filler related visual changes. Further, because the supratrochlear artery is so superficial, ophthalmic artery emboli are possible even with superficial filler injection. The superficial location of the supratrochlear artery, the rich vasculature surrounding it, and the variability of the anatomy make the possibility of filler-induced blindness real.

Limitations of the Study

Disclosures

The numbers of the cadavers included in the study were small, and the age group did not represent the typical filler patients. Although our model simulates both physiologic blood pressure and flow rate, it falls short of true physiology. For example, it lacks a vasoconstriction or closure of the collateral circulation, which can occur under external stimuli in physiologic conditions.31 Also, the cannulation of the artery was done under direct vision, which affects the success rate of the intravascular injection of the HA filler. The mean age of the 7 cadavers included in this study was 71.7 years (range, 65-80). According to the cosmetic surgery national data bank statistics, in 2016, people aged 35 to 60 accounted for 37.4% of hyaluronic acid filler treatment followed by age group 51 to 64 (34.4%).32 People the age 65 and older accounted for 13.3% of the HA injection. Therefore, the mean age of the cadavers was older than the typical filler patients. Finally, this study sends a cautionary note to the clinician. The practicing plastic surgeon should not be reassured by the rarity of this event. This study suggests that because of the superficial location of the supratrochlear artery, the rich vascularity surrounding it, and the variability in this anatomy the possibility of filler-induced blindness is real. And while this is a minor office procedure, the clinician should be prepared for this major event.

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding This study was supported by an Aesthetic Surgery Education and Research Foundation (ASERF) grant. The perfusion system, hyaluronic acid filler, albumin, saline, catheter, histology, and medical illustrations were supported by the ASERF grant.

REFERENCES 1. Coleman SR. Avoidance of arterial occlusion from injection of soft tissue fillers. Aesthet Surg J. 2002;22(6):555-557. 2. Ozturk CN, Li Y, Tung R, Parker L, Piliang MP, Zins JE. Complications following injection of soft-tissue fillers. Aesthet Surg J. 2013;33(6):862-877. 3. Carruthers JD, Fagien S, Rohrich RJ, Weinkle S, Carruthers A. Blindness caused by cosmetic filler injection: a review of cause and therapy. Plast Reconstr Surg. 2014;134(6):1197-1201. 4. Beleznay K, Carruthers JD, Humphrey S, Jones D. Avoiding and treating blindness from fillers: a review of the world literature. Dermatol Surg. 2015;41(10):1097-1117. 5. Zhu GZ, Sun ZS, Liao WX, et al. Efficacy of retrobulbar hyaluronidase injection for vision loss resulting from hyaluronic acid filler embolization. Aesthet Surg J. 2017;38(1):12-22.

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

Injection pressure (mmHg)

0.93

Mean

106

20. Kelly CP, Yavuzer R, Keskin M, Bradford M, Govila L, Jackson IT. Functional anastomotic relationship between the supratrochlear and facial arteries: an anatomical study. Plast Reconstr Surg. 2008;121(2):458-465. 21. Li X, Du L, Lu JJ. A novel hypothesis of visual loss secondary to cosmetic facial filler injection. Ann Plast Surg. 2015;75(3):258-260. 22. Wu S, Pan L, Wu H, et al. Anatomic study of ophthalmic artery embolism following cosmetic injection. J Craniofac Surg. 2017;28(6):1578-1581. 23. Schwenn OK, Wüstenberg EG, Konerding MA, Hattenbach LO. Experimental percutaneous cannulation of the supraorbital arteries: implication for future therapy. Invest Ophthalmol Vis Sci. 2005;46(5):1557-1560. 24. Goldman MP, Weiss RA. Sclerotherapy E-book: Treatment of Varicose and Telangiectatic Leg Veins. 5th ed. Philadelphia, PA: Saunders; 2011. 25. Khan TT, Colon-Acevedo B, Mettu P, DeLorenzi C, Woodward JA. An anatomical analysis of the supratrochlear artery: considerations in facial filler injections and preventing vision loss. Aesthet Surg J. 2017;37(2):203-208. 26. Schmidt D, Adelmann G. Is it feasible to use the supratrochlear artery for inducing intra-arterial fibrinolysis in cases of central retinal artery occlusion? An anatomical investigation. Neuro Ophthalmol. 2009;15(5):265-270. 27. Paul S, Hoey MF, Egbert JE. Pressure measurements during injection of corticosteroids: in vivo studies. Med Biol Eng Comput. 1999;37(5):645-651. 28. Ilyin SO, Kulichikhin VG, Malkin AY. The rheological characterisation of typical injection implants based on hyaluronic acid for contour correction. Rheologica Acta. 2016;55(3):223-233. 29. Cowman MK, Schmidt TA, Raghavan P, Stecco A. Viscoelastic properties of hyaluronan in physiological conditions. F1000Res. 2015;4:622. 30. Egbert JE, Paul S, Engel WK, Summers CG. High injection pressure during intralesional injection of corticosteroids into capillary hemangiomas. Arch Ophthalmol. 2001;119(5):677-683. 31. Zheng H, Qiu L, Liu Z, et al. Exploring the possibility of a retrograde embolism pathway from the facial artery to the ophthalmic artery system in vivo. Aesthetic Plast Surg. 2017;41(5):1222-1227. 32. Cosmetic surgery national data bank statistics. Aesthet Surg J. 2017;37(Suppl 2):1-29.

