We evaluated the effect of acute mild light exposure at night on sleep architecture and glucose homeostasis.Twenty healthy normal-weight men took part in two conditions of a randomized, controlled, balanced cross-over experimental study: i) two-consecutive nights with 8-h of sleep under dLAN (5 lux) or ii) total darkness (CON). Sleep was evaluated by polysomnography. In the morning following 'night2', glucose homeostasis was assessed by an intravenous glucose tolerance test (ivGTT) with consecutive measures of glucose, insulin, and c-peptide. Plasma cortisol was measured at night before sleep, after morning awakening, and during mid-afternoon hours.There was no significant difference in total sleep time, sleep efficiency, and sleep latency between conditions (all p 0.66). However, NREM sleep stage N3 latency was prolonged after dLAN (p = 0.02) and NREM sleep stage 2 was decreased after two nights with dLAN (p = 0.04). During the first sleep hour, power in slow-oscillations, slow-waves, and delta bands diminished after dLAN (all p 0.04). Glucose, insulin, and c-peptide were not altered by dLAN (all p 0.14). Cortisol was reduced in the afternoon after 'night1' and in the morning after 'night2' (both p 0.03).dLAN slightly disturbed sleep architecture and quality without impairment of glucose homeostasis. Longer exposure to chronic dLAN might be needed to unmask its hypothesized metabolic consequences.
