Abstract WP135: Sinus Bradycardia as a Potential Predictive Factor for Paroxysmal Atrial Fibrillation in Stroke Patients
Background and Purpose The detection rate of paroxysmal atrial fibrillation (PAF) in stroke patients is limited because they are often asymptomatic, or presenting sinus rhythm on ECG. To estimate the likelihood of PAF as a cause of ischemic stroke, we attempted to identify predictive factors for PAF using the data of sinus heart rate (SHR) on monitoring ECG. Methods We enrolled 711 consecutive patients admitted to our hospital with acute ischemic stroke. Exclusion criteria were (1) persistent AF; (2) cardiac pacemaker; and (3) incomplete clinical investigations. Minimum and mean SHR on 24-hour Holter ECG were obtained. The presence of PAF was judged based on previous history, initial ECG, 24-hour Holter ECG, and cardiac monitoring by inpatient telemetry. The clinical characteristics were compared between patients with and without PAF (PAF and non-PAF group, respectively), and multiple logistic regression analysis was performed to identify predictors for PAF. Results Of all enrolled patients, 577 patients were eligible for analysis, and PAF was confirmed in 110 (19.1%). Clinical parameters showing a significant difference between PAF and non-PAF groups included: age (mean, 74.1 vs. 66.6, P <0.001); dyslipidemia (32.7% vs. 50.8%, P =0.001); lack of intra- or extracranial stenosis (54.0% vs. 79.1%, P <0.001); chronic heart failure (17.3% vs. 4.9%, P <0.001); and NIHSS score (median, 8 vs. 6, P =0.002). Minimum and mean SHR were lower in PAF group than in non-PAF group (46.4 vs. 54.1 bpm, P <0.001; 71.0 vs. 73.8 bpm, P =0.021, respectively). Percentages of patients with PAF were highest in the lowest quartiles of minimum and mean SHR (Figure). Multivariate analysis showed minimum SHR as one of independent predictive factors of PAF (OR 1.08; 95% CI 1.05 to 1.12; P <0.001). Conclusions Low SHR on monitoring ECG can be a novel predictive factor for PAF in ischemic stroke patients.