Late, but not early, wake therapy reduces morning plasma melatonin: relationship to mood in Premenstrual Dysphoric Disorder

Abstract

Wake therapy improves mood in Premenstrual Dysphoric Disorder (PMDD), a depressive disorder in DSM-IV. We tested the hypothesis that the therapeutic effect of wake therapy in PMDD is mediated by altering sleep phase with melatonin secretion. We measured plasma melatonin every 30 min (18:00-09:00 h) in 19 PMDD and 18 normal control (NC) women during mid-follicular (MF) and late luteal (LL) menstrual cycle phases, and during LL interventions with early wake therapy (EWT) (sleep 03:00-07:00 h)(control condition) vs. late wake therapy (LWT) (sleep 21:00-01:00 h)(active condition). Melatonin offset was delayed and duration was longer in the symptomatic LL vs. asymptomatic MF phase in both NC and PMDD subjects. LWT, but not EWT, advanced offset and shortened duration vs. the LL baseline, although they improved mood equally. Later baseline LL morning melatonin offset was associated with more depressed mood in PMDD patients, and longer melatonin duration in the MF phase predicted greater mood improvement following LWT. That LWT, but not EWT, advanced melatonin offset and shortened duration while they were equally effective in improving mood suggests that decreasing morning melatonin secretion is not necessary for the therapeutic effects of wake therapy in PMDD.

Figure 1

Flow diagram showing Baseline and Treatment Conditions over the three-month study period after the evaluation phase. Mood, sleep and melatonin (MLT) sampling were carried out during one Mid Follicular (MF) and three consecutive Late Luteal (LL) phases. Interventions during Months 2 and 3 included Late Wake Therapy (LWT) or Early Wake Therapy (EWT), presented in a randomized, crossover design.

Figure 2

Mean melatonin offset time (± SEM) during Late Luteal baseline, Late Wake Therapy (LWT), and Early Wake Therapy (EWT) conditions in patients with Premenstrual Dysphoric Disorder (PMDD; closed bars, N = 13) and Normal Control (NC; open bars, N = 12) subjects. P-values identify significant differences between LWT and LL baseline (p = .045), and LWT and EWT (p = .002); EWT did not differ from LL baseline (p > .05).

Figure 3

Mean melatonin duration (± SEM) during Late Luteal baseline, Late Wake Therapy (LWT), and Early Wake Therapy (EWT) conditions in patients with Premenstrual Dysphoric Disorder (PMDD; closed bars, N = 13) and Normal Control (NC; open bars, N = 12) subjects. P-values identify significant differences between LWT and LL baseline (p = .014), and LWT and EWT (p = .005); EWT did not differ from LL baseline (p > .05).

Figure 4

Change in Hamilton Depression Rating Scale (HDRS) score from the mid follicular (MF) to the late luteal (LL) menstrual cycle phase in relation to Melatonin Offset Time during the LL baseline phase in patients with Premenstrual Dysphoric Disorder (PMDD, N = 14).

Figure 5

Change (improvement) in Hamilton Depression Rating Scale (HDRS) score after Late Wake Therapy Recovery (LWT-R) sleep as a function of melatonin duration during the mid follicular (MF) phase in patients with Premenstrual Dysphoric Disorder (PMDD, N = 14).