Projection-Resolved Optical Coherence Tomographic Angiography of the Macular Ganglion Cell Layer Plexus

Abstract

Purpose : To characterize the macular ganglion cell layer plexus (GCLP) boundaries using projection-resolved optical coherence tomographic angiography (PR-OCTA) in healthy eyes.

Methods : Participants were scanned using a commercial OCTA system (RTVue-XR Avanti; Optovue Inc, Fremont CA) in a 6×6-mm area centered on the foveal avascular zone. The split spectrum amplitude decorrelation angiography algorithm (SSADA) was applied to detect flow signal. The GCLP anterior boundary was marked at the nerve fiber layer (NFL)-GCL junction. PR-OCTA algorithm was used to remove flow projection artifacts. Ganglion cell and inner plexiform layer (GCIPL) was divided into 20 equal slabs. In each slab, vessel density (VD) in each polar coordinate sector (Figure 1) were measured using a custom software with automatic shadow exclusion and reflectance compensation. Fifth-degree polynomial fit was used to analyze the correlation between VD and depth in the GCIPL and estimate the boundary between GCPL and intermediate capillary plexus.

Results : 38 normal participants (78.9% female) were enrolled, and one eye in each participant was studied. Mean age and standard deviation was 59.6±10.7. The watershed (depth of minimum VD) between the GCLP and the intermediate capillary plexus (ICP) is located at 75% depth within the GCIPL (Figure 2) throughout the macula. GCLP VD was significantly (p<0.0001) correlated with GCIPL and macular ganglion cell complex (GCC) thickness (r=0.443, and r=0.857, respectively). The correlation was significantly stronger for macular GCC compared to GCIPL (z=-3.3, p<0.001).

Conclusions : Macular GCLP supplies the anterior 75% of the GCIPL. Its density is better correlated with GCC, which also contain the NFL, than with the GCIPL, suggesting that it also supplies the posterior aspect of the NFL. Mapping the macular GCLP may be useful in evaluating ganglion cell perfusion in glaucoma and optic neuropathies.

This is a 2021 ARVO Annual Meeting abstract.

 

Figure 1. Polar coordinate sectors that were used to estimate the boundary between ganglion cell layer plexus and intermediate capillary plexus. N=nasal, S=superior, T=temporal, I=inferior

 

Figure 2. Scatter plot of vessel density and depth in each polar sector. Fifth degree polynomial model showing the correlation between vessel density and depth in ganglion cell inner plexiform layer. Depth is position of the normalized ganglion cell inner plexiform layer thickness.