Rhythm course over 5 years following surgical ablation for atrial fibrillation.

OBJECTIVES Reporting methods for surgical ablation (SA) of atrial fibrillation (AF) were standardized by the Heart Rhythm Society Guidelines, stating that results should be reported only for the first 2 years following SA. The purpose of this study was to assess the outcome of SA over 5 years and determine predictors for success over that period. METHODS Data were collected prospectively for all SA (n=787). Rhythm was verified by electrocardiogram and Holter monitoring at 3, 6, 9, 12, 18 and 24 months and yearly thereafter. Patients with rhythm status available at 2 and 5 years were included in the main analyses (n=137). Multivariate logistic regression was used for predictors of normal sinus rhythm (NSR). Receiver operating curves compared 2- and 5-year predicted probability against observed rhythm status by year. RESULTS Return to NSR at 2 years was 88% (80% off antiarrhythmic drugs) and at 5 years was 85% (71% off antiarrhythmic drugs). The majority of patients (64%) had stable NSR over 5 years. The only predictor for 2-year NSR was smaller left atrial size (odds ratio [OR]=0.40, P=0.044). Predictors for 5-year NSR were smaller left atrial size (OR=0.28, P=0.002), age (OR=0.91, P=0.031) and length of hospital stay (OR=0.85, P=0.026). CONCLUSIONS This study demonstrated stable results of SA for AF over time with somewhat different predictors for 2- and 5-year NSR in a group of patients with complete follow-up at both time points. Accurate models to determine predictors for success of SA more than 2 years after surgery are essential to better understand long-term outcome for patients with AF.