Limited reliability of five non-invasive biomarkers in predicting hepatic fibrosis in chronic HCV mono-infected patients opposed to METAVIR scoring

Abstract

BackgroundLiver biopsy is gold standard for fibrosis assessment in hepatitis C virus (HCV) infection but its limitations led to the identification of non-invasive biomarkers. This study assesses the reliability of five biomarkers in estimating the stage of liver fibrosis/cirrhosis in chronic HCV patients versus METAVIR scoring. MethodsOne hundred HCV monoinfected patients who underwent liver biopsy and blood sampling were included. Liver fibrosis was staged (F0–4) and required laboratory tests were performed. AAR, API, APRI, FIB-4 and Pohl score were calculated and their receiver operating curves (ROCs), sensitivities, specificities, predictive values and accuracies were evaluated. ResultsThere were 27, 44, and 29 patients at F0–F1, F2–F3, and F4 groups. Significant statistical differences were found regarding AST, vireamia, platelet count, prothrombin time and all biomarkers. From ROCs only Pohl score predicted significant fibrosis and cirrhosis but with low accuracy. AAR, API and APRI showed moderate performance at low cut-offs, but had limited predictive values or accuracies at higher cut-offs. FIB-4 was the least accurate test. The diagnostic reliability of these biomarkers was limited to patients with suspected insignificant fibrosis. ConclusionsThis study verified the limited reliability for AAR, API, APRI, FIB-4 and Pohl score in estimating the stage of hepatic fibrosis in HCV infected patients opposed to METAVIR scoring.