Association of Change in Iris Vessel Density in Optical Coherence Tomography Angiography With Anterior Segment Ischemia After Strabismus Surgery

This case series study analyzes the use of iris optic coherence tomography angiography (OCT-A) for patients undergoing strabismus surgery at risk for anterior segment ischemia.

Key Points

Question

Is optical coherence tomography angiography useful in patients undergoing strabismus surgery to determine whether anterior segment ischemia is present?

Findings

In this case series study of 9 participants, the mean vessel density for all quadrants had a relatively small decrease from preoperative to postoperative. Optical coherence tomography angiography detected filling defects in the quadrant adjacent to the operated muscle in 1 patient.

Meaning

The clinical relevance of these findings is unknown, and given the small sample size, further research to determine its use in clinical practice seems warranted.

Abstract

Importance

Anterior segment ischemia (ASI) is a rare but potentially serious complication of strabismus surgery. Indocyanine green angiography and fluorescein angiography have been used to reveal iris-filling defects for clinicians considering a patient’s risk of ASI. However, both are limited by invasive and time-consuming nature and potential adverse effects. Recently, optic coherence tomography angiography (OCT-A) has been introduced and used to image iris vasculature in individuals without abnormalities.

Objective

To determine the use of iris OCT-A for patients undergoing strabismus surgery and who are at risk for ASI.

Design, Setting, and Participants

This prospective case series study took place in an academic center. Adults undergoing strabismus surgery on at least 1 vertical muscle were prospectively recruited. The study took place from June to November 2017, and analysis began in January 2018.

Interventions

Indocyanine green angiography and OCT-A of the iris preoperatively and 1 day postoperatively.

Main Outcomes and Measures

A masked examiner evaluated all images and determined whether any filling defects were present qualitatively (lack of perfusion) and quantitatively (for OCT-A using internal software to calculate vessel density).

Results

Ten eyes of 9 individuals (mean [SD] age, 63 [11] years) were included. Two individuals (22.2%) identified as Hispanic, and 7 (77.8%) identified as white. There were 6 women (66.7%). The mean preoperative vessel density (percentage of the area occupied by vessels) averaged for all quadrants decreased from 57% preoperatively to 55% postoperatively (mean difference, 2%; 95% CI, 0.4%-4.2%; P = .05). When comparing quadrants adjacent to operated muscles, the mean vessel density decreased from 56% to 53% (mean difference, 2.6%; 95% CI, 0.17%-4.8%; P = .02). In addition, OCT-A detected vascular filling defects in the quadrant adjacent to the operated muscle on the patients in whom they were present (n = 1, inferior rectus recession).

Conclusions and Relevance

In this preliminary study, OCT-A determined iris vessel filling defects when present. In addition, OCT-A gives qualitative vessel density values that can be compared preoperatively and postoperatively although the clinical relevance of small differences is not known. While only 10 eyes were evaluated, and as such generalizability of these findings cannot be determined, the results suggest that OCT-A may be a useful tool in the evaluation of patients undergoing strabismus surgery to determine whether a patient is at risk to develop ASI.

Introduction

Anterior segment ischemia (ASI) is a serious but rare complication of strabismus surgery. Anterior segment ischemia occurs after strabismus surgery when the anterior segment blood supply is disrupted; the anterior segment blood supply is principally made up of 2 anterior ciliary arteries associated with each of the 4 rectus muscles, except for the lateral rectus muscle, which only has 1. The anterior ciliary vessels supply several vascular plexuses, including the episcleral limbal plexus, the intramuscular circulation within the ciliary body, and the major arterial circle in the iris root.^1,2 When the rectus muscles are disinserted from the sclera, the overlying anterior ciliary vessels are interrupted, leading to a decrease in the blood supply to the various anterior segment vascular plexuses that they support. Clinically, this decrease in blood flow results in a delay or absence of iris-vessel filling in the quadrant corresponding to the tenotomized muscle when imaged using iris fluorescein angiography and indocyanine green angiography (ICGA).³ However, fluorescein angiography and ICGA of the iris are limited by the necessity to image only 1 eye at a time and to acquire early and late images and by the invasive nature of the injection and the potential adverse effects of the contrast media. Optic coherence tomography angiography (OCT-A) has been introduced and used to image the iris vasculature in individuals without abnormalities and in eyes with iris tumors.^4,5 Optic coherence tomography angiography may be used to evaluate patients undergoing strabismus surgery to rule out ASI. Herein, we report a method of diagnosing ASI with the use of OCT-A and describe our initial findings.

