The purpose of the present study is to compare the efficacy of imipramine in the treatment of psychotic versus nonpsychotic depression. Previous studies report varying results of monotherapy with antidepressants in psychotic depression. Because psychotic depression is seriously underinvestigated, performing a post hoc analysis of randomized clinical trials consisting of both psychotic and nonpsychotic depressed patients may contribute to the discussion on the optimal treatment of depressed patients with mood-congruent psychotic features. A total of 112 patients diagnosed with major depression (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition) (40 with psychotic depression and 82 without psychotic features) received treatment with imipramine for 6 weeks after a washout period of 7 days. Imipramine doses were adjusted to attain a predefined fixed plasma level. Treatment response was assessed with the Hamilton Rating Scale for Depression (HAM-D). A logistic regression analysis showed a significantly larger reduction in HAM-D score in the sample with psychotic features compared with the nonpsychotic sample (regression coefficient, -3.47; SE, 1.7; P = 0.04). According to the primary outcome criterion, that is, the change in HAM-D score, imipramine was significantly more effective in the sample with psychotic depression compared with the nonpsychotic depressed patients. The contradiction between our results and those of several previous studies may be due to the fixed plasma level dosing of imipramine refraining from concurrent psychotropic medication or limiting our patient sample to patients with mood-congruent psychotic features.