A Computed Tomographic Investigation of the Ophthalmic Artery Volume and Its Relevance to Soft Tissue Filler Injections

Abstract

Background

The measured intraarterial volume of cadaveric ophthalmic arteries was utilized for safety recommendations during facial soft tissue filler injections. However, its clinical practicability and model applicability have become questionable.

Objectives

To measure the volume of the ophthalmic artery in living individuals by utilizing computed tomography (CT) imaging technology.

Methods

A total of 40 Chinese patients (23 males, 17 females) were included in this study with a mean age of 61.0 (14.2) years and a mean body mass index of 23.7 (3.3) kg/m². Patients were investigated with CT imaging technology to evaluate the length, diameter, and volume of the bilateral ophthalmic arteries as well as the length of the bony orbits, resulting in a total of 80 investigated ophthalmic arteries and orbits.

Results

Independent of gender, the average length of the ophthalmic artery was 80.6 (18.7) mm, the calculated volume of the ophthalmic artery was 0.16 (0.05) mL and the minimal and maximal internal diameter of the ophthalmic artery were 0.50 (0.05) mm and 1.06 (0.1) mm, respectively.

Conclusions

Based on the results obtained from the investigation of 80 ophthalmic arteries it must be concluded that current safety recommendations should be reevaluated. The volume of the ophthalmic artery appears to be 0.2 mL rather than 0.1 mL as previously reported. In addition, it appears impractical to limit the volume of soft tissue filler bolus injections to 0.1 mL due to the aesthetic requirements of each individual patient and treatment plan.

See the Commentary on this article here.

According to the annual report released by The Aesthetic Society, the number of soft tissue filler injections increased by 42% between 2020 and 2021, with a total of 1,857,339 procedures performed in the United States in 2021. The number of soft tissue filler reversal procedures also increased during the same period by 57%, with 23,031 corrective procedures performed in 2021, including the administration of hyaluronidase with or without ultrasound guidance.¹

Despite the increasing number of soft tissue filler injections performed worldwide, the number of reported serious adverse events involving the ophthalmic artery circulation are relatively low.^2–4 Potential reasons for this are the preventive measures conducted previous to or during the injection process, including knowledge of 3-dimensional (3D) facial anatomy, respecting injection biomechanics and product rheology, performing preinjection aspiration, injecting with low plunger pressure and injection speed, and injecting small boluses (<0.1 mL at a time).^5–19

The latter preventive measure is based on an anatomic study by Khan et al published in 2016, which estimated the volume of the ophthalmic artery in 4 latex-injected human body donors with the water displacement method. The intraluminal latex was utilized to determine the intraarterial volume of the investigated ophthalmic arteries, and the authors found an average volume of 0.085 mL (range of 0.04 to 0.12 mL). The authors concluded that this volume was sufficient to cause vascular occlusion of the ophthalmic artery circulation and that injectors should be mindful and inject less than this threshold volume to avoid injection-related visual compromise (IRVC).²⁰

The advised 0.1-mL limit to soft tissue filler bolus injections, however, seems to be of reduced clinical practicability for customary facial aesthetic procedures. Other factors such as patients’ aesthetic desires, aesthetic needs identified by the injector, the budget of the patient, or the overall facial aesthetic treatment plan may be more important to how much volume is administered during various bolus injections. In addition, it is questionable whether the advised 0.1-mL limit truly prevents embolization of retinal and periorbital vessels and helps avoid IRVCs. The objective of this study is therefore 2-fold: (1) to investigate the volume of the ophthalmic artery with contrast agent–enhanced computed tomography (CT) scanning methodology in a large sample (n = 40) of living individuals; and (2) to reevaluate and discuss the practicability of the 0.1-mL limit for bolus injections based on the current status of the literature. It is hoped that the investigations conducted and their analysis will help create awareness and reevaluate currently accepted standards for safety during facial soft tissue filler injections.

METHODS

Study Setup

This cross-sectional CT imaging–based study was conducted between April and November 2022 and enrolled consecutive patients of the Department of Radiology, Guangdong Second Provincial General Hospital, Guangzhou, Guangdong Province, China. Preexisting CT imaging data sets were retrospectively analyzed for the morphology and intraarterial volume of their ophthalmic arteries. Preexisting CT scans were sampled from the radiology database and were not acquired specifically for the purposes of this study.

CT imaging data sets were included in this study if the internal carotid artery (ICA) displayed normal anatomy, without contrast agent–filling defects, indicating the absence of atheromatous plaques and congenital or tumoral vascular pathologies. Presence of common carotid artery or ICA aneurysms, changes in vessel wall morphology, craniocervical neoplasms, or bony malformations of the skull base or the orbit were regarded as exclusion criteria.