Downloaded from https://academic.oup.com/asj/article-abstract/39/1/96/5033292 by guest on 25 August 2019

6. DeLorenzi C. Complications of injectable fillers, part 2: vascular complications. Aesthet Surg J. 2014;34(4):584-600. 7. Pham M, Kale A, Marquez Y, et al. A perfusion-based human cadaveric model for management of carotid artery injury during endoscopic endonasal skull base surgery. J Neurol Surg B Skull Base. 2014;75(5):309-313. 8. Wolff KD, Fichter A, Braun C, Bauer F, Humbs M. Flap raising on pulsatile perfused cadaveric tissue: a novel method for surgical teaching and exercise. J Craniomaxillofac Surg. 2014;42(7):1423-1427. 9. Carey JN, Minneti M, Leland HA, Demetriades D, Talving P. Perfused fresh cadavers: method for application to surgical simulation. Am J Surg. 2015;210(1):179-187. 10. Greenfield DS, Heggerick PA, Hedges TR 3rd. Color Doppler imaging of normal orbital vasculature. Ophthalmology. 1995;102(11):1598-1605. 11. Reece EM, Schaverien M, Rohrich RJ. The paramedian forehead flap: a dynamic anatomical vascular study verifying safety and clinical implications. Plast Reconstr Surg. 2008;121(6):1956-1963. 12. Yoshioka N, Rhoton AL. Forehead and orbital region. In: Hiscock T, Landis S, eds. Atlas of the Facial Nerve and Related Structures. 1st ed. New York: Thieme; 2016:23-26. 13. Kim YK, Jung C, Woo SJ, Park KH. Cerebral angiographic findings of cosmetic facial filler-related ophthalmic and retinal artery occlusion. J Korean Med Sci. 2015;30(12):1847-1855. 14. Wei Z, Garzon-Muvdi T, Yang W, et al. Utility of intraoperative diagnostic C-arm angiography for management of high grade subarachnoid hemorrhage. Interdiscip Neurosurg. 2015;2(2):98-102. 15. Garrett HE Jr. A human cadaveric circulation model. J Vasc Surg. 2001;33(5):1128-1130. 16. Carey JN, Rommer E, Sheckter C, et al. Simulation of plastic surgery and microvascular procedures using perfused fresh human cadavers. J Plast Reconstr Aesthet Surg. 2014;67(2):e42-e48. 17. Hayreh SS. The ophthalmic artery: III. Branches. Br J Ophthalmol. 1962;46(4):212-247. 18. Hollinshead WH. Anatomy for Surgeons: The Head and Neck. Vol 1. 1st ed. New York: Hoeber; 1954:156. 19. Tansatit T, Apinuntrum P, Phetudom T. An anatomic basis for treatment of retinal artery occlusions caused by hyaluronic acid injections: a cadaveric study. Aesthetic Plast Surg. 2014;38(6):1131-1137.

Aesthetic Surgery Journal Vol 39(1)