It is generally accepted that the clinical efficacy of monoamine oxidase inhibitors (MAOI) is related to inhibition of this enzyme. In order to evaluate the predictive ability of monoamine oxidase-A inhibition for therapeutic efficacy, the start of treatment effects of moclobemide, a selective, reversible monoamine oxidase-A inhibitor, on plasma concentrations of monoamines and monoamine metabolites were determined. The plasma levels of 3,4-dihydroxyphenylglycol (DHPG, deaminated metabolite of noradrenaline), 5-hydroxyindoleacetic acid (5-HIAA, deaminated metabolite of serotonin), 3,4-dihydroxyphenylacetic acid and homovanillic acid (DOPAC and HVA, deaminated metabolites of dopamine), L-dihydroxyphenylalanine (L-dopa) and noradrenaline were investigated and related to treatment outcome. This was a randomized double blind parallel group study in 47 patients with criteria of major depression according to DSM III R. Moclobemide 300 mg/day, 450 mg/day or 600 mg/day was administered continuously for 6 weeks. Plasma concentrations of monoamine metabolites and monoamines were determined just before treatment by moclobemide, 4 h after the first dose, 24 h after the first dose, before the first dose on day 7, and 4 h after the first dose, on day 7. Each moclobemide dose improved depression as measured by MADRS (Montgomery-Asberg Depression Rating scale) but there was no difference between the three doses. Moclobemide dose-dependently reduced plasma concentration of DHPG, L-dopa and HVA. No dose-dependent treatment effect was observed for plasma 5-HIAA, noradrenaline and DOPAC. The clinical outcome as defined by the final MADRS score was not related to any start of treatment changes in plasma monoamine metabolites reflecting inhibition of MAO-A. It is concluded that monoamine oxidase-A inhibition at the beginning of the treatment does not predict clinical outcome.