Distribution of liver stiffness in non-alcoholic fatty liver disease with higher fibrosis-4 index than low cut-off index

Background and Aim In the condition of high prevalence of non-alcoholic fatty liver disease (NAFLD), a new diagnostic algorithm to efficiently identify NAFLD patients with significant fibrosis is urgently required. We evaluated the predictive ability of the fibrosis-4 index (FIB-4 index) for significant liver fibrosis (F >= 2) in a cohort of Japanese patients with NAFLD. Methods We prospectively calculated the FIB-4 index in patients who were incidentally diagnosed as fatty liver in medical checkups and then conducted liver stiffness measurement by vibration-controlled transient elastography (VCTE) only in patients in whom the FIB-4 index was more than the low cut-off index (> 1.45). Results Of the 5929 people who underwent medical checkups, a total of 1374 people were identified as having fatty liver. Among these, we performed VCTE in 106 patients in whom the FIB-4 index was higher than 1.45. The distribution of the fibrosis stage as estimated by VCTE in the patients was as follows: F0, 52.8%; F1, 10.3%; F2, 21.6%; F3, 11.3%; and F4, 3.7%. The positive predictive value of the FIB-4 index for detecting NAFLD with significant fibrosis was 36.6%. The minimum value of the FIB-4 index was constant for each estimated fibrosis stage. Conclusions This is the first prospective study to evaluate the usefulness of the FIB-4 index as the first step to screen NAFLD patients with significant fibrosis. In Japan, addition of one further step that combined with the FIB-4 index is necessary to meaningfully reduce the number of patients needing liver stiffness measurement or liver biopsy.