Abstract
Purpose :
Previous reports of intensity-based optoretinography (iORGs) require dark- adaptation before acquisition, adding additional time per trial and reducing the assay’s clinical utility. Here we assess the feasibility and repeatability of a light-adapted iORG.
Methods :
Three subjects with normal vision were dilated and imaged over two separate experiments using a custom adaptive optics scanning laser ophthalmoscope. In each experiment, three trials of 10 acquisitions each were collected 2° temporal from the fovea. Experiment 1 consisted of 3min of dark adaptation preceding each trial and experiment 2 consisted of 3min of light adaptation in office lighting. Experiments were conducted at least one month apart. Individual acquisitions consisted of 2s of prestimulus recording, a 68ms 17.6μW/° stimulus (λ=554nm; Δλ=±45nm), and 4s of post-stimulus recording. Acquisitions were co- registered via custom software. Cone locations were found semi-automatically using an average image that was generated from co-aligned videos. Population iORGs were extracted using a previously described approach [PMC7771891], resulting in a single root mean square (RMS) signal per subject per iORG acquisition. For each RMS signal, we extracted amplitude, defined as the difference between the pre and post stimulus RMS. We assessed the differences between subjects, adaptation state, and amplitudes using a two-way ANOVA. To assess repeatability of each condition, we compared the geometric coefficient of variance (GCV) of the light- and dark-adapted amplitudes.
Results :
We obtained iORGs from a total of 8637 light- and 6925 dark- adapted conesbacross all three subjects. For all subjects, light-adapted amplitudes were significantly decreased from dark-adapted amplitudes (0.57±0.1 and 0.4±0.06, respectively; p<0.001), but did not decrease by the same amount (p<0.001). The cross-subject GCV of the light- and dark- adapted amplitudes was 0.37 and 0.43, respectively.
Conclusions :
There was a significant difference between RMS amplitudes in light vs dark adapted iORGs, but similar repeatability. Our results demonstrate that light adapted iORGs could be used as a similarly representative measure of cone function in place of dark adapted iORGs. To establish the clinical utility of this approach, future studies should assess other metrics and validate this relationship in individual cone iORGs and retinal pathology.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.