TSH response to TRH and EEG sleep in non-bipolar major depression: a multivariate approach

The TSH response to TRH and selected sleep EEG variables were studied in a homogeneous sample of 280 non-bipolar major depressed inpatients (95 males and 185 females). The TSH response to TRH was blunted in 28% of the sample. delta max TSH was correlated negatively with age, Hamilton rating scale, Newcastle scale, percentage of wake, and positively with basal TSH, percentage of stage II, slow wave sleep, REM sleep and REM latency. delta max TSH was also lower in male patients and in patients suffering from an endogenous or a psychotic subtype of major depression. Basal TSH was only correlated negatively with the Newcastle score. In view of intercorrelations between all these variables, and because of the confounding effect of age, gender and severity on both the TSH response to TRH and sleep EEG variables, a multiple regression analysis was performed and demonstrated that basal TSH and gender were the two variables with the highest contribution to the delta max TSH variance, followed by age and the presence of psychotic symptoms. When controlling strictly for these significant effects, correlation with the severity or with the endogenous character of depression, and with sleep EEG parameters disappeared.