Minimally invasive epicardial surgical left atrial appendage exclusion for atrial fibrillation patients at high risk for stroke and for bleeding

BACKGROUND Atrial fibrillation (AF) patients at high risk for stroke and for bleeding may be unsuitable for either oral anticoagulation or endocardial left atrial appendage (LAA) occlusion. However, minimally invasive, epicardial left atrial appendage exclusion (LAAE) may be an option. OBJECTIVE The purpose of this study was to evaluate outcomes of LAAE in high-risk AF patients not receiving oral anticoagulation. METHODS A retrospective analysis of Medicare claims data was conducted to evaluate thromboembolic events in AF patients who underwent LAAE compared to a 1:4 propensity score-matched group of patients who did not receive LAAE (control). Neither group was receiving any oral anticoagulation at baseline or follow-up. Fine-Gray models estimated hazard ratios and evaluated between-group differences. Bootstrapping was applied to generate 95% confidence intervals (CIs). RESULTS The LAAE group (n = 243) was 61% male (mean age 75 years). AF was nonparoxysmal in 70% (mean CHA2DS2-VASc score 5.4; mean HAS-BLED score 4.2). The matched control group (n = 972) had statistically similar characteristics. One-year adjusted estimates of thromboembolic events were 7.3% (95% CI 4.3%-11.1%) in the LAAE group and 12.1% (95% CI 9.5%- 14.8%) in the control group. Absolute risk reduction was 4.8% (95% CI 0.6%-8.9%; P = .028). Adjusted hazard ratio for thromboembolic events for LAAE vs non-LAAE was 0.672 (95% CI 0.394-1.146). CONCLUSION In AF patients not taking oral anticoagulation who are at high risk for stroke and for bleeding, minimally invasive, thoracoscopic, epicardial LAAE was associated with a lower rate of thromboembolic events.