Testing the hypothesis of a circadian phase disturbance underlying depressive mood in nonseasonal depression

In a crossover design, 8 nonseasonal depressed subjects, selected on the presence of diurnal mood variations, and 8 sex- and age-matched controls were exposed to dim light (< 10 lux) in the evening (18:00-21:00 h) and bright light (2500 lux) in the morning (ML, 6:00-9:00 h), to dim light in the morning and bright light in the evening (EL), or to dim light both in the evening and in the morning (DL) during 3 consecutive days in each of these conditions. There were no initial phase differences between depressed and healthy subjects in the timing of dim light melatonin onset, sleep termination, and body temperature. The phase shifts after EL and ML in both healthy and depressed subjects were as expected on the basis of a human phase response curve. On average, there was no therapeutic effect of the light exposure in the depressed patients. Two patients improved, but these effects do not seem to be related to shifts in the circadian system.