We sought to determine whether fluvoxamine and fluoxetine, two different antidepressants with in vitro selectivity for the serotonin uptake transporter also demonstrated similar selectivity in vivo. To accomplish this, we measured cerebrospinal fluid (CSF) concentrations of 5-hydroxyindoleacetic acid (5-HIAA), 3-methoxy-4-hydroxyphenylglycol (MHPG), and homovanillic acid (HVA) before and after 6 weeks of treatment with these two drugs. Twenty-four subjects who had major depression according to DSM-III-R criteria gave written, informed consent for the collection of CSF during a double-blind comparative treatment trial of fluvoxamine (50-150 mg/day) and fluoxetine (20-80 mg/day). The symptoms of subjects were assessed clinically on a weekly basis throughout the treatment trial. CSF samples were obtained after a 7- to 14-day washout period before treatment and again at the end of treatment. CSF samples were analyzed for 5-HIAA, HVA, and MHPG using high-pressure liquid chromatography coupled to electrochemical detection. Fluvoxamine- and fluoxetine-treated patients did not differ in clinical outcome or in the CSF concentrations of monoamine metabolite levels before or after treatment. Therefore, the CSF data were pooled. Drug treatment, overall, was associated with significant decreases in 5-HIAA and MHPG and a trend toward a reduction in HVA levels. Levels of 5-HIAA, MHPG, and HVA were reduced by 57%, 48%, and 17%, respectively. In addition, the magnitude of the decreases in 5-HIAA and MHPG appeared to be correlated (r = 0.83) across the subjects, although a Spearman rank correlation indicated that outlying values had an undue effect on this relationship. These results suggest that treatment with selective serotonin reuptake inhibitors, which are selective for serotonin uptake in vitro, does not show a similarly selective effect on serotonin in vivo during treatment of patients.