Partial sleep deprivation reduces phase advances to light in humans

Abstract

Partial sleep deprivation is increasingly common in modern society. This study examined for the first time if partial sleep deprivation alters circadian phase shifts to bright light in humans. Thirteen young healthy subjects participated in a repeated-measures counterbalanced design with 2 conditions. Each condition had baseline sleep, a dim-light circadian phase assessment, a 3-day phase-advancing protocol with morning bright light, then another phase assessment. In one condition (no sleep deprivation), subjects had an 8-h sleep opportunity per night during the advancing protocol. In the other condition (partial sleep deprivation), subjects were kept awake for 4 h in near darkness (<0.25 lux), immediately followed by a 4-h sleep opportunity per night during the advancing protocol. The morning bright light stimulus was four 30-min pulses of bright light (~5000 lux), separated by 30-min intervals of room light. The light always began at the same circadian phase, 8 h after the baseline dim-light melatonin onset (DLMO). The average phase advance without sleep deprivation was 1.8 ± 0.6 (SD) h, which reduced to 1.4 ± 0.6 h with partial sleep deprivation (p < 0.05). Ten of the 13 subjects showed reductions in phase advances with partial sleep deprivation, ranging from 0.2 to 1.2 h. These results indicate that short-term partial sleep deprivation can moderately reduce circadian phase shifts to bright light in humans. This may have significant implications for the sleep-deprived general population and for the bright light treatment of circadian rhythm sleep disorders such as delayed sleep phase disorder.

Figure 1

A sample protocol for a subject who typically slept from midnight to 0800 h. There were 6 days of baseline sleep at home (days 1–6), a baseline phase assessment in the laboratory (day 7), a further week of baseline sleep at home (days 8–14), followed by a 3-day phase-advancing protocol in the laboratory (days 15–17) and then a final phase assessment (day 18). After a 9-day break, this sequence was repeated. In the no sleep deprivation condition (top), there was an 8-h sleep opportunity per night in the phase-advancing protocol. In the partial sleep deprivation condition (bottom), there was 4 h of enforced wakefulness in near darkness, followed by a 4-h sleep opportunity per night in the phase-advancing protocol. Thus, the light-dark cycle was similar in both conditions, but the duration of the sleep opportunity varied. The 2 conditions were counterbalanced across subjects. Black shading = sleep/dark episodes at night. * = at least 10 min of morning outdoor light. ↑= time of dim light melatonin onset. L = advancing bright light stimulus: four 30-min bright light pulses alternating with room light, starting 8 h after the baseline dim light melatonin onset, and advancing by 1 h on the second and third mornings. For clarity, the phase assessments are shown as starting and ending at 1200 h.

Figure 2

The average difference from baseline in ratings of sleepiness on the Stanford Sleepiness Scale and number of lapses and mean reaction time on the Psychomotor Vigilance Test when subjects had an 8-h sleep opportunity (no sleep deprivation condition) and a 4-h sleep opportunity per night (partial sleep deprivation condition). Subjects reported higher levels of sleepiness and had more lapses and slower reaction times in the partial sleep deprivation condition (*p < 0.02 on all paired t tests). Error bars represent SEMs.

Figure 3

The phase advances observed in each individual subject who underwent a 3-day phase-advancing protocol with either an 8-h (no sleep deprivation condition) or 4-h (partial sleep deprivation condition) sleep opportunity per night. The phase shifts for an individual subject are connected by a line. The subject identified as an outlier is represented with the open circles. Ten of the remaining 13 subjects showed reduced phase advances when sleep deprived.