Human Adolescent Phase Response Curves to Bright White Light

Abstract

Older adolescents are particularly vulnerable to circadian misalignment and sleep restriction, primarily due to early school start times. Light can shift the circadian system and could help attenuate circadian misalignment; however, a phase response curve (PRC) to determine the optimal time for receiving light and avoiding light is not available for adolescents. We constructed light PRCs for late pubertal to postpubertal adolescents aged 14 to 17 years. Participants completed 2 counterbalanced 5-day laboratory sessions after 8 or 9 days of scheduled sleep at home. Each session included phase assessments to measure the dim light melatonin onset (DLMO) before and after 3 days of free-running through an ultradian light-dark (wake-sleep) cycle (2 h dim [~20 lux] light, 2 h dark). In one session, intermittent bright white light (~5000 lux; four 20-min exposures) was alternated with 10 min of dim room light once per day for 3 consecutive days. The time of light varied among participants to cover the 24-h day. For each individual, the phase shift to bright light was corrected for the free-run derived from the other laboratory session with no bright light. One PRC showed phase shifts in response to light start time relative to the DLMO and another relative to home sleep. Phase delay shifts occurred around the hours corresponding to home bedtime. Phase advances occurred during the hours surrounding wake time and later in the afternoon. The transition from delays to advances occurred at the midpoint of home sleep. The adolescent PRCs presented here provide a valuable tool to time bright light in adolescents.

Keywords: adolescent; bright light; circadian; phase response curve; phase shift; school start time; sleep.

Figure 1

Protocol for the two 5-day laboratory sessions. After 8 or 9 days of sleeping on a fixed 9-h sleep schedule at home, participants completed this protocol in the laboratory. On days 1 and 5, participants completed a circadian phase assessment in dim light (< 5 lux) to determine the salivary dim light melatonin onset (DLMO). Baseline saliva sampling began at 15:30 and ended at 03:30. Final saliva sampling began at 14:30 and ended at 08:00 the next day. An ultradian light-dark (LD) cycle occurred in the intervening 3.4 days; participants were exposed to a 4-h LD cycle with a 2-h sleep opportunity in the dark alternating with 2 h of required wake in room light (~20 lux). Participants completed this protocol of home sleep followed by the 5-day laboratory session twice in a counterbalanced order. During one laboratory session, participants received 80 mins (four 20-minute exposures) of bright white light (~5000 lux) during one of the 2-h wake periods on days 2, 3, and 4. During the other laboratory session, participants remained in room light.

Figure 2

Spectral power distribution for the fluorescent lamps (Sylvania Octron 5000 K FB031/50K 31 W) contained within the bright light boxes used to construct the phase response curves to bright light in the current sample of adolescents.

Figure 3

Melatonin profiles. Baseline and final melatonin profiles from a 16-year old male participant during the room light condition (top) and the bright light condition (bottom). The solid horizontal line illustrates the 4 pg/mL threshold to determine the time of the DLMO. This participant was in a group in which bright light exposure (gray box and vertical dotted lines) started at 02:05 and ended at 03:55.

Figure 4

Adolescent phase response curve to bright light plotted relative to each individual’s circadian phase marked by the dim light melatonin onset (DLMO; upward facing arrow at 0 h). Bright light (~5000 lux) was produced with a single light box. An intermittent bright light pattern (four 20-min bright light exposures with 10 mins of dim room light (~20 lux) in between) was repeated at the same clock time for three consecutive days while participants lived on an ultradian sleep/wake (2 h dark/2 h light) schedule. DLMOs were measured before and after the three days of ultradian sleep/wake schedules and bright light exposures, from which the DLMO phase shift size (in h) and direction (advance or delay) was computed for each individual. Participants completed the same 5-day laboratory session without bright light (in a counterbalanced order) to measure the natural drift of the system due to the protocol. This shift was subtracted from the phase shift during the bright light session to isolate the shift due to bright light alone (net phase shift). Each point represents an individual’s net phase shift (y-axis) in response to when bright light exposures started relative to their DLMO (x-axis). Open symbols represent the three participants for whom a different DLMO threshold was used (10 pg/mL for two high melatonin secretors and 1 pg/mL for one low melatonin secretor; see text). A 2-harmonic curve was fit to the data. Phase shifts that occurred 21 to 24 h after baseline DLMO are double plotted.

Figure 5

Adolescent phase response curve to bright light plotted relative to the midpoint of the fixed 9-h home sleep schedule participants followed before each 5-day laboratory session. The sleep schedule was tailored to each participant’s current summertime sleep schedule and is illustrated with vertical lines. Bright light (~5000 lux) was produced with a single light box. An intermittent bright light pattern (four 20-min bright light exposures with 10 mins of dim room light (~20 lux) in between) was repeated at the same clock time for three consecutive days while participants lived on an ultradian sleep/wake (2 h dark/2 h light) schedule. DLMOs were measured before and after the three days of ultradian sleep/wake schedules and bright light exposures, from which the DLMO phase shift size (in h) and direction (advance or delay) was computed for each individual. Participants completed the same 5-day laboratory session without bright light (in a counterbalanced order) to measure the natural drift of the system due to the protocol. This shift was subtracted from the phase shift during the bright light session to isolate the shift due to bright light alone (net phase shift). Each point represents an individual’s net phase shift (y-axis) in response to when bright light exposures started relative to their schedule midpoint of sleep time at home (x-axis). Open symbols represent the three participants for whom a different DLMO threshold was used (10 pg/mL for two high melatonin secretors and 1 pg/mL for one low melatonin secretor; see text). A 2-harmonic curve was fit to the data. Phase shifts that occurred 12 to 18 h after home midsleep time are double plotted.