Abstract
OBJECTIVE: Societal guidelines support the concomitant surgical ablation of atrial fibrillation in patients undergoing cardiac surgery. Recent evidence has highlighted the stroke reduction of left atrial appendage obliteration with or without surgical ablation in similar populations. To inform clinical decision-making, we evaluated real-world outcomes of patients with atrial fibrillation undergoing cardiac surgery by comparing no atrial fibrillation management with left atrial appendage obliteration alone versus surgical ablation + left atrial appendage obliteration. METHODS: By using the US Centers for Medicare and Medicaid Services inpatient claims database, we evaluated all beneficiaries aged 65 years and older with a diagnosis of atrial fibrillation undergoing coronary artery bypass grafting or mitral/aortic/tricuspid valve repair or replacement between January 2018 and December 2020. Diagnosis-related group and International Classification of Diseases, 10th Revision procedure codes were used to define variables. Risk adjustment was performed with regression analysis using inverse probability weighting of propensity scores and Cox proportional hazards models. Subgroup analyses stratified patients by primary operation and paroxysmal or persistent atrial fibrillation. RESULTS: A total of 103,382 patients with preoperative atrial fibrillation were stratified by surgical ablation + left atrial appendage obliteration (10,437; 10.1%), left atrial appendage obliteration alone (12,901; 12.5%), or no atrial fibrillation management (80,044; 77.4%). Patients with persistent atrial fibrillation (21,076; 20.4%) received the highest proportion of surgical ablation + left atrial appendage obliteration (4661 19.4%) and left atrial appendage obliteration alone (3%724%; 15.4%) versus no atrial fibrillation management (15,688; 65.2%). Likewise, patients undergoing open atrial operations (mitral/tricuspid; 17,204; 16.6%) had higher proportions of atrial fibrillation treatment (surgical ablation + left atrial appendage obliteration 5267 30.6%; left atrial appendage obliteration alone 4259 24.8%; no atrial fibrillation management 7678 44.6%). After robust risk adjustment, surgical ablation + left atrial appendage obliteration was independently associated with reduced 3-year mortality compared with no atrial fibrillation treatment (hazard ratio, 0.68, P < .001) and left atrial appendage obliteration alone (hazard ratio, 0.90, P < .001). Compared with no atrial fibrillation treatment, readmissions for embolic stroke were lower with both surgical ablation + left atrial appendage obliteration (hazard ratio, 0.77, P = .009) and left atrial appendage obliteration alone (hazard ratio, 0.73, P < .001). Reduction in 3-year composite mortality or stroke after surgical ablation + left atrial appendage obliteration was superior to left atrial appendage alone (hazard ratio, 0.90, P = .035). CONCLUSIONS: In Medicare beneficiaries with atrial fibrillation undergoing cardiac surgery, the surgical management of atrial fibrillation was associated with lower 3-year mortality and readmission for stroke, with surgical ablation + left atrial appendage obliteration being associated with higher survival compared with left atrial appendage obliteration alone.