Embolic events after left atrial appendage closure for recurrent embolic events despite anticoagulation

Background Left atrial appendage closure (LAAC) is an established alternative to oral anticoagulation for stroke prevention in atrial fibrillation. LAAC is increasingly considered for recurrent embolic events despite oral anticoagulation (OAC), yet data on long-term outcomes in this emerging indication remain limited. Purpose To assess the embolic event rate after LAAC for recurrent embolic events despite adequate OAC compared to estimated event rates in patients managed with OAC alone and to other indications for LAAC. Methods We conducted a retrospective bi-center study (2012–2025) of patients undergoing successful LAAC. The primary endpoint was a composite of ischaemic stroke and systemic embolism. Secondary endpoints included all-cause death, major bleeding, device-related thrombosis (DRT), and ischaemic stroke. Patients were matched using propensity score matching (PSM) on age, CHA2DS2-VASc, HAS-BLED, diabetes, and peripheral artery disease. Survival analysis was adjusted for CHA2DS2-VASc and HAS-BLED scores in the overall population. Theoretical event rates were extracted from large literature cohorts. Results Of 814 patients, 66 underwent LAAC for recurrent embolic events and 748 for other indications (92.4% for bleeding under OAC). The recurrent embolic events group was younger (74.5 vs. 78 years, p=0.018), had lower HAS-BLED scores (3 vs. 4, p<0.001), and more peripheral artery disease (42.4% vs. 26.1%). At discharge, 92.4% received anticoagulant therapy versus 28.8% in patients with other LAAC indications (p<0.001). After a median follow-up of 20 [7-43] months, 52 embolic events occurred: 10 (15.2%) in the LAAC for recurrent embolic events group versus 42 (5.6%) for other indication (aHR 2.88, 95% CI [1.39-5.97], p=0.005), driven by a higher rate of stroke (aHR 3.92, 95% CI [1.86–8.26], p<0.001). All-cause mortality was lower in the recurrent embolic events group (aHR 0.46, 95% CI [0.26–0.82], p=0.008). No differences were found for major bleeding (p=0.16) or DRT (p=0.72). After PSM, results were unchanged (aHR 3.65, 95% CI [1.31–10.15], p=0.013 for the primary outcome). Compared to estimated annualized event rate in patients with recurrent stroke despite OAC, patients who underwent LAAC had lower annualized event rates (5.2% per patient-year vs 9.7% in RENO-EXTEND and 8.9% in the STR-OAC LAAO control group). Conclusion Patients undergoing LAAC for recurrent embolic events despite anticoagulation had a nearly 3-fold increased risk of embolic recurrence compared with other LAAC indications, consistent after PSM, highlighting the heterogeneity of patients undergoing LAAC for bleeding versus ischaemic indications. While the annualized event rate in this high-risk ischaemic subset appears lower than historical rates for patients maintained on OAC alone, it remains higher than rates reported in other LAAC registries. This variability suggests substantial clinical heterogeneity in this high-risk subset of patients.