Periprocedural stroke risk in patients undergoing catheter ablation for atrial fibrillation on uninterrupted warfarin.

BACKGROUND: Catheter ablation is an effective treatment for symptomatic individuals with atrial fibrillation (AF) but is associated with a risk of periprocedual stroke. Recent data suggest that this risk may be abolished if catheter ablation is performed with uninterrupted warfarin (UW). We sought to compare the incidence, severity and timing of periprocedural stroke between 2 periprocedural anticoagulation protocols: bridging low-molecular-weight heparin (LMWH) and UW. METHODS AND RESULTS: Periprocedural stroke (≤14 days) was assessed in 2,855 ablations performed in 1,813 patients. Thromboembolic stroke occurred in 11/1,653 (0.7%) procedures with bridging LMWH and in 5/1,202 (0.4%) procedures on UW (P = 0.5). Four of the 5 strokes (80%) on UW occurred despite a therapeutic INR and a mean activated clotting time of ≥300 seconds and 4/5 strokes (80%) occurred in patients with a CHADS2 score of 0. Eleven of 16 (69%) strokes overall occurred within 24 hours of the procedure. All 4 strokes resulting in major neurological deficit occurred in the LMWH group. Major bleeding complications occurred in 6.0% of patients in the bridging LMWH group compared to 4.0% in the UW group (P = 0.02). CONCLUSIONS: In contrast to existing data, periprocedural stroke still occurs despite therapeutic anticoagulation throughout the operative period. The optimal strategy to protect patients against thromboembolic stroke remains unclear.