Methods

This study was approved by the University of California, Los Angeles institutional review board and conformed to the requirements of the US Health Insurance Portability and Accountability Act of 1996. All patients gave written informed consent prior to participating. Patients older than 40 years scheduled to undergo strabismus surgery involving at least 2 rectus muscles were recruited from a clinic. Patients undergoing surgery on 1 rectus muscle were included if they had previous surgery on a separate rectus muscle in the same eye. Data collected included age, history of ocular or strabismus surgery, and comorbid conditions.

Initially, all patients underwent preoperative iris ICGA and OCT-A 1 to 7 days prior to surgery, followed by a postoperative iris ICGA and OCT-A within 3 days after surgery. However, after the first 4 patients, we determined that patients were reluctant to undergo an injection of dye especially on their postoperative visit. Therefore, we enrolled 5 additional patients who underwent OCT-A alone 1 to 7 days prior to surgery, followed by a postoperative OCT-A within 3 days of surgery. The additional patients also provided written informed consent. Iris ICGA and iris OCT-A were analyzed for filling defects by a masked examiner (S.L.P.).

Optic coherence tomography angiography images of the iris were obtained using a spectral-domain OCT device (RTVue-XR Avanti; Optovue, Inc) that operated with a central wavelength of 840 nm, an acquisition speed of 70 000 A-scans per second, and a bandwidth of 45 nm. Image acquisition was as previously described.⁶ Image acquisition of the iris was obtained using an anterior segment lens and a manual focusing adjustment in the iris using the retina imaging software. Each cube consisted of 2 repeated volumes of 304 B-scans, each containing 304 A-scans. Split spectrum amplitude decorrelation technology was used to improve the signal-to-noise ratio by splitting the spectrum to generate multiple repeat OCT frames from 2 original repeat OCT frames. A 3 × 3 mm and a 6 × 6 mm scan area was chosen to fully capture the iris vasculature in the superior, nasal, inferior, and temporal quadrants. A full frontal 8 × 8 mm iris scan was also performed. Internal projection artifact software was used to reduce motion artifacts. Optic coherence tomography angiography scans were captured in both eyes and always by the same masked operator (J.P.D.). Total imaging time for OCT-A of the iris typically was 3 minutes per eye. Automated vascular density software analysis, which included manual segmentation in retina imaging mode and commercially available vascular density software analysis, was used to calculate vessel density for each quadrant preoperatively and postoperatively. The vascular density analysis portion of the examination took approximately 5 minutes per patient. The images were then randomly assigned a study identification number by J.P.D. and placed into files to be reviewed by the masked examiner (S.L.P.) who was not involved in the care of any of the study patients.

Statistical Analysis

All data were analyzed using JMP Pro, version 12 (SAS Institute). Descriptive statistics were used to summarize patient characteristics and vessel density. The percentage of vessel density was compared among groups; vessel density was obtained using the Optovue density software analysis and represents the percentage of the area occupied by vessels (arteries and veins). Means were compared using a 2-sided paired t test.

Results

Ten eyes of 9 patients (mean [SD] age, 63 [11] years) were included. Of the 9 patients, 2 (22.2%) identified as Hispanic, and 7 (77.8%) identified as white. There were 6 women (66.7%) and 3 men (33.3%). Every patient underwent surgery on at least 2 muscles in the study eye (either at the time of the surgery or combined with previous surgeries) using either recession, resection, plication, or partial tenotomy procedures. Four patients (44.4%) had previous strabismus surgery. Three additional patients (33.3%) had initially consented to take part in the study but were later excluded owing to poor-quality images, extremely dark iris pigmentation, or inability to maintain fixation for the study. Surgical procedures are summarized in the Table.

Table. Patient Characteristics.

Patient	Age, y	Eye Analyzed	Previous Surgeries	Surgical Procedurea
1	83	Left	None	MR and IR full tenotomy
2	53	Left	None	LR full tenotomy; IR full plication
3	66	Left	None	SR full tenotomy; IR partial plication
4	60	Left	None	MR and SR full tenotomy
5	46	Right	LR full tenotomy	IR full tenotomy
6	68	Right	None	MR and LR full tenotomy
7	69	Left	None	MR full tenotomy; IR partial tenotomy
8	55	Left	MR tenotomy	MR and SR full tenotomy
		Right	MR tenotomy	LR full tenotomy
9	74	Right	IR full tenotomy	IR full tenotomy; SR plication

Abbreviations: IR, inferior rectus; LR, lateral rectus; MR, medial rectus; SR, superior rectus.

