To date, no large-scale, controlled trial comparing a serotonin-norepinephrine reuptake inhibitor and selective serotonin reuptake inhibitor with placebo for the treatment of panic disorder has been reported. This double-blind study compares the efficacy of venlafaxine extended-release (ER) and paroxetine with placebo. A total of 664 nondepressed adult outpatients who met DSM-IV criteria for panic disorder (with or without agoraphobia) were randomly assigned to 12 weeks of treatment with placebo or fixed-dose venlafaxine ER (75 mg/day or 150 mg/day), or paroxetine 40 mg/day. The primary measure was the percentage of patients free from full-symptom panic attacks, assessed with the Panic and Anticipatory Anxiety Scale (PAAS). Secondary measures included the Panic Disorder Severity Scale, Clinical Global Impressions--Severity (CGI-S) and--Improvement (CGI-I) scales; response (CGI-I rating of very much improved or much improved), remission (CGI-S rating of not at all ill or borderline ill and no PAAS full-symptom panic attacks); and measures of depression, anxiety, phobic fear and avoidance, anticipatory anxiety, functioning, and quality of life. Intent-to-treat, last observation carried forward analysis showed that mean improvement on most measures was greater with venlafaxine ER or paroxetine than with placebo. No significant differences were observed between active treatment groups. Panic-free rates at end point with active treatment ranged from 54% to 61%, compared with 35% for placebo. Approximately 75% of patients given active treatment were responders, and nearly 45% achieved remission. The placebo response rate was slightly above 55%, with remission near 25%. Adverse events were mild or moderate and similar between active treatment groups. Venlafaxine ER and paroxetine were effective and well tolerated in the treatment of panic disorder.