Patent Foramen Ovale Closure in Stroke Trials: Pooled Analysis
What is the comparative effectiveness of patent foramen ovale (PFO) closure versus medical therapy in patients with cryptogenic stroke?
A pooled analysis of data from the three completed randomized trials of PFO closure versus medical therapy was performed. Two studies of the Amplatzer PFO occluder (RESPECT and PC) and one study of the STARFlex septal closure system (CLOSURE I) were included. The primary composite outcome was stroke, transient ischemic attack (TIA), or death. The secondary outcome was stroke. Log-rank tests and (unadjusted and covariate-adjusted) Cox regression models were used to compare device closure versus medical therapy.
The analysis included 2,303 patients, of which 909 were randomized to STARFlex versus medical therapy, and 1,394 were randomized to the Amplatzer PFO occluder versus medical therapy (980 in the RESPECT trial and 414 in the PC trial). Closure was not significantly associated with the primary composite outcome in the unadjusted analysis, but did become significant after covariate adjustment (hazard ratio [HR], 0.68; p = 0.049). For the outcome of stroke, all comparisons were statistically significant, with unadjusted and adjusted HRs of 0.58 (p = 0.043) and 0.58 (p = 0.044), respectively. In analysis limited to the Amplatzer occluder device trials, the effect of closure was not significant for the composite outcome, but was for the stroke outcome (unadjusted HR, 0.39; p = 0.013). Atrial fibrillation was more common among closure patients.
The authors concluded that among patients with PFO and cryptogenic stroke, closure reduced recurrent stroke and had a statistically significant effect on the composite endpoint of stroke, TIA, and death in adjusted, but not unadjusted analysis.
This study is the sixth meta-analysis of data from the existing randomized trials of PFO closure versus medical therapy in patients with cryptogenic stroke. The methodology is different in that individual patient data were used, which permitted covariate adjustment, proper handling of missing data, and exploration of effect heterogeneity. Overall, the risk of recurrent stroke was relatively low (1% with medical therapy and approximately 0.5% with device closure). The authors estimated the number needed to treat to prevent one ischemic stroke at 67. The number needed to harm in terms of the risk of atrial fibrillation was approximately 100, although the atrial fibrillation risk was higher with the since-discontinued STARFlex device. Previously stated limitations of the data set continue to apply, specifically concerns about the STARFlex device as well as potential for enrollment bias with the highest risk patients receiving devices outside of the randomization process. Longer-term analysis of the completed studies is underway and may be helpful in informing the decision regarding device closure to prevent recurrent neurologic events in patients with cryptogenic stroke.
< Back to Listings