Abstract 17804: Echocardiographic and Radiologic Predictors of Atrial Fibrillation in Acute Ischemic Stroke

Abstract

Introduction: Atrial fibrillation (AF) is an important risk factor for stroke. Approximately 15-25% of cryptogenic ischemic strokes (CIS) are due to AF. Hypothesis: Patients with echocardiographic markers including left atrial dilatation (LAD) and valvular disease and/or radiologic findings including a hyper-dense vessel sign (HDV sign), prior cortical or cerebellar stroke, and hemorrhagic transformation (HT) are more likely to have stroke due to AF. Methods: This is a single center, retrospective case control study of patients admitted with acute ischemic stroke (AIS) from 2008-2013. Patients with a new diagnosis of stroke and a defined stroke etiology (large vessel, cardio-embolic, small vessel and other, TOAST classification) were included. CIS patients or those without imaging/echocardiographic studies were excluded. Images were reviewed for prior stroke, HT, a HDV sign and stroke location (left or right anterior, posterior, multifocal). Echocardiograms were reviewed for LAD, ejection fraction<35%, moderate-severe mitral regurgitation (MR), moderate-severe aortic regurgitation, mitral stenosis, aortic stenosis, left ventricular segmental akinesis, diastolic dysfunction, LVH and patent foramen ovale. Results: Charts of 600 patients were reviewed and 383 patients were included in the analysis. 185 patients had AF as the etiology of their stroke. The remainder had no known AF. Univariate and multivariate logistic regression was performed using SPSS. LAD, moderate to severe MR, HDV sign, and HT were found to accurately predict AF in patients with AIS. LAD had an OR of 6.9 (CI 4.4-10.8, p<.0001) and HDV predicted AF with an OR of 2.2 (CI 1.2-4.2, p<.016) even after controlling for LAD, MR, and HT. Conclusion: Echocardiographic and radiologic markers help predict AF as a stroke etiology in AIS patients. This is the first study, to the best of our knowledge, to show that HDV sign is a positive predictor of AF. Prospective studies with larger cohorts are needed to confirm these findings in CIS patients. These markers could be added to demographic information to enhance prediction of occult AF and the need for chronic cardiac monitoring.