Serotonergic antidepressants have been shown to be effective treatments for premenstrual dysphoric disorder (PMDD). The efficacy of nonserotonergic antidepressants is less well studied. This study was a two-center, parallel design, placebo-controlled, randomized trial of fluoxetine, bupropion, and placebo in women with PMDD. Thirty-four women with PMDD completed 1 month of single-blind placebo and 2 months of fluoxetine 20 mg/day (N = 10), bupropion 100 mg three times daily (N = 12), or placebo (N = 12). Clinical Global Impressions (CGI) Scale, an expanded form of the Hamilton Rating Scale for Depression (HAM-D), and Global Assessment Scale (GAS) ratings were obtained premenstrually in each of the three treatment cycles. The three treatment groups differed significantly in efficacy by CGI ratings. Fluoxetine was superior to both bupropion and placebo. Comparison of posttreatment to pretreatment HAM and GAS scores demonstrated significant superior efficacy of fluoxetine compared with placebo. Posttreatment HAM and GAS scores for bupropion were intermediate between but not significantly different from fluoxetine or placebo. In summary, fluoxetine was significantly superior to bupropion and placebo as an effective treatment for PMDD. Although some improvement with bupropion was noted, and both medications were well tolerated, patient satisfaction was far greater with fluoxetine. (J Clin Psychopharmacol 1997;17:261-266).