While data dealing with neurobiological effects of sleep deprivation (SD) are mainly restricted to the acute effects of a single night, only few studies have investigated mid-term effects after repeated SD. We therefore examined the clinical and hormonal characteristics of depressive patients before and after serial SD to determine potential sustained effects, focusing especially on serotoninergic functions. One tool to investigate serotoninergic dysfunction in depression is the use of serotoninergic agents to stimulate hormonal secretion, which is assumed to normalize during a clinically effective therapy. Eighteen drug-free inpatients with unipolar major depression received cognitive-behavioral treatment for three weeks and - according to a randomized control design - additional SD therapy (six nights of total SD within three weeks, separated by nights of recovery sleep) or no SD therapy (control group). Serotoninergic function was assessed by measuring cortisol and prolactin in response to intravenously administered clomipramine (12.5mg) before and after the treatment period. The post-treatment challenge test was performed three days after the last SD night. Apart from of a transient overnight improvement of mood induced by SD, both groups showed a comparable clinical course during the three-week treatment period. Compared to the control group, the SD-treated patients exhibited significantly decreased pre-stimulation cortisol levels and significantly increased cortisol responses to clomipramine, whereas no treatment effects were observed for prolactin. In conclusion, our findings suggest that the mid-term effects of serial SD therapy lead to a normalization of serotoninergic dysfunction, although an obvious impact on clinical symptoms was not detected.