0024 Measuring Melanopsin-driven Retinal Responses in Sleep and Circadian Health: Performance of the Pupil Light Reflex

Abstract Introduction Alterations in the melanopsin-driven post-illumination pupil response (PIPR) may reflect a physiological vulnerability for circadian mood disruptions. However, the PIPR is confounded by other factors affecting retinal irradiance, like how much light reaches the retina. The pupil light reflex (PLR) measures effects on pupil responses that are nonspecific to melanopsin. Controlling for nonspecific factors related to retinal irradiance (i.e., age, entrance pupil diameter, iris pigmentation) is necessary to isolate melanopsin-driven retinal response in PIPR studies. The present study re-analyzed data using group, season, and a group-by-season interaction to predict the PIPR with different covariates testing nonspecific effects on the PIPR. We hypothesized that the PLR alone would explain more variance in the PIPR then the above set of indirect measures of nonspecific effects. Methods Participants ages 19-65 (M=34.9, SD=10.6) with seasonal depression (n=25) and non-depressed controls (n=22) completed assessments in the winter and summer months. The PLR was measured as the difference between baseline pupil diameter and the acute minimum constriction in millimeters and expressed as a percent of baseline. The Net PIPR was calculated as the difference between the red and blue trials averaged 10-40 seconds post-light stimulus. Iris pigmentation was rated using the Franssen (2008) scale. Model performance for different covariate sets was assessed using Akaike Information Criteria (AIC). All models included time of the PIPR relative to midsleep. Results Significant associations between group, season, and a group*season interaction in predicting the PIPR were found in every model. Model fit varied based on covariates included. The PLR following red light performed best in models predicting the PIPR (AIC=-161.63) compared to a model including age, entrance pupil diameter, and iris pigmentation (AIC=-144.98). The next best fit model included the PLR in response to blue light (AIC=-157.57). Conclusion Studies aiming to isolate melanopsin-driven retinal responsivity should adopt the PLR as an individually specific, objective, and contemporaneous measure of nonspecific influences of light on pupil responsivity. Methodology separating the effects of light incident on the retina (e.g., PLR) from melanopsin-driven retinal responses (e.g., PIPR) can inform treatments like bright light therapy, allowing titration of light intensity to account for individual differences in retinal irradiance. Support (if any)