5-HT brain function in affective disorder: d,l-fenfluramine-induced hormone release and clinical outcome in long-term...

Prolactin (PRL) and Cortisol (CORT) responses to d,l-fenfluramine (FEN) challenge (60 mg) were examined in patients with affective disorders on two occasions under euthymic conditions: drug-free before admission to prophylactic treatment and after about 9 months of medication with lithium or carbamazepine. Response to treatment was assessed by a complex algorithm using continuous ratings in outpatient clinic over a period of 2 years. In general, treatment resulted in a delayed and diminished CORT release (n.s.); subgroup analysis revealed an attenuated CORT response (P < 0.05) in responders, whereas nonresponders showed no change in CORT secretion pattern except an enhanced CORT baseline value (P < 0.05). Cross-sectional comparison of responders with nonresponders under medication yielded a trend for greater CORT stimulation in nonresponders. This result was not affected by FEN/NorFEN or lithium/carbamazepine serum levels, baseline CORT values, age, sex, diagnostic distribution, number of appointments to the outpatient clinic or duration of medication at the time of FEN test session. Before onset of prophylactic medication responders and nonresponders could not be discriminated significantly regarding stimulated hormone release, probably due to the small sample size (n = 17). CORT response to FEN was increased in drugfree unipolar patients compared to bipolar (P < 0.05) and to schizoaffective patients (P < 0.1). In accordance with its well-documented presynaptic 5-HT-agonistic action lithium medication resulted in a significantly greater increase in CORT release than carbamazepine (P < 0.05). Evaluation of PRL stimulation showed patterns of secretion quite similar to those of CORT, without reaching statistical significance in most cases. Perhaps due to methodological differences in assessing treatment response, these data do not confirm former results, which supposed an enhanced 5-HT net activity in long-term prophylactic lithium treatment. Because of high interindividual variances of hormone parameters, the FEN-test procedure is not a useful tool for the prediction of therapeutical outcome in terms of clinical routine use. Relations of stimulated hormone response as a marker of central serotoninergic activity and clinical outcome are discussed.