Background: Complications of percutaneous patent foramen ovale (PFO) closure using the AmplatzerTM PFO Occluder were reported in 2012 after a median follow-up of 2.1 years. The FDA requested an analysis of long-term device safety observed in the Randomized Evaluation of Recurrent Stroke Comparing PFO Closure to Established Current Standard of Care Treatment (RESPECT) trial.
Methods: We randomized patients with cryptogenic stroke and PFO in a 1:1 ratio between PFO closure and medical management alone (MM). Post procedure medical management in the closure arm was 30 days of aspirin and clopidogrel followed by 5 months of aspirin alone and continued medical therapy as per the site neurologist. Medical treatment in the MM arm was a guideline-directed antiplatelet regimen or warfarin. An independent data and safety monitoring board adjudicated all adverse events as serious (SAE) or non-serious, and as procedure, device or protocol related.
Results: We enrolled 980 patients (mean age, 45.9 years) at 69 sites who were followed for a median of 5.9 years (IQR 4.2-8.0). Follow-up was unequal – 3141 patient-years in the closure arm vs. 2669 in the MM arm, due to a higher dropout rate in the MM arm. There was less warfarin use in the closure arm vs. the MM arm (109 vs 578 patient-years). There were no study-related deaths. Two patients had procedure-related ischemic strokes (7 days and 3 months post-procedure). SAEs were equally distributed between the two arms: 13.7 in the closure arm vs. 12.4 per 100 patient-years in the MM arm, p=0.17. There was no device thrombus or erosion. The incidence of post procedure AF was not increased in the closure arm. Venous thromboembolic events (VTE) occurred at a higher rate in the closure arm, 0.87 per 100 patient-years vs. 0.22 in the MM arm, p=0.0008. In univariable logistic regression modeling, a remote history of DVT prior to randomization was a significant predictor of VTEs in the closure arm. No VTEs occurring beyond 6 months were attributed to the procedure or the device.
Conclusions: These new long-term data reaffirm that the AmplatzerTM PFO Occluder has a low rate of procedure (2.4%) and device (2%) related SAE’s. A small subset of cryptogenic stroke patients have an underlying proclivity to VTE and may need long-term anticoagulation.