Noninvasive tests predict liver-related events and mortality in patients with nonalcoholic fatty liver disease: sub-analysis of the CLIONE-Asia study.

Abstract

Noninvasive tests (NITs) have prognostic potential, but whether NITs are comparable with liver biopsy is unclear. This study aimed to examine the prognostic accuracy of NITs for liver-related mortality (LRM) and events (LREs) in patients with biopsy-proven nonalcoholic fatty liver disease (NAFLD). We investigated 1313 patients with NAFLD. Patients were assigned to low-risk, indeterminate-risk, and high-risk groups using conventional cutoff values of each FIB-4 and NAFLD fibrosis score (NFS) and to stage 0-2 and stage 3-4 groups using the fibrosis stage. Survival and Cox regression analyses of the prognostic potential of NITs for LRM/LREs were conducted. During a median follow-up of 4.5years, regarding to FIB-4, the incidence rate (/1000 person-years) in the low risk was zero for LRM and 0.5 for LREs. In contrast, the rate in stage 0-2 was 1.3 for LRM and 2.8 for LRE. The adjusted hazard ratios (aHRs) for LREs in the high risk compared with the low risk were 32.85 (P<0.01). The aHRs in stage 3-4 compared with stage 0-2 were 2.68 (P=0.02) for LREs and 2.26 (P=0.582) for LRM. In the same fibrosis stage, the incidence of LRM/LREs was more frequent with a higher risk stratification. The same trend was observed for NFS. NITs accurately predict LRM and LREs as well as a liver biopsy in Japanese patients with NAFLD. Patients in the low risk may not require close follow-up for at least 5years. The simple NITs could be an acceptable alternative method to performing a liver biopsy for the prognosis of NAFLD.