Diurnal variation of mood and the cortisol rhythm in depression and normal states of mind.

A large scale chronobiological investigation was undertaken in 20 drug-free psychiatric inpatients displaying RDC major depression (endogenous subtype) in comparison to 10 healthy control subjects and 10 of the patients after clinical recovery. A series of measurements was taken 6 times a day and, in 8 of a total of 14 variables, also once a night over a period of 10 to 14 days. The following variables were assessed: mood (three different scales), performance (two tests), motor activity (three measures), salivary flow, urinary excretion of water, sodium, potassium, and free cortisol (UFC), and rectal temperature. A phase chart of the acrophases of the 8 variables with measurements taken during day and night revealed two clusters in the depressives and three in the non-depressed subjects. In the depressives, the acrophases of the mood scales clustered around the time of awakening in the morning, together with the acrophase of UFC, whereas all other acrophases clustered in the afternoon. In the non-depressed subjects, however, the mood scales reached their circadian maxima in the middle of the night around the time when sleep was interrupted to take measurements. All other acrophases corresponded roughly with those found in the depressives. The coincidence of the time course of depressed mood and cortisol excretion in the patients was interpreted as reflecting a temporal relationship between diurnal mood swings in depression and the cortisol rhythm. This interpretation was supported by the significant correlation between the acrophases of the two respective rhythms in patients showing a significant diurnal variation in mood. The mood curves of non-depressed subjects seemed unrelated to the cortisol rhythm. Probably, they mirror diurnal fluctuations of vigilance rather than fluctuations of mood. According to the literature, this rhythm is temporally related to the rhythm of melatonin secretion.