Percutaneous closure of patent foramen ovale in patients with cryptogenic embolism: a network meta-analysis

Abstract
Up to 40% of ischaemic strokes are cryptogenic. A strong association between cryptogenic stroke and the prevalence of patent foramen ovale (PFO) suggests paradoxical embolism via PFO as a potential cause. Randomized trials failed to demonstrate superiority of PFO closure over medical therapy. Randomized trials comparing percutaneous PFO closure against medical therapy or devices head-to-head published or presented by March 2013 were identified through a systematic search. We performed a network meta-analysis to determine the effectiveness and safety of PFO closure with different devices when compared with medical therapy. We included four randomized trials (2963 patients with 9309 patient-years). Investigated devices were Amplatzer (AMP), STARFlex (STF), and HELEX (HLX). Patients allocated to PFO closure with AMP were less likely to experience a stroke than patients allocated to medical therapy [rate ratio (RR) 0.39; 95% CI: 0.17–0.84]. No significant differences were found for STF (RR 1.01; 95% CI: 0.44–2.41), and HLX (RR, 0.71; 95% CI: 0.17–2.78) when compared with medical therapy. The probability to be best in preventing strokes was 77.1% for AMP, 20.9% for HLX, 1.7% for STF, and 0.4% for medical therapy. No significant differences were found for transient ischaemic attack and death. The risk of new-onset atrial fibrillation was more pronounced for STF (RR 7.67; 95% CI: 3.25–19.63), than AMP (RR 2.14; 95% CI: 1.00–4.62) and HLX (RR 1.33; 95%-CI 0.33–4.50), when compared with medical therapy. The effectiveness of PFO closure depends on the device used. PFO closure with AMP appears superior to medical therapy in preventing strokes in patients with cryptogenic embolism.