PFO closure vs. medical therapy in cryptogenic stroke or transient ischemic attack: A systematic review and meta-analysis

Background/objectives: This study aims to assess whether patent foramen ovale (PFO) closure is superior to medical therapy in preventing recurrence of cryptogenic ischemic stroke or transient ischemic attack (TIA). Methods: We searched PubMed for randomized trials which compared PFO closure with medical therapy in cryptogenic stroke/TIA using the items: "stroke or cerebrovascular accident or TIA"and "patent foramen ovale or paradoxical embolism" and "trial or study". Results: Among 650 potentially eligible articles, 3 were included including 2303 patients. There was no statistically significant difference between PFO-closure and medical therapy in ischemic stroke recurrence (1.91% vs. 2.94% respectively, OR: 0.64, 95% CI: 0.37-1.10), TIA (2.08% vs. 2.42% respectively, OR: 0.87, 95% CI: 0.50-1.51) and death (0.60% vs. 0.86% respectively, OR: 0.71, 95% CI: 0.28-1.82). In subgroup analysis, there was significant reduction of ischemic strokes in the AMPLATZER PFO Occluder arm vs. medical therapy (1.4% vs. 3.04% respectively, OR: 0.46, 95% CI: 0.21-0.98, relative-risk-reduction: 53.2%, absolute-risk-reduction: 1.6%, number-needed-to-treat: 61.8) but not in the STARFlex device (2.7% vs. 2.8% with medical therapy, OR: 0.93, 95% CI: 0.45-2.11). Compared to medical therapy, the number of patients with new-onset atrial fibrillation (AF) was similar in the AMPLATZER PFO Occluder arm (0.72% vs. 1.28% respectively, OR: 1.81, 95% CI: 0.60-5.42) but higher in the STARFlex device (0.64% vs. 5.14% respectively, OR: 8.30, 95% CI: 2.47-27.84). Conclusions: This meta-analysis does not support PFO closure for secondary prevention with unselected devices in cryptogenic stroke/TIA. In subgroup analysis, selected closure devices may be superior to medical therapy without increasing the risk of new-onset AF, however. This observation should be confirmed in further trials using inclusion criteria for patients with high likelihood of PFO-related stroke recurrence. c 2013 Elsevier Ireland Ltd. All rights reserved