Neuroendocrine profiles in mood disorders.

The study of neuroendocrine abnormalities in major mental illness, such as the unipolar and bipolar affective syndromes, has been the focus of interest in the past few years. The neuroendocrine window into the brain has been considered as a fruitful and promising approach to the study of mental disorders, as suggested by studies of some neuroendocrine challenge tests in depression that demonstrated their potential use as biological markers. The modern approach to hormonal dynamics focuses on the circadian and pulsatile profiles that truly represent physiological modulation and tests the hypothesis that an abnormality in circadian rhythms is present in affective illness. From the fundamental point of view, such studies performed using a frequent sampling interval over the 24-h cycle aim to clarify the control and significance of the temporal sleep and wake fluctuations of neuroendocrine system activities. Twenty-four-hour hypersecretion of cortisol, diurnal hypersecretion of growth hormone, and normal 24-h levels of prolactin have been reported in careful chronobiological studies of depressed patients, along with sleep recordings. In addition, a nocturnal quiescent period, and a subsequent increase towards the morning maximum, have been consistently found in a subset of depressed patients suffering from endogenous depression. After successful treatment with antidepressants, most of these abnormalities (with the exception of those found in the prolactin study) tend to correct. The normalization of the timing of hormonal secretion was accompanied by a correction of sleep abnormalities and in particular, a lengthening of the REM latencies. Normalization of cortisol secretion was associated with a decrease in the magnitude of episodic cortisol pulses whereas normalization of growth hormone secretion was due to a diminished number of secretory pulses. In conclusion, a disorder of circadian time-keeping seems to characterize acute episodes of major endogenous depression in some patients. This abnormality as well as the associated increases in adrenocorticotropic and somatotropic activities seem to be a state-, rather than trait-dependent phenomenon.