Predictors of sinus rhythm return during defibrillation testing in patients with permanent atrial fibrillation undergoing implantation of a cardioverter-defibrillator.

Background: Atrial fibrillation (AF) is present in a significant proportion of patients treated with an implantable cardioverter-defibrillator (ICD). Defibrillation testing may lead to sinus rhythm (SR) restoration which may be hazardous due to the increased risk of thromboembolic complications in these patients.

Aim: To identify predictors of SR restoration during defibrillation testing in patients with permanent AF undergoing ICD implantation.

Methods: Permanent AF was present in 79 (12%) of 671 consecutive patients who received ICD in our institution between 2005 and 2010. In this group, 47 patients (mean age 64 ± 12 years, 38 males) underwent defibrillation testing during the implantation procedure and in the remaining 32 patients defibrillation testing was not performed due to various contraindications. Sinus rhythm was restored in 17 (36%) patients, while AF was still present after defibrillation testing in the remaining 30 patients. We analysed demographic, clinical, echocardiographic and electrophysiological parameters which could identify those patients in whom SR was restored, using univariate and multivariate analysis as well as constructing ROC curves.

Results: Demographic parameters and clinical history were similar in both groups. Patients in whom SR was restored had smaller left atrial diameter (49.9 ± 5.9 vs 59.4 ± 6.1 mm, p 〈 0.001), more often received a double-coil defibrillation lead (83% vs 10%, p 〈 0.001), had less advanced heart failure as assessed using NYHA classification (p 〈 0.05), and were more frequently treated with amiodarone (47% vs 23%, p 〈 0.025). The chance of SR return was increased 11 times in patients receiving amiodarone, 6 times in patients with a double-coil lead, 2 times in patients with lower NYHA class, and 1.36 times in patients with smaller left atrium diameter (for each 1 mm increase). The ROC curves showed that using the cut-off value for the left atrial diameter of 47 mm, patients prone to SR restoration were identified with a sensitivity of 65%, specificity of 100%, positive predictive value of 83%, and negative predictive value of 90% (area under curve 0.904, 95% CI 0.809-1.0). Multivariate analysis showed that NYHA class, amiodarone usage, type of defibrillating lead and left atrial diameter were independent predictors of SR restoration.

Conclusions: Atrial fibrillation was present in 12% of consecutive patients undergoing ICD implantation and was terminated by defibrillation testing in 36% of those who underwent this test. The NYHA class, amiodarone usage, type of defibrillating lead and left atrial diameter were independent predictors of SR restoration.