Surgery for atrial fibrillation in patients with mitral valve disease: results at five years from the International Registry of Atrial Fibrillation Surgery.

OBJECTIVES: We sought to assess the clinical and survival benefit of atrial fibrillation surgery in patients submitted to mitral valve surgery after stabilization of postoperative rhythm at 1 year. METHODS: One thousand seven hundred twenty-three patients were enrolled. Patients with follow-up of longer than 1 year (n = 972) were divided into 3 groups according to surface electrocardiographic rhythm during follow-up visits: stable sinus rhythm, stable atrial fibrillation, and intermittent rhythms. Adverse cardiac event incidence and predictors of long-term outcome were compared among the 3 groups. RESULTS: In-hospital mortality was 2.6%. Risk factors for mortality were the cut-and-sew technique (odds ratio, 8.92; 95% confidence interval, 1.71-46.50; P = .009) and isolated left atrial procedure (odds ratio, 0.16; 95% confidence interval, 0.04-0.56; P = .004). At 1 year, 63.4% patients were in stable sinus rhythm. Stable sinus rhythm was found to be associated with early and late survival (P = .01, log-rank analysis). Multivariate binary logistic regression analysis found that left atrial dimension (odds ratio, 0.97; 95% confidence interval, 0.96-0.99; P = .005) and concomitant coronary revascularization (odds ratio, 0.48; 95% confidence interval, 0.25-0.92; P = .027) were independent predictors of stable sinus rhythm at 1 year after surgical intervention. At 48 months' follow-up, predictors for stable sinus rhythm were biatrial surgical approach and absence of preoperative permanent atrial fibrillation (odds ratio, 3.56; 95% confidence interval, 1.62-7.83; P < .002). Left atrial size (each millimeter) has a borderline statistical significance (odds ratio, 0.97; 95% confidence interval, 0.93-1.00; P = .065). Thromboembolic events were found to be associated with absence of stable sinus rhythm (P = .010, log-rank analysis). CONCLUSIONS: The achievement of stable sinus rhythm is a predictor of better survival and lower incidence of thromboembolic events. Predictors of stable sinus rhythm were smaller dimensions of the left atrium, biatrial approach, absence of preoperative permanent atrial fibrillation, and absence of concomitant coronary artery bypass grafting.