Progesterone Might Be an Active Component of the Sleep-Wake Homeostatic Mechanism

Abstract

Abstract Background: Using a randomized, double blind, placebo-controlled, crossover protocol, we have shown previously that progesterone may prevent sleep disturbances but has no action on undisturbed sleep. In that study, 8 healthy postmenopausal women took daily at 2300 h for 3 wk a capsule of either 300 mg of progesterone or placebo. Sleep was polygraphically recorded during the last two nights and, during the second night, blood samples were collected at 15-min intervals using an iv catheter,. During the first night, sleep was normal under placebo and progesterone had no effect. During the second night, blood sampling procedure was associated under placebo with marked sleep disturbances, which were considerably reduced under progesterone: at the group level, mean ± SEM duration of wake after sleep onset (WASO) dropped from 152 ± 37 min to 71 ± 19 min (P = 0.01), and slow-wave sleep (SWS) duration increased from 53 ± 6 min to 79 ± 10 min (P = 0.04). Objective: To submit individual data collected in that study to new analyses designed to further investigate possible mechanism(s) of progesterone actions on sleep architecture. Results: Among individual subjects, no relation could be evidenced between progesterone levels and sleep variables, or between individual progesterone-associated improvements and absolute values of corresponding sleep variables under placebo. By contrast, for WASO and SWS, significant positive correlations (Spearman test) were evidenced between individual responses to progesterone (i.e. the difference, during night 2, between value under progesterone and value under placebo) and corresponding individual alterations caused under placebo by the blood sampling procedure (i.e. the difference, under placebo, between value during night 1 and value during night 2): WASO: rs = 0.74, P = 0.037, n = 8; SWS: rs = 0.86, P = 0.014, n = 7. (Pearson test yielded similar results: WASO: r = 0.85, P = 0.008; SWS: r = 0.76, P = 0.047). Conclusions: Although they obviously need to be confirmed by larger studies performed in a variety of clinical conditions, the present findings suggest that progesterone action on sleep architecture is specifically tailored to restore individual normality rather than group normality. Since SWS is mainly regulated by the sleep-wake homeostatic mechanism relating sleep pressure to the duration of prior wakefulness, it is tempting to speculate that progesterone might be an active component of this mechanism.