Methods Subjects with NAFLD who underwent liver biopsy and a valid Fibroscan assessment within 6 months of each BMN 673 supplier other at two tertiary hospitals were retrospectively

evaluated. Biopsies were scored according to the NAFLD Clinical Research Network staging system with F3-4 considered as advanced fibrosis. The diagnostic utility of NFS and LSM were studied seperately, then in combination according to the algorithm suggested by Castera et al. using previously published cut-offs. Results The cohort consisted of 98 adults, (43% male) with a mean (SD) age of 52 (11) years and mean body mass index of 37 (6.5) kg/m2. The prevalence of advanced fibrosis (F3,4) was 17%. NFS and LSM were significantly correlated (Spearman rho=0.35, p<0.001). For predicting advanced fibrosis, the area under the receiver operator characteristic curve (AUROC) of LSM alone was 0.841 (95% CI, 0.762-0.920) and NFS alone was 0.779 (95% CI, 0.663-0.895). Using recommended cut-offs, NFS alone, LSM alone and the sequential combination all had sensitivities and negative predictive values (NPV's) greater than 90% for excluding advanced fibrosis

(see Table). A greater proportion of individuals had advanced fibrosis excluded with the combination algorithm (43%) compared to LSM (31%) or NFS (22%) alone (p<0.001). The specificity of each algorithm for predicting advanced fibrosis was modest (51-79%) and the positive predictive values poor (33-35%). The percentage of subjects correctly classified (true positives plus true negatives) was significantly higher with the combination medchemexpress Deforolimus cell line of NFS plus LSM (65%) compared to LSM (47%) and NFS (27%) (p<0.001). Conclusions The combination of NFS and LSM can reliably and effectively exclude advanced fibrosis in a greater proportion of patients with NAFLD than either method alone. Disclosures: Stuart K. Roberts - Board Membership: Jannsen, Roche, Gilead, BMS Matthew T. Kitson - Consulting: MSD, Roche; Grant/Research Support: MSD; Speaking and Teaching: Janssen-Cilag Gary P. Jeffrey - Advisory Committees or Review Panels: MSD, Novartis The following people have nothing to disclose: Edric Hee, William

W. Kemp, Bastiaan de Boer, Jeffrey M. Hamdorf, Gerry C. MacQuillan, George Garas, Helena Ching, Ross Mac Nicholas, Leon Adams BACKGROUND: Nonalcoholic fatty liver disease (NAFLD) may result in variable spectrum of cardiac abnormalities. The aim of this study was to evaluate whether NAFLD is associated with left ventricular diastolic dysfunction independently of other risk factors. METHODS: A total of 3,306 subjects who underwent health check-up including echocardiography between January 1 and December 31, 2010 at a single health screening center in South Korea were enrolled. Diastolic dysfunction was diagnosed and graded using echocardiographical parameters including E/A ratio, e′/a′, deceleration time, and peak e′.