The effect of comedication with fluvoxamine on the plasma concentrations of the enantiomers of citalopram and its metabolites in dextromethorphan/mephenytoin phenotyped patients pretreated with citalopram (CIT) was studied: seven female patients (45.1 +/- 13.9 years) suffering from a major depressive episode [ICD-10: F32.2 (n = 3 patients), F33.2 (n = 2), F32.10 (n = 1) or F32.11 (n = 1)], who were non-responders to a 3-week treatment with 40 mg/day CIT (From day-21 to day 0) (day 0: MADRS score > or = 12), were co-medicated for another 3 weeks with fluvoxamine (50 mg/day from day 1-7, 100 mg/day from day 14-21). All patients were extensive metabolizers of mephenytoin (CYP2C19) and dextromethorphan (CYP2D6), except one patient, who had a genetic deficiency of CYP2D6. There was a significant increase of the plasma concentrations of S- and R-citalopram from day 0 (27 +/- 14 micrograms/l and 55 +/- 23 micrograms/l, respectively) to day 21 (83 +/- 38 micrograms/l and 98 +/- 44 micrograms/l, respectively), after addition of fluvoxamine (P < 0.02, for each comparison), and the mean ratio S/R-citalopram increased from 0.48 to 0.84. S-Citalopram inhibits more potently 5-HT uptake than R-citalopram: therefore, fluvoxamine increases the pharmacologically more active S-citalopram with some stereoselectivity. According to a previous in vitro study, this pharmacokinetic interaction occurs on the level of CYP2C19, but also of CYP2D6 and CYP3A4 which, in contrast to CYP1A2, contribute to the N-demethylation of citalopram and which are stereoselectively inhibited by fluvoxamine. All but one patient showed clinical improvement by a decrease of the MADRS score by at least 50% and a final score < or = 13 (mean +/- SD: day 0:30.6 +/- 9.2; day 21:11.0 +/- 6.5). Some patients showed minor symptoms, such as nausea and tremor, but the combined treatment was generally well tolerated.