Abstract
OBJECTIVES: To investigate the predictive values of surface electrocardiogram-derived parameters in patients with atrial fibrillation who underwent thoracoscopic epicardial ablation. METHODS: The present study included 102 patients with atrial fibrillation who underwent thoracoscopic epicardial ablation and whose baseline 12-lead electrocardiograms were available. Frequency domain analysis was performed to calculate the electrocardiogram-derived parameters. Cox proportional hazards regression was used to identify predictive risk factors for postoperative recurrence. RESULTS: At 36-month interval, the overall rate of freedom from atrial tachyarrhythmia without antiarrhythmic drugs was 58.5%. The results of the univariable and multivariable analyses showed that larger left atrial diameter (hazard ratio: 1.074, 95% confidence interval: 1.021-1.130, P = 0.006) was an independent risk factor for atrial fibrillation recurrence, while higher fibrillatory wave amplitude was a protective factor (hazard ratio: 0.292, 95% confidence interval: 0.157-0.542, P < 0.001). The associations were clarified by the restricted cubic splines. The dominant frequency and organization index were not identified as statistically significant parameters. CONCLUSIONS: The fibrillatory wave amplitude has the highest predictive value for atrial fibrillation recurrence in electrocardiogram-derived parameters. Together with left atrial diameter, it may help identify patients in whom thoracoscopic ablation is likely to be effective.