Validation of APRI and FIB-4 score in an Antwerp cohort of chronic hepatitis C patients

Background and aims : Evaluation of liver fibrosis in chronic hepatitis C patients guides clinical decision-making. The aim of this study is to validate APRI and FIB-4, two easily calculated non- invasive tests to predict fibrosis, in chronic HCV patients using biopsy as a gold standard and to compare accuracy between HCV monoinfected and HIV/HCV coinfected patients. Patients and methods : We retrospectively studied HCV patients of two centres who underwent liver biopsy. Liver fibrosis was staged according to METAVIR. Results : 136 patients were included. The AUROC of FIB-4 (0.896) to discriminate F0-F2 vs. F3-F4 was significantly higher (p = 0.0186) than the AUROC of APRI (0.842). The difference in AUROC between HIV-negative and positive patients was not significant for APRI (p = 0.471), nor for FIB-4 (p = 0.495). Perfor- mance status was lower in HIV-positive patients with 46.7% and 69.0% of patients correctly classified using APRI and FIB-4, com- pared to 56.6% and 73.6% in HIV-negative patients, respectively. Conversion of transaminase values from one hospital to the other did not significantly change the AUROC of FIB-4 (p = 0.928). Conclusions : APRI and FIB-4 have a better performance status in HCV monoinfected patients compared to HIV/HCV coinfected patients. FIB-4 has a better AUROC compared to APRI and is the preferred noninvasive fibrosis score to discriminate between F0-F2 and F3-F4. Different hospitals should use their local absolute serum transaminase values without conversion. (Acta gastroenterol. belg., 2015, 78, 373-380).