Abstract
Background: Many noninvasive models based on serum markers composition are used for the assessment of liver fibrosis, reducing the need for liver biopsy. However, most of the models have rarely been validated in Chinese hepatitis B patients. We aim to evaluate and validate chronic hepatitis B(CHB) patients with normal ALT, mildly and significantly elevated ALT levels.Methods: This single-center retrospective study enrolled 285 patients with CHB who underwent liver biopsy. There were 156 patients in normal ALT group, 85 patients in mildly elevated ALT group, and 44 patients in significantly elevated ALT group. The diagnostic accuracy of APRI and FIB-4 was evaluated by areas under the characteristic curves (AUROC) using the histological assessment of the fibrosis stages of the biopsy specimens as the standards.Results: Among 285 patients with CHB, 156 patients had normal ALT level, of which 65 (41.7%) had significant fibrosis(S2-4). The evaluation of significant fibrosis, AUROC in APRI and FIB-4 were 0.608, 0.634, 0.708 and 0.638, 0.679, 0.734 in normal ALT, mildly and significantly elevated ALT, respectively. The assessment of advanced fibrosis, AUROC in APRI and FIB-4 were 0.636, 0.751, 0.708 and 0.652, 0.763, 0.734 in normal ALT, mildly and significantly elevated ALT groups, respectively. Conclusions: APRI and FIB-4 may not be ideal noninvasive markers for evaluating liver fibrosis in Chinese HBV-infected patients with normal ALT levels. Compared with HBV-infected patients with normal ALT, APRI and FIB-4 had high accuracies in diagnosing liver fibrosis in patients with mildly and significantly elevated ALT.