Background and purpose: Electrocardiographic left atrial abnormality has been associated with stroke independently of atrial fibrillation (AF), suggesting that atrial thromboembolism may occur in the absence of AF. If true, we would expect an association with cryptogenic or cardioembolic stroke rather than noncardioembolic stroke.
Methods: We conducted a case-cohort analysis in the Northern Manhattan Study, a prospective cohort study of stroke risk factors. P-wave terminal force in lead V1 was manually measured from baseline ECGs of participants in sinus rhythm who subsequently had ischemic stroke (n=241) and a randomly selected subcohort without stroke (n=798). Weighted Cox proportional hazard models were used to examine the association between P-wave terminal force in lead V1 and stroke etiologic subtypes while adjusting for baseline demographic characteristics, history of AF, heart failure, diabetes mellitus, hypertension, tobacco use, and lipid levels.
Results: Mean P-wave terminal force in lead V1 was 4452 (±3368) μV*ms among stroke cases and 3934 (±2541) μV*ms in the subcohort. P-wave terminal force in lead V1 was associated with ischemic stroke (adjusted hazard ratio per SD, 1.20; 95% confidence interval, 1.03-1.39) and the composite of cryptogenic or cardioembolic stroke (adjusted hazard ratio per SD, 1.31; 95% confidence interval, 1.08-1.58). There was no definite association with noncardioembolic stroke subtypes (adjusted hazard ratio per SD, 1.14; 95% confidence interval, 0.92-1.40). Results were similar after excluding participants with a history of AF at baseline or new AF during follow-up.
Conclusions: ECG-defined left atrial abnormality was associated with incident cryptogenic or cardioembolic stroke independently of the presence of AF, suggesting atrial thromboembolism may occur without recognized AF.
Keywords: atrial fibrillation; cardiomyopathies; cohort studies; embolism; stroke.
© 2015 American Heart Association, Inc.