OBJECTIVE To identify predictors of left atrial appendage stunning after the use of electrical cardioversion to restore sinus rhythm in patients with non-valvular atrial fibrillation. METHODS A total of 68 consecutive patients (45 men, 23 women, 60.5 +/- 8.7 years of age) with non-valvular atrial fibrillation undergoing electrical cardioversion were enlisted in this study. Clinical and echocardiographic variables were analyzed by univariate regression and multivariate logistic regression to investigate the relationship between occurrences of left atrial appendage stunning and these factors. RESULTS Univariate analysis revealed that, in comparing patients without and with left atrial appendage stunning, there were significant differences in the duration of atrial fibrillation > 8 weeks (32.3% vs 75.5%, P < 0.001), left atrial diameter > 50 mm (29.0% vs 54.1%, P < 0.05), left atrial emptying fraction (31.5% +/- 7.8% vs 27.1% +/- 8.5%, P < 0.05), left ventricular ejection fraction < 50% (38.7% vs 67.6%, P < 0.05), maximum electrical energy (96.8 J +/- 65.8 J vs 156.8 J +/- 100.8 J, P < 0.01), cumulative electrical energy 146.8 J +/- 142.6 J vs 290.5 J +/- 242.1 J, P < 0.01) and number of electrical cardioversion shocks (1.7 +/- 0.9 vs 2.43 +/- 1.20, P < 0.05). However, backward stepwise multivariate logistic regression analysis identified as significant and independent predictors of left atrial appendage stunning only duration of atrial fibrillation > 8 weeks (OR = 7.249, 95% CI = 1.998 - 26.304, P < 0.01), left atrial diameter > 50 mm (OR = 3.896, 95% CI = 1.105 - 13.734, P < 0.05), left ventricular ejection fraction < 50% (OR = 4.465, 95% CI = 1.51713.140, P < 0.01) and cumulative energy of electrical cardioversion (OR = 1.004, 95% CI = 1.000 - 1.008, P < 0.05). CONCLUSIONS Duration of atrial fibrillation > 8 weeks, left atrial diameter > 50 mm, left ventricular ejection fraction < 50%, and cumulative energy of electrical cardioversion are independent predictors of left atrial appendage stunning. Anticoagulation treatment should be individualized for patients undergoing electrical cardioversion to reduce the risk of both cardioversion-related thromboembolic events and hemorrhagic complications caused by warfarin treatment.