Noninvasive scoring systems in patients with nonalcoholic fatty liver disease with normal alanine aminotransferase levels

Abstract

Background: The severity of liver fibrosis must be estimated to determine the prognosis, for surveillance, and for optimal treatment of nonalcoholic fatty liver disease (NAFLD). However, the severity of hepatic fibrosis tends to be underestimated in patients with normal ALT.

Methods: We investigated histological data and scoring systems (FIB-4 index, NAFLD fibrosis score, BARD score, and AST/ALT ratio) of 1,102 liver-biopsy-confirmed NAFLD patients.

Results: A total of 235 NAFLD patients with normal ALT were estimated to exist. The ratio of advanced fibrosis (stage 3-4) was seen in 16.1 % of subjects with normal ALT. Scoring systems, especially the FIB-4 index and NAFLD fibrosis score, were clinically very useful (AUROC >0.8), even in patients with normal ALT. Furthermore, with resetting of the cutoff values, the FIB-4 index (>1.659) and NAFLD fibrosis score (>0.735) were found to have a higher sensitivity and higher specificity for the prediction of advanced fibrosis, and all of these scoring systems (FIB-4 index, NAFLD fibrosis score, BARD score, and AST/ALT ratio) had higher negative predictive values (>90.3 %). By using the resetting cutoff value, liver biopsy could have been avoided in 60.4 % (FIB-4), 66.4 % (NAFLD fibrosis score), 51.9 % (BARD score), and 62.1 % (AST/ALT ratio).

Conclusions: We reset the cutoff values of numerous non-invasive scoring systems to improve their clinical usefulness in the prediction of liver fibrosis in NAFLD patients with normal ALT, and these non-invasive scoring systems with the reset cutoff values could be of substantial benefit to reduce the number of liver biopsies performed.