From 3% to 5% of women of reproductive age develop functional impairment of the degree found in premenstrual dysphoric disorder (PMDD)-a severe form of premenstrual syndrome. Changes in mood, behavior, and appetite suggest that central serotonergic activity has a role in the development of PMDD. Antidepressants that inhibit serotonin reuptake have proved to be a consistently effective treatment for PMDD in randomized, double-blind trials, but a sizable number of patients fail to respond. This study evaluated venlafaxine, a new antidepressant that selectively inhibits reuptake of both serotonin and norepinephrine. Using a randomized, double-blind, placebo-controlled design, 164 women given a diagnosis of PMDD, who were free of other major psychiatric disorders, were assigned to receive venlafaxine in a daily dose of 50 to 200 mg or placebo over four menstrual cycles. Premenstrual symptoms were evaluated by a daily symptom report (DSR) and the Hamilton Rating Scale for Depression. Apart from meeting criteria for PMDD, the participants were 18 to 45 years of age, had had regular cycles for the past 6 months, and probably had ovulated. Active medication or placebo was given throughout the entire menstrual cycle. Of 157 evaluable patients, 56 withdrew during the 4-month treatment phase-roughly 35% of those in each treatment group. There were no serious adverse events. Total DSR scores indicated significantly greater improvement with venlafaxine than with placebo. The active drug reduced symptoms of PMDD in the first treatment cycle by 42%, compared with 14% for placebo recipients, and improvement increased in the second cycle. Venlafaxine remained significantly more effective than placebo in all four double-blind treatment cycles. DSR postmenstrual scores were comparable in the two groups. Venlafaxine provided improvement in emotion, function, pain, and physical symptoms but not in appetite. Hamilton depression scores also were significantly improved in the venlafaxine group. Rates of treatment response, with at least a 50% reduction from the pretreatment baseline in total premenstrual DSR scores, were 60% in the venlafaxine group and 35% for placebo patients. Analysis of Hamilton depression scores yielded very similar results. Whichever test was used, symptom remission was observed substantially more often in patients given venlafaxine. All these findings were ratified by the clinician-rated Clinical Global Impression scale. The new antidepressant venlafaxine, which blocks reuptake of serotonin and norepinephrine, seems to be an effective means of treating PMDD. Responses occur rapidly and treatment is generally well tolerated. Additional study may show that treatment only in the symptomatic luteal phase is a reasonable approach.