Comparison of Optical Coherence Tomography Angiography to Indocyanine Green Angiography and Slit Lamp Photography for Corneal Vascularization in an Animal Model

Tisha P Stanzel,Leopold Schmetterer,Nyein C Lwin,Gary H Yam,Marcus Ang,Kavya Devarajan,Jodhbir S Mehta

doi:10.1038/s41598-018-29752-5

Abstract

Corneal neovascularization (CoNV) could be treated by novel anti-angiogenic therapies, though reliable and objective imaging tools to evaluate corneal vasculature and treatment efficacy is still lacking. Optical coherence tomography angiography (OCTA) –currently designed as a retinal vascular imaging system— has been recently adapted for anterior-segment and showed good potential for successful imaging of CoNV. However, further development requires an animal model where parameters can be studied more carefully with histological comparison. Our study evaluated the OCTA in suture-induced CoNV in a rabbit model compared to indocyanine green angiography (ICGA) and slit-lamp photography (SLP). Overall vessel density measurements from OCTA showed good correlation with ICGA (0.957) and SLP (0.992). Vessels density by OCTA was higher than ICGA and SLP (mean = 20.77 ± 9.8%, 15.71 ± 6.28% and 17.55 ± 8.36%, respectively, P < 0.05). OCTA was able to depict CoNV similarly to SLP and ICGA, though it could better detect small vessels. Moreover, the depth and growth of vessels could be assessed using en-face and serial-scans. This study validated the OCTA in a rabbit model as a useful imaging tool for translational studies on CoNV. This may contribute to further studies on OCTA for anterior-segment including serial evaluation of emerging anti-angiogenic therapies.

Highlights

Photography for Corneal neovascularization (CoNV) are still under validation[11,12]
This proof of concept study was conducted following our previous work in humans, and in an effort to establish Optical coherence tomography (OCT) angiography (OCTA) as an imaging tool for animal models of CoNV for future translational studies
A total of 240 OCTA scans (5 scans/rabbit/week) were taken during the 8 weeks study period, and images with CoNV were analyzed from week 3 onwards where the peak of CoNV developed beyond 3 millimeters from the limbus into the cornea

Summary

Introduction

Photography for CoNV are still under validation[11,12]. Fluorescein and indocyanine green angiography techniques have demonstrated better vessel delineation than slit-lamp photograph despite the presences of corneal scars[13]. The videographic function showing blood flow and leakage is a useful tool to evaluate dynamic vessel properties[13] Both techniques are invasive and expose subjects to common gastrointestinal side effect or serious adverse reaction like anaphylactic shock[14,15,16]. Another limitation of the above procedures is that they measure the vascularized area only in two dimensions. We have previously described the use of OCTA for corneal vascularization in patients, demonstrating good repeatability and reproducibility[24,25] This technique showed good agreement with indocyanine green angiography (ICGA) for vessel delineation and could quantitatively monitor vascularized area after antiangiogenic treatment[26,27,28]. We assessed first time, the use of OCTA for depicting and measuring corneal vascularization in a rabbit model, as well as comparisons to ICGA and slit-lamp photography

Methods

Results

Conclusion