0842 Feasibility of daytime bright light exposure and urinary 6-sulfatoxymelatonin profiling in the medical ICU: A pilot study

Introduction Bright daytime light, dim evening light, and near darkness during sleep are optimal lighting conditions to support circadian physiology, wakefulness, and sleep. Such light patterns are absent in the medical intensive care unit (MICU) and are considered one factor contributing to the delayed circadian timing observed in MICU patients. We, therefore, conducted a pilot trial assessing the use of daytime bright light (DBL) in the MICU and tested the feasibility of using urinary 6-sulfatoxymelatonin (aMT6s) as a marker of circadian phase and amplitude. Methods We enrolled patients aged >50 years admitted to the MICU without pre-existing circadian disruption. Urine samples were collected every 4 hours starting on study day 1 (D1). Patients were randomized to receive bright light (>1200lux at the angle of gaze, 5000K) for 4 hours or 8 hours starting at 09:00 on D2. Circadian metrics of aMT6s amplitude and acrophase were calculated via gamma distribution cosinor analysis. Results Eight (6 males, 2 females) out of 14 pilot participants produced sufficient urine for sampling; one was subsequently excluded due to exogenous melatonin use. Of the seven patients in the analysis, one did not have significant cosinor fit. Mean (SD) age was 74 (9) years and APACHE II illness severity score was 24 (4). D1 median (IQR) aMT6s amplitude was 29,389ng (447-125,874ng) and acrophase was 12:42 (8:30-13:54). D2 (after DBL start), median amplitude was 847,381ng (372-1,694,391ng) and acrophase was 04:36 (03:06-06:06). Six patients received 4-hours and two patients received 8-hours of DBL. Sample size was inadequate for testing differences in circadian outcomes between light assignments and study days. Conclusion DBL may promote earlier circadian timing and greater circadian amplitude in critically ill patients. Findings support feasibility of aMT6s in assessing circadian phase and amplitude in MICU patients who produce urine; however, a significant proportion of MICU patients are anuric. Consequently, sample size limitations, potential for masking, and infrequent urine sampling warrant further investigation. These pilot findings have influenced the frequency of aMT6s sampling (e.g., change to hourly) in an ongoing RCT evaluating the impact of DBL versus usual light exposure on circadian rhythms in critical illness. Support (if any) K23HL138229, AASM Foundation