Optimal outcomes from depression treatment are long-term recovery and, in the case of recurrent depression, prevention of new episodes. However, few data are available concerning the long-term efficacy of antidepressants in prophylactic treatment to prevent recurrences of depression. The efficacy and safety of fluoxetine 20 mg/day was evaluated in reducing the number of depressive episodes and in extending the time free of symptoms in patients with recurrent unipolar major depression. Patients with recurrent unipolar major depression according to DSM-III-R criteria and who responded to 32 weeks of open-label fluoxetine were randomly assigned to receive fluoxetine 20 mg/day (N = 70) or placebo (N = 70) for 48 weeks of double-blind maintenance treatment. Outcome measures were the percentage of recurrences and time to recurrence. Safety assessments included treatment-emergent adverse events, reasons for discontinuation, vital signs, and laboratory measures. Fluoxetine was associated with a statistically significantly smaller percentage of patients who had a recurrence compared with placebo (20% vs. 40%; χ2analysis,p= 0.010). The symptom-free period was significantly longer for patients treated with fluoxetine versus placebo (295 vs. 192 days; Kaplan-Meier estimates, log-rank test,p= 0.002). Treatments were well tolerated during maintenance treatment. The only statistically significant difference in adverse events between treatment groups was anxiety, which was more frequent in the placebo group (fluoxetine, 12.9% vs. placebo, 30%; χ2analysis,p= 0.013). Two placebo-treated patients and no fluoxetine-treated patients were withdrawn because of adverse events. In conclusion, fluoxetine at 20 mg/ day was effective and well tolerated for the prophylactic treatment of recurrent unipolar major depression.