Background and purpose: Paroxysmal atrial fibrillation (PAF) is often difficult to detect in patients with acute ischemic stroke. We aimed to assess the predictive value of a prolonged QT interval corrected for heart rate (QTc) in PAF detection after acute ischemic stroke.
Methods: We enrolled 972 patients with acute ischemic stroke consecutively extracted from our observational stroke registry system. Exclusion criteria were as follows: (1) AF on the initial 12-lead ECG (n=171); (2) previously diagnosed PAF (n=47); and (3) the use of a cardiac pacemaker (n=10). Of the 972 patients, 744 (mean age, 67.6 years; men, 62.6%) were eligible for analysis. The clinical characteristics and 12-lead ECG findings of the patients with and without PAF were compared, and multiple logistic regression analysis was performed to identify predictors of poststroke PAF.
Results: The poststroke cardiac work-up yielded 69 (9.3%) de novo PAF cases among the 744 patients. The QTc interval was significantly longer in patients with PAF than in those without PAF (436 versus 417 ms; P<0.001). Each 10-ms increase in the QTc interval was associated with an increased risk of PAF after multivariate adjustments (odds ratio, 1.41; 95% confidence interval, 1.24-1.61; P<0.001). The optimal threshold value of QTc interval calculated by a receiver-operating characteristic curve was 438 ms, and the area under the curve was 0.73 in this data set.
Conclusions: The QTc interval prolongation is potentially a strong and useful predictor for poststroke PAF.
Keywords: atrial fibrillation; electrocardiography; stroke.
© 2014 American Heart Association, Inc.