Diagnostic accuracy of FibroScan-AST score in nonalcoholic steatohepatitis with significant activity and fibrosis / 临床肝胆病杂志

ABSTRACT

Objective To investigate the diagnostic accuracy of FibroScan-AST (FAST) score in patients with high-risk nonalcoholic steatohepatitis (NASH) with a nonalcoholic fatty liver disease (NAFLD) activity score of ≥4 and significant liver fibrosis (F ≥2), along with comparison with other serological models. Methods A total of 84 consecutively admitted patients hospitalized in Ruijin Hospital from January 2015 to December 2020 and biopsy-confirmed NAFLD/NASH were included in this study, and FibroScan (liver stiffness measurement and controlled attenuation parameter) and blood biochemical tests were performed at one week before and after liver biopsy. A Kruskal-Wallis H analysis of variance was used for comparison between multiple groups, and Spearman's correlation coefficient was used to analyze the correlation between variables. The receiver operating characteristic (ROC) curve was plotted with pathological results as the "gold standard", and the area under the ROC curve (AUC) was calculated. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and classification accuracy were calculated based on the cut-off values determined by previous studies. In subgroup analysis, the patients were divided into subgroups based on different clinical indices to evaluate the diagnostic accuracy of each model, which was expressed as AUC (95% confidence interval [ CI ]). Results Among the 84 patients, 43 had high-risk NASH. The FAST score was 0.54(0.04-0.93) for all patients, and the FAST score for liver fibrosis stages F0-F4 was 0.26(0.06-0.73), 0.48(0.04-0.82), 0.61(0.13-0.75), 0.64(0.09-0.93), and 0.82(0.75-0.89), respectively, with a significant difference between stages ( H =23.360, P < 0.001). FAST score was positively correlated with liver fibrosis stage ( r =0.491, P < 0.001). NAFLD fibrosis score (NFS), fibrosis-4 (FIB-4), and aspartate aminotransferase-to-platelet ratio index were positively correlated with liver fibrosis stage ( r =0.230, 0.346, and 0.281, all P < 0.05), with a weaker correlation than FAST score. FAST score had an AUC of 0.725 (95% CI 0.617-0.834, P < 0.001) in evaluating high-risk NASH. According to the low cut-off value determined by previous studies, FAST score ≤0.35 excluded high-risk NASH in 21 patients (25%) with an NPV of 71%; according to the high cut-off value, FAST score ≥0.67 helped to make a confirmed diagnosis of high-risk NASH in 19 patients (22.6%) with a PPV of 74%. NFS and FIB-4 had an AUC of 0.633(95% CI 0.513-0.753) and 0.686(95% CI 0.570-0.803), respectively, in the diagnosis of high-risk NASH ( P < 0.05). Conclusion FAST score can accurately determine the presence or absence of high-risk NASH in NAFLD patients with or without metabolic risk factors, and selection of appropriate cut-off values can help some patients avoid liver biopsy.