A 1-week sleep and light intervention improves mood in premenstrual dysphoric disorder in association with shifting melatonin offset time earlier

A potentially disabling condition, premenstrual dysphoric disorder (PMDD) causes extensive personal suffering, occupational impairment, and disruption of interpersonal and family relationships. While 20–80% of women report mood, cognitive, and behavioral disturbances associated with their menstrual cycle (Hamilton et al. 1984; Parry and Wehr 1987), PMDD can progress from premenstrual syndrome (PMS) to major depression (MD) (Halbreich and Endicott 1985; Hamilton et al. 1984), increase risks for peripartum and menopausal depression (Parry et al. 1995), and exacerbate bipolar illness (Dias et al. 2011; Payne 2011). Premenstrual mood symptoms meet criteria for a mental disorder in 5–8% of menstruating women (Accortt et al. 2008; American Psychiatric Association 2013; Yonkers et al. 2008). Current PMDD treatments are only marginally efficacious (< 60% response) (Halbreich et al. 2006), and many women do not want to use, or cannot tolerate, taking chronic antidepressant medications for a periodic illness. Pharmacological interventions are not efficacious in over 40% of women (Halbreich et al. 2006), are associated with problematic side effects, increase the risk of breast and ovarian cancer (Cosgrove et al. 2011), and do not improve health outcomes (Borenstein et al. 2007). Consequently, enhanced PMDD treatments are needed (Borenstein et al. 2007).

A single night of total or partial “wake therapy” (WT) (previously called therapeutic sleep deprivation) produces a rapid, albeit transient, antidepressant response in 40–60% of patients (Giedke and Schwarzler 2002; Schilgen and Tolle 1980; Wirz-Justice and Van den Hoofdakker 1999; Wirz-Justice and Terman 2012; Wirz-Justice et al. 2013; Wu and Bunney 1990). Early-night wake therapy (EWT: i.e., remaining awake until 3:00 am, then sleeping from 3:00–7:00 am), or late-night wake therapy (LWT: i.e., sleeping from 9:00 pm–01:00 am and remaining awake until the following night), benefits mood as much as “total-night” WT (Leibenluft and Wehr 1992; Parry and Wehr 1987; Wirz-Justice et al., 2013). LWT is more efficacious than EWT in some, but not all, MDs (Parry and Wehr 1987; Wirz-Justice et al. 2013). We confirmed wake therapy efficacy in PMDD (Parry and Wehr 1987; Parry 1995) and peripartum depression (Parry et al. 2000, 2019).

Light treatment sustains wake therapy benefits often lost after recovery sleep, and wake therapy hastens and potentiates light treatment effects taking 5–10 weeks to significantly improve mood in peripartum non-seasonal depression (Corral et al. 2000, 2007; Epperson et al. 2004; Oren et al. 2002; Wirz-Justice et al. 2011). We previously reported and replicated the independent antidepressant effects of light (Parry et al. 1989; 1993; 1997a) and wake therapy (Parry and Wehr, 1987; Parry et al. 1995) in PMDD. AM light significantly improved scores on the Hamilton Rating Scale for Depression (HRSD) by 50% (Parry et al. 1997a), replicated using a shorter, brighter light pulse (Parry et al. 2011). LWT significantly reduced mean HRSD depression scores by 62.2% (Parry et al. 1995). In the current study, we combined wake and light interventions to enhance their individual benefits.

As misaligned circadian rhythms characterize mood disorders (Goel et al. 2013; Monteleone et al. 2011; Srinivasan et al. 2006; Wehr and Wirz-Justice 1982), we sought to test the hypothesis that wake and light interventions exert antidepressant effects by correcting misaligned circadian rhythms (CRs), best measured by melatonin in humans. In PMDD, we previously found plasma melatonin CR were phase delayed (shifted later) in the symptomatic luteal, compared with the asymptomatic follicular, menstrual cycle phase, which correlated with more depressed mood (Parry et al. 1997b, 2008a). Consistent with the internal coincidence model of sleep deprivation and depression (Wehr and Wirz-Justice 1981), baseline timing disturbances between sleep and melatonin became normalized after treatment (Parry et al. 1989, 1997a, 1999). We reasoned that advancing and restricting sleep with 1 night of LWT followed by 7 days of the morning (AM) bright white light (BWL), which phase advances CRs (active intervention), would benefit mood more than delaying and restricting sleep with a single night of EWT followed by 7 days of the evening (PM) BWL, which phase delays CRs (control intervention).