Written and informed consent was obtained at the initial CT scan from all patients; data were not accessed without a signed consent form for the release of demographic and imaging-related data. The study protocol was approved by the institutional review board (IRB) under the number 2022-KY-KZ-310-01.

Computed Tomographic Imaging

All CT examinations were performed with the same scanner (Brilliance 6000 iCT, Philips Healthcare; Amsterdam, the Netherlands) with the following parameters: slice thickness 0.9 mm; slice increment 0.45 mm; pitch 1.0; table feed 40 mm/rotation; 0.9 mm reconstruction interval; matrix 512 × 512; 180- to 240-mm field of view; 120 kV; and 200 mA. Contrast agent was administered by cubital venous access with a power injector at a rate of 5.0 mL/s, and a total of 50.0 mL of contrast agent (Iohexol, 350 mg iodine/mL; GE Healthcare, Shanghai) was injected per patient scanned. Scan delay was individually adapted by bolus tracking, with a region of interest placed at the level of the aortic arch and automatic triggering at 150 HU (Hounsfield units). Once identified, relevant CT scans were transferred to a workstation (IntelliSpace, Philips Healthcare) for processing.

The ophthalmic artery was isolated and identified with the subtraction algorithm of the workstation software, which removed nonenhanced imaging data from contrast-enhanced data, revealing the filling of the arterial system exclusively. This exposed filling represented the intraarterial volume of the arterial system as visualized during CT scanning. The multiplanar reformation algorithm was employed for tracking tortuous arteries, accounting for a nonlinear arterial pathway and for nonfilling artifacts (Figures 1-3).

Conducted Measurements

The ophthalmic artery was identified and measured between the optic canal and the superior orbital rim, including the branching pattern into the most prominent arterial stem; this included the branching into the intraorbital segment of the supratrochlear or supraorbital artery. The following measurements were conducted:

• Length of the ophthalmic artery;

• Maximal and minimal internal diameters of the ophthalmic artery;

• Volume of the ophthalmic artery;

• Length of the bony orbit.

Statistical Analysis

Differences between genders and ages were calculated by independent t test. Bivariate correlations relied on the calculation of Pearson's correlation coefficient. All calculations were run with SPSS Statistics 25 (IBM; Armonk, NY), and differences were considered statistically significant at a probability level of ≤.05 to guide conclusions.

RESULTS

Patient Demographic Data

A total of 40 patients (23 males, 17 females) were included in this study, with 100% Chinese ethnic background. Mean age was 61.0 (14.2) years (range: 24-88) and the mean body mass index was 23.7 (3.3) kg/m² (17.5-32.4). Bilateral measurements were conducted resulting in a total of 80 investigated ophthalmic arteries.

Orbital Length

The overall average length of the orbit was 46.2 (2.6) mm (41.2-53.5), with 47.1 (2.6) mm (41.5-53.5) in males, 45.0 (2.2) mm (41.2-49.4) in females, and P < .001 representing a statistically significant shorter bony orbital conus in females (Table 1). Higher age did not statistically significantly correlate with a longer distance between the optic canal and superior orbital rim, with r_p = 0.194 and P = .085.

Ophthalmic Artery Length

The overall average length of the ophthalmic artery was 80.6 (18.7) mm (48.4-143.6), with 81.5 (16.4) mm (57.0-124.6) in males and 79.4 (21.7) mm (48.4-143.6) in females, and P = .617 indicating a similar length in both genders (Table 1, Figure 4). Higher age and higher BMI did not statistically significantly correlate with an increased vessel length, with r_p = −0.051/0.159 and P = .654/.159, respectively. Interestingly, no statistically significant correlation was found between the length of the artery and the length of the bony orbit, with r_p = 0.076 with P = .504.

Internal Diameters of the Ophthalmic Artery

The overall average minimal internal diameter of the ophthalmic artery was 0.50 (0.05) mm (0.45-0.64), with males having a minimum diameter of 0.50 (0.06) mm (0.45-0.64) and females having a minimum diameter of 0.50 (0.05) mm (0.45-0.56), without statistical difference between genders (P = .085). The overall average maximal internal diameter of the ophthalmic artery was 1.06 (0.1) mm (0.70-1.60), with males having a maximum diameter of 1.07 (0.1) mm (0.79-1.20) and females having a maximum diameter of 1.05 (0.2) mm (0.70-1.60), without a statistical difference between genders (P = .407) (Table 1, Figure 4). The minimum diameter showed a statistically significant negative correlation with higher age (r_p = −0.228 with P = .042) indicating a smaller internal circumference at higher age. Interestingly, the maximum diameter showed a statistically significant positive correlation with higher age (r_p = 0.358 with P = .001) indicating a greater internal circumference at higher age.

