Abstract WP233: Epicardial Fat Thickness, Free Fatty Acid and Pro-BNP Predict Atrial Fibrillation in Acute Ischemic Stroke

Background/Objectives: Detection of atrial fibrillation (AF) is important for diagnosis of cardioembolic stroke. Few studies that cardioembolic stroke relate to some biomarkers such as high sensitive C-reactive protein (hs-CRP), free fatty acid (FFA), pro-brain natriuretic peptide (pro-BNP) and D-dimer have been reported. FFAs are major components of epicardial fat. Increased epicardial fat thickness (EFT) is known to be associated with the presence of AF. The object of this study is to find the useful markers to distinguish stroke with AF in acute ischemic stroke. Design/Method: Total 214 consecutive patients (mean age, 66.8±12.3 years; 39.7% women; 16.4% AF group) with acute ischemic stroke within 72 hours of onset were retrospectively enrolled. We analyzed the EFT and the several biomarkers between stroke with and without AF. Results: Age, heart rate, FFA, pro-BNP, hs-CRP, HDL and the frequency of woman were significantly higher in AF group (p <0.05). Estimated glomerular filtration rate (eGFR) was significantly lower in AF group (p<0.05). D-dimer and myoglobin were no significant difference. EFT (6.45±1.18 vs 5.30±1.23 mm, p<0.001), Left atrium (LA) diameter and LA volume were significantly higher in AF group. The respective cut-off value of EFT, FFA and pro-BNP for prediction of the AF were 5.58mm, 1226.5 μEq/ml and 295.0 pg/ml (sensitivity 85%, specificity 61%, AUC 0.76 for EFT; 67%, 93%, 0.79 for FFA; 82%, 84%, 0.88 for pro-BNP). Multivariate logistic regression analysis demonstrated that age, FFA and EFT were independently associated with the presence of AF (p<0.05). Conclusions: The combination of EFT, FFA and pro-BNP can be a useful marker for the prediction of the AF in acute ischemic stroke.