A double-blind multicenter randomized parallel group study comparing paroxetine and maprotiline was carried out in a total of 544 outpatients. Included were patients with varying degrees of severity of depressive symptoms who fulfilled modified RDC criteria for either Minor or Major Depression and showed a HAMD-17 score of > or = 13. No concomitant benzodiazepine treatment was allowed. Duration of treatment was 6 weeks, after an initial wash-out period. Doses were fixed during the first 3 weeks of treatment, patients receiving either 20 mg paroxetine or 100 mg maprotiline daily. An option for dose escalation was provided for insufficient responders after 3 weeks. The weekly assessments comprised rating of the HAMD-17, MADRS, BRMS, RDS, HAMA, CAS, and CGI scales and registration of adverse events by non-leading questions. An intention-to-treat and a completer analysis were performed. Response was defined as a HAMD-17 reduction of > or = 50% or a HAMD-17 score of < or = 9 at the end of the study or at dropout. The treatment groups were comparable according to demographic data. Overall evaluation indicated equieffective and good antidepressant and anxiety-reducing properties for paroxetine and maprotiline. No persistent significant differences between treatment groups were observed on any assessment instrument. There was no difference in the frequency of observed side-effects, but side-effect profiles were markedly different, as maprotiline patients had more anticholinergic and paroxetine patients more SSRI-typical side-effects.