Of cryptogenic stroke patients younger than 55 years of age, up to 61% have had a patent foramen ovale (PFO). Observational studies have revealed reductions in recurrent neurologic events through PFO closure versus medical therapy, and randomized controlled trials have shown nonsignificant trends toward benefit. We systematically searched for randomized controlled trials of percutaneous PFO closure with medical therapy versus medical therapy alone in patients with cryptogenic stroke and performed a meta-analysis of treatment outcomes. The primary endpoint was combined death, stroke, and transient ischemic attack. We included 3 trials. Of 2,303 total patients, 1,150 underwent PFO closure and 1,153 received medical therapy (median follow-up period, 2.6 yr). The pooled incidence of the primary endpoint was 1.2 events per 100 patient-years in the closure group (95% confidence interval [CI], 0.2-2.3) and 1.8 in the therapy group (95% CI, 0.7-2.9) (P=0.32); the number needed to treat was 167 (range, 100-500). The corresponding pooled hazard ratio was 0.67 (95% CI, 0.44-1.01; P=0.054) in favor of closure. Closure was associated with an increased risk of atrial fibrillation: relative risk=3.51 (95% CI, 1.44-8.55; P=0.006). When stratified by device, use of the Amplatzer™ PFO Occluder resulted in significant stroke-prevention benefit over medical therapy alone: hazard ratio=0.44 (95% CI, 0.21-0.95; P=0.037). When compared with medical therapy alone, PFO closure with medical therapy showed a trend toward a decreased hazard of combined events, although the absolute event reduction was small and the number needed to treat was high.