Ophthalmic Artery Volume

The overall average contrast agent–filled volume of the ophthalmic artery was 0.16 (0.05) mL (0.10-0.35), with 0.17 (0.05) mL (0.11-0.35) in males and 0.16 (0.04) mm (0.10-0.24) in females, with P = .226 (Table 1, Figure 4). No relationship with higher age was identified, with r_p = 0.186 and P = .099.

DISCUSSION

This study was designed to verify and expand on a previous anatomic study published by Kahn and colleagues in 2016. That study investigated 4 human body donors with the water displacement method and reported that the average volume of the latex-injected ophthalmic arteries was 0.085 mL.²⁰ Partially due to these findings, preventive safety recommendations include limiting soft tissue filler boluses injected to less than 0.1 mL to avoid embolizing the crucial ophthalmic artery circulation.²¹ However, such recommendations need to be critically evaluated based on the methodology of the initial study, on the practicality for regularly performed soft tissue filler injections, and on the suggested pathologic mechanism that results in IRVCs.

Regarding the study methodology, the lack of systolic blood pressure in human body donors results in postmortem (relative) vasoconstriction and a smaller internal diameter of arteries.^22,23 Additionally, incomplete filling of the arterial tree with latex during the injection process and the presence of blood clots if not flushed with anticoagulant substances (heparin) previous to the latex injection process can result in false lower values for measured intraarterial volumes in the human body donor study design. Moreover, a small study sample size (n = 4) could increase the inaccuracy of the previous results presented.²⁰ These limitations were accounted for in the present study: the study sample included n = 40 (living) individuals, allowing for the measurement of n = 80 ophthalmic arteries. The technology for measuring the anatomy and the volume of the ophthalmic artery was based on CT scanning with a slice thickness of 0.9 mm and on computed software algorithms that utilized standardization of the Hounsfield units starting at the aortic arch and multiplanar reconstructions to account for the torturous intraorbital arterial course. The results of both studies were similar, with an average ophthalmic artery volume of 0.085 mL (anatomic study) and 0.16 mL (CT study). The greater values obtained in the present study are consistent with the aforementioned differences between living and cadaveric tissue and with the greater accuracy of the methodology applied. No gender differences were observed in the volume measurements obtained, with 0.17 mL in males and 0.16 mL in females (P = .226), allowing for generalizability of results. Similarly, no gender differences were identified in the measured length of the investigated ophthalmic arteries, 81.5 mm in males and 79.4 mm in females with P = .617. The CT-based evaluated ophthalmic artery length was longer than the measurements reported in the anatomic model (51.6 mm), which is consistent with the absence of systolic blood pressure and the tortuous pathway of the ophthalmic artery.

Regarding the practicality of the aforementioned recommendations, aesthetic practitioners do not routinely limit the injected soft tissue filler volume to 0.1 mL or less when performing large-volume treatments. The volume is rather based on the targeted aesthetic defect. Some injection techniques require larger volumes to be injected to achieve the desired outcome; this is the case, for instance, for the temporal lifting technique, which was reported to require 1.0 mL of high G prime filler when administered with a cannula in the subdermal plane of the posterior superior temple.^24–26 Additionally, the rheology of the product needs to be factored into the aesthetic treatment plan, with some areas such as the piriform fossa or the tear trough requiring undercorrection and others requiring slightly more volume, such as the mandibular angle or the jawline, rendering the 0.1 mL per injected bolus recommendation of limited practicality.

Furthermore, a previous study by Zhang and colleagues reported that the volumes of the supraorbital and supratrochlear arteries based on calculations of length and external diameters were 0.083 and 0.089 mL, respectively.²⁷ Three cases of IRVC were presented in their study, and it was suggested that the severity and the pattern of the clinical presentation depended on the vessel embolized (central retinal vs posterior ciliary vs ophthalmic artery) and that an amount as little as 0.08 mL of soft tissue filler was sufficient to cause an IRVC. It should be noted that the volume was calculated based on the external (not internal) diameter of the arteries, which was given as 1 mm for both vessels. This indicates that the true intraarterial volume of the vessels was most likely smaller than the presented 0.08 mL, which is also smaller than the minimal volume an injector can carefully administer during facial aesthetic procedures. Despite the limitations of the methodology applied it seems most likely that even smaller volumes than 0.08 mL can cause IRVCs and their catastrophic consequences for the patient. This is supported by the anatomic analysis of the facial arterial system, which has been shown to have a plethora of connections between the internal and the external carotid artery systems that can occur in any anatomic layer.^6,28 The arteries decrease in diameter to become a capillary bed for organ supply before they join to other arteries with a larger diameter, resulting in a large-small-large internal diameter pathway for the arterial blood stream. It is therefore plausible that even smaller amounts (smaller than 0.08 mL) of soft tissue filler could cause adverse vascular events, and recommendations for use of 0.1 mL or 0.2 mL might create a false sense of safety. It is hoped that future studies will focus on this knowledge gap and determine the smallest amount that can cause adverse vascular events. Until robust scientific evidence is available it has to be assumed that any amount greater than zero is sufficient to result in tissue loss or IRVCs.

