Abstract
Purpose :
Optical Coherence Tomography Angiography (OCTA) is widely used to image retinal and choroidal vasculature in ophthalmology. To interpret clinical data, it is essential to understand the conditions under which a microvessel can be detected by OCTA. Several methods have been applied in OCTA validation studies, each of which has limitations. Fluorescence angiography enables visualization of perfusion, but does not provide depth-resolved information. This limitation can be overcome with adaptive optics, which in turn limits the field-of-view and spatiotemporal co-registration with OCTA. Endothelial labeling and histology remain the gold standard but are subject to distortions during tissue processing and provide no information about perfusion. A depth-resolved, co-registered, and in vivo validation of OCTA is therefore needed. In this study, we use a highly scattering plasma tracer to assess the ability of OCTA to visualize microvessels depending on their angular orientation.
Methods :
Rats were imaged under isoflurane anesthesia, using a custom-built OCT ophthalmoscope operating at 1300 nm. Intralipid-20%, an FDA-approved fat emulsion dietary supplement that acts as a plasma tracer, was injected intravenously (3 mL/kg body weight). OCTA volumes were acquired with 1500 axial scans at 1500 y-locations with 3 repeats per y-location, over a 2.9 mm (x) by 2.9 mm (y) by 1.9 mm (z) range, before and after injection. Microvessel visibility was assessed across OCTA volumes.
Results :
Figure 1A shows the angiogram after tracer injection, which is assumed to visualize all plasma-perfused microvessels. To assess the ability of OCTA to visualize microvessels with different orientations, polar angle was color-coded from 0 degrees (parallel to the optic axis) to 90 degrees (en face) (Figure 1B-C). Thus retinal vessels mainly lie in en face laminae (red), while oblique vessels connect laminae. Clearly, more oblique microvessels are visualized after tracer injection.
Conclusions :
Our results suggest that the ability of OCTA to visualize a microvessel depends on its orientation with respect to the OCT beam. Specifically, we find that vessels that do not lie in en face laminae are more likely to fall below the OCTA detection limit. While orientation dependence of erythrocyte backscattering likely contributes to the observed findings, we cannot exclude the possibility that hematocrit differences may also contribute.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.