Abstract WP380: Diastolic Dysfunction Predicts In-Hospital Death in Stroke and Transient Ischemic Attack Patients with Atrial Fibrillation
Background: The aim of this study is to identify whether diastolic dysfunction predicts in-hospital death in stroke and transient ischemic attack (TIA) patients with atrial fibrillation.
Method: Consecutive ischemic stroke patients within 24 hours of onset were analyzed. Then, we enrolled stroke and TIA patients with atrial fibrillation. All patients underwent MRI or CT and blood test on admission and transthoratic echocardiography within 24 hours of admission. Diastolic filling pressure was estimated as the ratio of early transmitral flow velocity (E) to mitral annular velocity (e’) on transthoracic echocardiography. Plasma brain natriuretic peptide and D-dimer were measured on admission. We calculated initial diffusion weighted imaging (DWI) lesion volume and assessed Albert Stroke Program Early CT score of DWI (ASPECTS-DWI). Stroke severity was assessed by the National Institute of Health Stroke Scale (NIHSS) score. Patients were divided into two groups: the deceased group and the survival group. Univariate and multivariate analysis were conducted to identify predictors of in-hospital death.
Results: Four-hundred sixty three patients with ischemic stroke were analyzed. Of these, 204 patients were categorized into stroke subtype of cardioembolic infarction with atrial fibrillation. Thirty-two patients (16%) could not undergo MRI due to pacemaker implantation. The median length of stay in hospital was 19 days (interquartile range 15 - 27). Twenty five patients (12%) deceased during hospitalization. Patients with the deceased group had higher NIHSS score (median, 20 versus 8, p<0.001), D-dimer level (median, 2.8 versus 1.7 mg/dl, p=0.019), DWI lesion volume (median, 197 versus 7 ml, p<0.001), ASPECTS-DWI (median, 3 versus 9, p<0.001), and E/e’ (median, 25.2 versus 13.6, p<0.001) than patients with the survival group. In multivariate analysis, higher E/e’ was an independent predictor of in-hospital death (OR 0.711, 95% CI 0.52 to 0.98, p=0.036). The cutoff value of E/e’ for prediction in-hospital death was 20 with the sensitivity of 100% and specificity of 88%.
Conclusion: Diastolic dysfunction may be associated with in-hospital death in stroke and TIA patients with atrial fibrillation.
- © 2012 by American Heart Association, Inc.