Atrial time and voltage dispersion are both needed to predict new-onset atrial fibrillation in ischemic stroke patients

Background: Atrial fibrillation (AF) is a known risk factor for ischemic stroke. Electrocardiographic predictors of AF in population studies such as the Framingham Heart Study, as well as in hypertensive patients have demonstrated a predictive value of the P-wave duration for development of AF. QRS vector magnitude has had a predictive value in ventricular arrhythmia development. We aimed to assess the value of the three-dimensional P-wave vector magnitude and its relationship to P-wave duration for prediction of new-onset AF after ischemic stroke. Methods: First-ever ischemic stroke patients without AF at inclusion in the Lund Stroke Register were included. Measurements of Pwave duration (Pd), QRS duration, corrected QT interval, and PQ interval were performed automatically using the University of Glasgow 12-lead ECG analysis algorithm. The P-wave vector magnitude (Pvm) was calculated automatically as the square root of the sum of the squared P-wave magnitudes in leads V6, II and one half of the P-wave amplitude in V2(root PV6(2) + PII2 + (0.5*PV2)(2)), based on the P-wave magnitude (Pvm) as defined by the visually transformed Kors' Quasi-orthogonal method. Results: The median age was 73 (IQR 63-80) years at stroke onset (135 males, 92 females). Multivariate predictors of new-onset atrial fibrillation included age > 65 years, hypertension, and Pd/Pvm. A cut-off value of 870 ms/mV gave sensitivity, specificity, positive and negative predictive values of 51, 79, 30 and 87%, respectively. The Pd/Pvm was the only ECG predictor of AF with a significant multivariate hazard ratio of 2.02 (95% CI 1.18 to 3.46, p = 0.010). Conclusion: P-wave dispersion as measured by the Pd/Pvm was the only ECG parameter measured which independently predicted subsequent AF identification in a cohort of stroke patients. Further prospective studies in larger cohorts are needed to validate its clinical usefulness.