^aFull tenotomy indicates either full recession or resection.

Figure 1 illustrates an example of a preoperative iris ICGA and OCT-A for patient 3. All 4 of the patients who underwent ICGA and OCT-A had similar appearances with regard to the comparability of the 2 modalities.

Figure 1. Preoperative Iris Scans.

Preoperative iris indocyanine green angiography (ICGA) (A) and optical coherence tomography angiography (OCT-A) (B) for a patient undergoing strabismus surgery. C, The OCT-A image is divided into quadrants, and vessel density is calculated for each quadrant.

The mean preoperative vessel density for all patients averaged for all quadrants decreased from 57% (preoperatively) to 55% (postoperatively) (mean difference, 2%; 95% CI of mean difference, 0.4%-4.2%; P = .05). When comparing quadrants adjacent with operated muscles, the mean vessel density decreased from 56% to 53% (mean difference, 2.6%; 95% CI of mean difference, 0.17%-4.8%; P = .02). In addition, OCT-A detected a large vascular filling defect in the inferior quadrant adjacent to the operated muscle (inferior rectus recession) in 1 patient (Figure 2).

Figure 2. Scans of Eye Before and After Inferior Rectus and Lateral Rectus Muscle Recession.

Preoperative (A) and postoperative (B) iris optical coherence tomography angiography (OCT-A) for a patient undergoing inferior rectus and lateral rectus muscle recession. The postoperative image reveals an iris filling defect (arrowhead) that is not present on the preoperative image.

Discussion

The results of this study suggest that OCT-A may be a substitute for iris ICGA without the invasive need for dye injection, although we did not have a large number of cases with which we could make direct comparison. In the 4 cases who underwent imaging with both modalities, the appearances were remarkably similar (Figure 1). Moreover, OCT-A provided quantitative iris vascular density analysis for comparison between the preoperative and postoperative eye and even identified in 1 patient postoperative quadrantic iris vascular filing defects, adjacent to the muscle surgery (Figure 2). This patient underwent lateral rectus recession combined with an inferior rectus plication; this combination of surgery has previously been shown by our group to result in temporary iris filling defects.⁷ The use of OCT-A to detect iris vessel filling defects may allow surgeons to plan vessel-sparing procedures on patients found to have preoperative abnormalities of anterior segment perfusion. Furthermore, in postoperative cases, it may assist surgeons in diagnosing ASI before more dangerous symptoms become manifest.

Despite the lack of patients with obvious qualitative filling defects on OCT-A, this technique did illustrate a difference in the overall vascular filling density in the postoperative group of patients vs the preoperative group. Although the small difference may not be clinically significant, it shows the sensitivity of OCT-A in identifying small changes in postoperative iris vessel density. This finding may be important in validating the preoperative vs postoperative outcomes. The value of comparing preoperative vs postoperative vessel density has not yet been defined; however, this technique may be useful in the diagnosis of ASI, especially in high-risk patients. We purposely selected higher-risk patients who were older and who underwent surgical procedures on at least 2 rectus muscles (either at the time of the surgery or combined with previous surgeries). Therefore, the use of OCT-A iris vessel density analysis in lower-risk patients requires further validation.

Limitations

The results of this study must be understood within the context of its limitations. The main limitation of our study is the small sample size. Given the small number of patients, the spectrum of surgical techniques, patient characteristics, and severity of iris vascular abnormalities was limited. Additionally, none of the patients had been clinically diagnosed as having ASI, and we therefore cannot state with certainty that OCT-A would be comparable with ICGA in those cases. Lastly, we do not know the reliability and reproducibility of our masked examiner or any other type of examiner of iris OCT-A, and therefore, we cannot evaluate reproducibility of our results.

Conclusions

This study represents a description of the use of OCT-A to evaluate strabismus surgery patients for ASI. We have demonstrated that OCT-A may be a useful tool in the evaluation of these patients and to determine whether a patient is at risk of developing ASI, which is crucial in protecting patients from vision-threatening complications after strabismus surgery. Further research seems warranted to determine the ability of OCT-A to diagnose ischemia of various levels of severity.