Background: Transesophageal echocardiography (TEE) is often performed to explore an embolic source in ischemic stroke patients without atrial fibrillation (AF). The aim of this study was to elucidate the relationship between the ratio of transmitral flow velocity (E) and mitral annular velocity (e’) measured by using TEE (E/e’ TEE) and prediction of new AF during hospitalization.
Method: We prospectively enrolled 170 patients with acute ischemic stroke without known atrial fibrillation at admission. Transthoracic echocardiography was performed to assess left atrial diameter, ejection fraction, and E/e’. A subset of patients underwent TEE to explore an embolic source. We obtained the e’ at the lateral wall located near the left atrial appendage by using TEE. Baseline characteristics, stroke features, initial National Institutes of Health Stroke Scale (NIHSS) score, plasma and serum biomarkers included brain natriuretic protein (BNP), and imaging findings were recorded. We investigated factors to predict new documented AF by multivariate logistic regression analysis. To elucidate the cut off value of factors for predicting new AF, the area under the curve of the receiver operating characteristic curve was evaluated.
Results: AF was detected in 14 patients (8%). In univariate analysis, older age (median, 75 years vs. 71 years; p=0.022), higher initial NIHSS score (median, 8 vs. 3; p=0.017), elevated plasma BNP level (median, 183 pg/mL vs. 36 pg/mL; p<0.001), larger left atrial diameter (median, 38 mm vs. 35 mm; p=0.024) and higher E/e’ TEE (median, 22.4 vs. 14.4; p=0.004) were positively associated with new documented AF. In multivariate analysis, E/e’ TEE was an independent predictor of newly diagnosed AF during hospitalization (Odds ratio, 1.680; 95% confidence interval, 1.174 to 3.420; p<0.001). The cut off value of E/e’ TEE for predicting new AF was 21.5 with the sensitivity of 80 % and the specificity of 98 %.
Conclusions: Diastolic dysfunction evaluated by TEE may be a good predictor of newly diagnosed AF.