Abstract
Purpose :
Capillary perfusion is spatially and temporally heterogeneous in the retina. We introduce a serial-acquisition imaging method, where sequentially acquired OCT angiography (OCTA) scans are processed and analyzed for potential correlation with a 670nm red light exposure stimulus. The stimulus was chosen for its known impact on cone photoreceptor metabolism, which we hypothesize may result in microvasculature changes detectable with OCTA.
Methods :
We investigated the ability of an OCTA device (Zeiss PLEX® Elite 9000) to detect retinal microvasculature responses to 670nm red light stimulus. Seven healthy human volunteers were exposed to a red light (670nm) shone directly into one eye for 3 minutes at energy levels one log unit greater than that found in environmental light, (with UBC REB approval). One OCTA imaging session was conducted before stimulus, and multiple imaging sessions were conducted after stimulus (0 min, 5 min, 15 min, 60 min). In each session, five 3x3mm OCTA scans centered on the fovea were acquired, registered and segmented, and then the pixel-intensity coefficient of variation (PICoV) and perfusion density (PD) were calculated for each imaging session and vascular layer (superficial and deep; SVC and DVC).
Results :
The image acquisition and processing pipeline is shown in Fig. 1, revealing a dynamic retinal microvasculature where OCTA signal variations are quantified. A significant inverse relationship was found between PICoV and PD for both the SVC (p<0.001) and DVC (p<0.05). However, changes in either image metric were not correlated with time from stimulus onset, nor were layer-specific temporal trends observed.
Conclusions :
This study found an inverse relationship between the perfusion density of a retinal capillary bed and its temporal variance in healthy eyes. However, no relationships between image metrics and red-light stimulus were found, nor were there layer-specific trends in the stimulus response. Further work should be done to expand this analysis on human subjects with retinal diseases or metabolic deficiencies.
This abstract was presented at the 2023 ARVO Imaging in the Eye Conference, held in New Orleans, LA, April 21-22, 2023.