Left Atrial Appendage Closure During Cardiac Surgery is Safe and Underutilized in California
Joseph Hadaya1, Yas Sanaiha1, Beate Danielsen2, Joseph Carey3, *Richard J. Shemin1, *Peyman Benharash1
1Cardiac Surgery, David Geffen School of Medicine at UCLA, Los Angeles, California, United States, 2Health Information Solutions, Rocklin, California, United States, 3Cardiothoracic Surgery, University of California Irvine Medical Center, Orange, California, United States
Objective: Left atrial appendage (LAA) closure has been associated with reduced long-term risk of thromboembolism among patients with atrial fibrillation (AF). In 2017, the Society of Thoracic Surgeons recommended LAA closure at the time of cardiac surgery in patients with preexisting AF. The present work evaluated recent trends and the safety profile of concomitant LAA closure in a statewide cohort of cardiac surgical patients with AF. We further examined the association of LAA closure with stroke or arterial thromboembolism and tested whether institutional practice patterns influence its use.
Methods: Adults undergoing coronary artery bypass grafting (CABG) and/or valve surgery were identified in the 2016-2019 California Office of Statewide Health Planning and Development databases. Preoperative AF was defined as paroxysmal, persistent, or permanent AF present prior to the surgical hospitalization. The primary endpoints were ischemic stroke or arterial thromboembolism at the index hospitalization or within 1 year from discharge. We secondarily evaluated for the need for blood transfusion and surgical re-exploration. Two-level hierarchical regression models were fit to study the association of LAA closure with outcomes. Hospital-level variation was studied using intraclass correlation coefficients and random intercepts were estimated using Bayesian approaches.
Results: During the study period, 87,928 patients underwent CABG or valve surgery at 127 hospitals, among which 21.0% had preoperative AF. Among patients with AF, rates of LAA closure were 34.6% for isolated CABG, 43.9% for AVR±CABG, 65.5% for MVR±CABG, and 64.9% for tricuspid, pulmonic, or multi-valve cases. Over time, concomitant LAA closure significantly increased for all operative categories (Figure 1). Patients receiving LAA closure were older (69.7 vs 69.1 years, p<0.001), more commonly female (32.0% vs 29.4%, p<0.001), and had a lower Charlson Comorbidity Index (2.5 vs 2.9, p<0.001). After risk adjustment, LAA closure was associated with reduced odds of stroke at the index hospitalization (0.76, 95% confidence interval, CI, 0.60-0.95) but similar odds of other arterial thromboembolism (1.19, 95% CI 0.75-1.90), blood transfusion (0.97, 0.90-1.05) and re-exploration (0.98, 95% CI 0.76-1.27). At 1 year, patients who received LAA closure had lower rates of stroke (3.6% vs 3.0%, p=0.01). A significant degree of risk-adjusted hospital level variation in LAA closure was evident (Figure 2), with 17.8% of total variation explained by the operating hospital.
Conclusions: In a contemporary, real-world cohort, LAA closure was associated with minimal surgical morbidity, and protected against short- and mid-term risk of stroke. Although the overall use of LAA closure has increased, substantial variation exists among surgical programs in California, suggesting the need for further standardization of care.
Back to 2022 Abstracts