Baseline demographic characteristics of study participants are reported in Table 1.

The aim of this study was to determine whether PMDD depressed mood could be improved using phase-advancing sleep and light intervention (SALI) to correct the phase delay in melatonin circadian rhythms (CR) we found in PMDD (Parry et al. 1997a, 2008a). We hypothesized that since estradiol advances, and progesterone delays CR (Albers et al. 1981; Morin et al. 1977), fluctuations in these hormones during the menstrual cycle could dysregulate CR, and correcting CR misalignment could thereby improve depressed mood. In support of this hypothesis, we found the following: (1) In the baseline luteal phase, PMDD atypical depression symptoms were associated with a greater phase delay of 6-SMT offset; (2) A phase-advance intervention (PAI: late-night wake therapy-LWT plus AM bright white light-BWL) produced significantly greater mood improvement than a phase-delay intervention (PDI: early-wake therapy-EWT plus PM BWL); and (3) After PAI, improvement in HRSD and SIGH-ADS mood scores correlated significantly with phase advance in 6-SMT offset.

Importantly, we designed the study arms to create identical sleep restriction durations (4 h), plus identical bright light exposures (60 min/day for 7 days), but at different times of day, to effect maximal CR realignment. Thus, the treatments varied only in the times of day when sleep restriction and light exposure were instituted, with PAI designed to advance 6-SMT timing vs. PDI. That mood benefits and 6-SMT phase-advanced timing were, indeed, significantly correlated in the PAI vs. PDI study arms confirmed that hypothesis.

We discuss implications of these findings in relation to previous literature below.

These results suggest potential benefits for other reproductively related mood disorders when a circadian pathophysiology is identified. For example, in pregnancy, we showed plasma and urinary melatonin were phase advanced in depressed vs. healthy participants (Parry et al. 2008b), and mood improved significantly more after phase-delaying than phase-advancing SALI. In contrast, in postpartum-depressed participants plasma and urinary melatonin were phase delayed (Parry et al. 2008b), and mood improved more after phase-advancing than phase-delaying SALI. In menopausal, like postpartum-depressed participants, plasma and urinary melatonin were phase delayed (Parry et al. 2008c), and mood improved more after phase-advancing vs. phase-delaying SALI. Thus, each reproductively related depressive disorder has a specific chronobiological abnormality that provides a basis for optimizing interventions targeted to specific pathogenic mechanisms.

The advantage of combining sleep and light interventions is that the sleep intervention hastens and potentiates mood benefits of light treatment, and light treatment maintains mood benefits of sleep intervention, which otherwise may be lost after a night of recovery sleep (Wirz-Justice 2011). Initially, we used the sleep intervention alone in PMDD (Parry and Wehr 1987; Parry et al. 1995) where, in contrast to other mood disorders, women with PMDD did not relapse after a night of recovery sleep; i.e., they were more night 2, rather than night 1, responders, which may indicate a more serotonergic, rather than noradrenergic, pathophysiology, supported by extensive studies (see review Parry 2001). As light treatment alone in non-seasonal depression may require at least 5 weeks to improve mood significantly (Parry and Maurer 2003), light treatment, without the priming effect of wake therapy, is less efficacious for PMDD when administered during the necessarily limited time period of 7 days in the symptomatic late luteal phase (Parry et al. 1989, 1993). The advantage of combining interventions is that sleep interventions may produce rapid improvements, which are subsequently sustained by light interventions.

These sleep and light interventions are safe, efficacious, rapid-acting, affordable, readily repeatable, well-tolerated with few side effects, non-pharmacological and non-hormonal home treatments that can be administered by paraprofessionals without office or clinic visits.

Study limitations include a relatively small sample size, from only a single study site, and melatonin measured by 6-SMT rather than plasma. Further, as 37.5% of participants did not complete the full protocol, we confined conclusions about primary study outcomes to results of only participants who completed the full protocol. Although in our earlier study of wake therapy (Parry et al. 1995), the duration of benefit lasted up to 6 months, the durability of sleep + light benefits in the present study needs to be tested.

This work was supported by the National Institutes of Health R01 AT007169.