Prediction of atrial fibrillation after ischemic stroke using P-wave signal averaged electrocardiography

Background: Atrial fibrillation (AF) is highly prevalent in patients with ischemic stroke, but the diagnosis is often difficult. Methods: This study consisted of 68 stroke patients in sinus rhythm without history of AF. All patients underwent P-wave signal-averaged electrocardiography (P-SAECG), echocardiography, 24-h Holter monitoring, and measurement of plasma B-type natriuretic peptide (BNP) concentrations at admission. Results: An abnormal P-SAECG was found in 34 of 68 stroke patients. In the follow-up period of 11 +/- 4 months, AF developed in 17 patients (AF group). The remaining 51 patients were classified as the non-AF group. The prevalence of atrial late potentials (ALP) on P-SAECG, and the number of premature atrial contractions (PACs) were significantly higher in the AF group than those in the non-AF group (88.2% vs 37.3%; p < 0.001, 149 +/- 120 vs 79 +/- 69; p = 0.030, respectively). However, there were no significant differences in age, left atrial dimension, or BNP concentrations between both groups. Cox proportional hazards analysis revealed that the presence of ALP (risk ratio 11.15; p = 0.002) and frequent PACs (more than 100/24 h) (risk ratio 4.53; p = 0.007) had significant correlation to the occurrence of AF. Conclusions: ALP may be a novel predictor of AF in stroke patients. P-SAECG should be considered in stroke of undetermined etiology. (c) 2012 Japanese College of Cardiology. Published by Elsevier Ltd. All rights reserved.