Abstract 14839: Association of Silent Brain Infarction With CHADS2 Score and Left Atrial Abnormalities in Patients With Nonvalvular Atrial Fibrillation

Abstract

Introduction: The CHADS2 score is widely used for risk stratification of thromboembolism in patients with nonvalvular atrial fibrillation (NVAF). Furthermore, LA abnormalities detected by transesophageal echocardiography (TEE) are also well known risk factors for thromboembolism in NVAF patients. On the other hand, silent brain infarction (SBI) relates to subsequent symptomatic brain infarction; however little is known regarding association among CHADS2 score, TEE parameters and SBI. Methods: The study population consisted of 103 neurologically asymptomatic patients with NVAF (mean age 63 ± 10 years) who underwent TEE. All participating patients underwent brain MRI and were calculated CHADS2 score. SBI was defined as an area of hypointense lesions that measured >3 mm on T1-weighted images and hyperintense lesions on T2-weighted images by brain magnetic resonance imaging (MRI). LA abnormalities were defined as LA thrombus, spontaneous echo contrast, or low LA appendage emptying velocity (<20 cm/s) detected by TEE. Results: Of 103 patients, 31 (30%) showed SBI on brain MRI. Over all mean CHADS2 score was 1.1 ± 0.9, and LA abnormalities were detected in 24 patients by TEE (23%). Mean CHADS2 score in patients with SBI was significantly higher than that in patients without SBI (P<0.05). The prevalence of SBI increased along with increases in CHADS2 score (CHADS2 score of 0, 1, 2 and 3-6 in 13%, 23%, 47% and 80%, respectively, P<0.05). Patients with SBI had a higher prevalence of LA abnormalities than those without SBI (P<0.05). In patients with SBI with CHADS2 score of 0-1, LA abnormalities were detected in 5 of 14 patients (36%). Conclusions: In this study, both CHADS2 score and prevalence of LA abnormalities by TEE were higher in patients with SBI on MRI. LA abnormalities detected by TEE may discriminate risk for SBI even in patients with low CHADS2 score.