Comparison of efficacy and safety of milnacipran and fluoxetine in Korean patients with major depression

OBJECT: To compare efficacy and safety of milnacipran and fluoxetine in a population of Korean patients with major depression. RESEARCH DESIGN AND METHODS: The design was a multi-centre, randomised, comparative clinical study. Patients with major depression (DSM-IV diagnostic criteria) scoring over 17 points on the 17-item Hamilton Depression Scale (HAM-D) and over 21 points on the Montgomery-Asberg Depression Rating Scale (MADRS) were recruited and randomised to receive milnacipran (50 mg/day increasing after 1 week to 100 mg/day) or fluoxetine (20 mg/day) for 6 weeks. All previous medication was stopped at least 7 days before entry into the study. Patients were evaluated (HAM-D, MADRS and clinical global impression scale, CGI) at baseline and after 1, 2, 4 and 6 weeks of treatment. All adverse events which developed during the study period were recorded. RESULTS: 70 patients (milnacipran 39; fluoxetine 31) were included in the study. Total score on both HAM-D, MADRS and CGI decreased significantly in both groups after 1 week and continued to decrease throughout the study. There was no significant difference between the two groups for any measurement at any time point. Both antidepressants were well tolerated. In the milnacipran group, 13 patients reported 28 adverse reactions, and in the fluoxetine group 11 patients reported 18 adverse reactions. Two patients discontinued due to adverse events in the milnacipran group and three in the fluoxetine group. There were no clinically significant modifications in vital signs, routine blood laboratory tests, biochemistry or ECG throughout the study. Nausea and headache were the most frequently reported adverse events with milnacipran while digestive disturbances, diarrhoea and insomnia were more common with fluoxetine. CONCLUSION: Milnacipran, like fluoxetine, was found to be effective and well tolerated for the treatment of major depression in this population of depressed Korean patients. Principal limitations of the study were its open design, its small sample size and its relatively short duration.