The results of the meta-analysis of studies comparing the efficacy of moclobemide, imipramine and so-called sedative antidepressants (amitriptyline, mianserin and maprotiline) in 2416 patients are described. The results demonstrated that in agitated-anxious depressive patients (defined by HAMD factor score or HAMD item 9) a nonsedative, reversible MAO-A inhibitor moclobemide has about equal efficacy as imipramine or sedative antidepressants. All antidepressants were clearly superior to placebo, irrespective of the outcome measures applied (> 50% HAMD decrease, CGI improvement). The efficacy of antidepressants in agitated patients was unrelated to the severity of agitation and did not appear to be inferior to the efficacy in nonagitated patients. Comedication with benzodiazepines had no impact on overall efficacy of either moclobemide or other antidepressants in this patient population. Previous treatment with antidepressants, however, always negatively influenced the outcome with trial drugs, e.g., reduced their efficacy. Placebo response in agitated depressives appeared generally to be low (20-30%) and was clearly reduced with increased severity of agitation, irrespectively of how the agitation was defined.