Abstract
Background Transcatheter closure of patent foramen ovale (PFO) has reduced the risk of recurrent stroke in patients with cryptogenic strokes in randomized clinical trials. Whether PFO closure in clinical practice is associated with similar benefit remains unknown. Methods and Results We identified patients with PFO and a history of ischemic stroke or transient ischemic attack who were treated with PFO closure or medical therapy in the OptumLabs database. The primary end point was recurrent ischemic stroke or systemic embolization. Secondary outcomes included mortality, all stroke, transient ischemic attack, and major bleeding. A total of 6668 propensity-matched patients were included (PFO closure n=4111; medical therapy n=2557). The incidence of stroke or systemic embolization per 100 person-years was 2.38 after PFO cohort and 2.99 with medical therapy (hazard ratio [HR], 0.85 [95% CI, 0.68-1.05], P=0.13). Mortality was lower in the PFO closure cohort (1.78 versus 2.59 per 100 person-years: HR, 0.69 [95% CI, 0.55-0.87], P=0.002). Falsification end points showed that this difference is unlikely to be completely explained by residual confounders. There were no significant differences between the groups in secondary end points including intracranial hemorrhage and major bleeding except for an increase in nonintracranial hemorrhage bleeding among patients treated with oral anticoagulation (1.42 versus 2.16 per 100 person-years: HR, 0.69 [95% CI, 0.48-0.99], P=0.043). The main end point was consistent in subanalyses including patients <60 years of age, patients with prior stroke, and those treated after the publication of the positive PFO trials in 2017. Conclusions In contemporary US practice, PFO closure is not associated with lower rates of recurrent ischemic stroke or systemic embolization compared with medical therapy. Potential reasons for this discrepancy warrant further investigation.