Lastly, the recommendation for injecting small boluses with a volume of 0.1 mL to avoid severe vascular events affecting the ophthalmic artery circulation is principally based on the mechanism theorized to cause an IRVC. This theory assumes that the injected soft tissue filler material is inadvertently administered into the arterial blood stream and then transported to the ophthalmic artery and its many branches, causing a mechanical block of the arterial pathway resulting in reduced oxygenation of the supplied soft tissues; for the ophthalmic artery this is the highly sensitive retina, where reduced oxygenation results in IRVC.^29,30 Recent research has revealed that even if intraarterial therapy is performed, the recovery rate from soft tissue filler–induced IRVC is below 50%.^31,32 This unsatisfactory outcome despite such invasive therapies may indicate that the cause for IRVC or facial vascular adverse events is more complex than a mechanical obstruction of continuous arterial flow caused by an embolus of soft tissue filler material. Interestingly, Schelke and colleagues recently provided information on ultrasound-guided hyaluronidase injections for treating soft tissue filler–related vascular adverse events and found that in all of the investigated cases the soft tissue filler material, when visualized by real-time ultrasound, was located in the periarterial space and not intrarterially.^33,34 The authors went on to propose the concept of a perforasome, and that vascular adverse events might be caused by either compression or vasospasm rather than an obstructing soft tissue filler bolus. Although this novel idea is still a theory, if true it would imply that the theoretical mechanism behind suggesting safer bolus volumes during soft tissue filler injections based on the volume of the ophthalmic artery (0.1 mL by anatomic measurement or 0.2 mL by CT imaging) is of limited applicability. This would be supported by the fact that soft tissue filler boluses need to pass along their migratory path from larger-caliber vessels in the face to arterial connections with small diameters to enter the ophthalmic artery circulation.

This study however was not free of limitations: the study cohort was wholly comprised of Chinese patients, and therefore its findings may not be generalizable to other populations. However, to date no study has shown ethnic differences in the ophthalmic artery circulation, which might allow an assumption that the results are applicable to all ethnic groups. The correlation calculated between orbital length and arterial length revealed no statistically significant relationship (r_p = 0.076 with P = .504). This suggests that even if anthropometric data are variable as a result of ethnic influences (skull size and the respective diameters and distances), the length of the artery is not influenced by such bony alterations. Another limitation was the relatively mature study population, with a mean age of 61 years. Given the methodology applied (CT scanning) and the study design (retrospective CT data analysis), it was natural that CT scans were predominantly available from patients of an older age, with more medical conditions. No CT scans were conducted explicitly for the purposes of this study to prevent unnecessary radiation exposure in participants.

Changing clinical practice is a long journey and requires substantiated scientific evidence. Many more studies are needed at this point to increase safety in the aesthetic field, but we are confident that scientists and clinicians will continue to work together for the benefit of aesthetic patients.

CONCLUSIONS

Based on the results obtained from the investigation of 80 ophthalmic arteries it must be concluded that current safety recommendations should be reevaluated. The volume of the ophthalmic artery appears to be 0.2 mL rather than 0.1 mL as previously reported. In addition, it seems impractical to limit the volume of soft tissue filler bolus injections to 0.1 mL, due to the aesthetic requirements of each individual patient and treatment plan. Moreover, it is questionable whether a generalized limit for bolus injections (0.2 mL or 0.1 mL) can truly prevent IRVCs, given the emerging body of literature. It may be more likely that any volume greater than zero can cause severe adverse vascular events and that other preventive measurements might be of greater practicability and relevance, such as understanding 3D facial (vascular) anatomy, performing preinjection aspiration, or conducting ultrasound-based preinjection screening.

Acknowledgments

The authors would like to thank Dr Jeremy B Green of Skin Associates of South Florida, Skin Research Institute (Miami, FL) for his critical revision of this manuscript.

Disclosures

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

Funding

The authors received no financial support for the research, authorship, and publication of this article.

REFERENCES

Author notes