Clues on electrocardiography to predict the presence of paroxysmal atrial fibrillation in patients with acute ischemic stroke: A propensity score-matched study

Objectives: In this study, we aimed to detect surface electrocardiography (ECG) markers that could be predictive of paroxysmal atrial fibrillation (PAF) attacks in patients with ischemic stroke. Patients and methods: Between November 2017 and April 2021, a total of 112 patients (65 males, 47 females; mean age: 70.5±6.8 years; range, 51 to 84 years) hospitalized for acute ischemic stroke with sinus rhythm on surface ECG who underwent Holter ECG monitoring for PAF were retrospectively analyzed. The patients were divided into two groups of 56 patients in each (Group 1: those with PAF on Holter ECG and Group 2: those without PAF). Both groups were matched according to demographic, clinical, and echocardiographic features using the propensity score matching method. Results: Demographic, clinical, and echocardiographic features were similar between groups (p>0.05). The mean maximum P-wave duration (PWD) and P-wave dispersion (PWDis) were longer in Group 1 than Group 2 (108.4±9.9 vs. 102.5±10.2 ms; p=0.002, 49.4±13.6 vs. 36.8±11.7 ms; p<0.001). Similarly, the mean P-wave terminal force in lead V1 (PTFV1) was higher in Group 1 than Group 2 (4415±909 vs. 3826±568 μV·ms; p<0.001). Logistic regression analysis revealed high PWDis (odds ratio [OR]: 1.164; 95% confidence interval [CI]: 1.069-1.268; p<0.001) and PTFV1 (OR: 1.156; 95% CI: 1.065-1.254; p=0.001) as independent predictors of PAF. Conclusion: PWDis and PTFV1 are independent predictors of PAF in patients with acute ischemic stroke. These simple and easily accessible predictors that can be detected via surface ECG may be used as a guide to identify patients who require longer rhythm monitoring to better detect occult PAF, thereby preventing recurrent strokes.