0196 Retinal responsivity is associated with circadian phase and circadian alignment but not sleep timing

Abstract

Abstract Introduction Light entrains the central circadian clock, with projections from the retina to the SCN through melanopsin-containing retinal ganglion cells (ipRGCs). Altered responsivity to light reflects ipRGC functioning and may be a physiological vulnerability for disrupted photoentrainment. Understanding how retinal responsivity relates to sleep and circadian timing may inform who might most benefit from sleep and circadian interventions. Methods 64 participants (85 observations) ages 20-66 years old were recruited during winter (n=35) and summer months (n=50), and included individuals with seasonal depression (n=33) and nonseasonal, never depressed controls (n=31). The post-illumination pupil response (PIPR) to red and blue light was used to measure the responsivity of ipRGCs (average 1:30pm; 10am-7pm). Circadian phase was assessed using Dim Light Melatonin Onset (DLMO), collected every 30-minutes on Friday evenings. Midsleep timing was measured using actigraphy (average number nights=4), and circadian alignment was calculated as the DLMO-midsleep phase angle. We performed a multilevel regression to determine the relationship between PIPR and markers of sleep and circadian timing, accounting for repeated seasonal assessments with a random intercept of participant. Covariates included age, gender, diagnostic group, and circadian time of PIPR assessments. We ran sensitivity analyses including photoperiod length on the day of PIPR assessment to account for potential light exposure on the PIPR. Results Greater retinal responsivity was associated with later DLMO (b=4.45; partially standardized b=0.28; SE=1.84; p=0.03), and shorter DLMO-midsleep phase angle (b= -7.33; partially standardized b= -0.32; SE=2.51; p=0.004), but not midsleep (b= -3.44; partially standardized b= -0.14; SE=2.35; p>0.05). Individuals with later DLMO had PIPR assessments at earlier circadian times (b= -0.12; SE=0.04; p=0.01). Older participants (b= -0.04; SE=0.02; p=0.04) and controls (b= -0.95; SE=0.44; p=0.04) had earlier sleep midpoints, but covariates were not associated with circadian markers. The association between circadian timing and PIPR became nonsignificant (b=4.15; partially standardized b=0.26; SE=1.88; p=0.06) when including photoperiod. Conclusion Retinal responsivity was associated with circadian but not behavioral sleep timing, suggesting ipRGC functioning may have downstream effects on circadian entrainment. Assessing circadian variation of retinal responsivity remains a crucial next step prior to testing whether retinal responsivity impacts response to circadian-focused interventions. Support (If Any) NIMH K.A.R.MH103303