Diagnostic accuracy of APRI and FIB-4 for predicting hepatitis B virus-related liver fibrosis accompanied with hepatocellular carcinoma

Background Aspartate aminotransferase to platelet ratio index (APRI) and the fibrosis index based on four factors (FIB-4) are the two most focused non-invasive models to assess liver fibrosis. Aims We aimed to examine the validity of these two models for predicting hepatitis B virus (HBV)-related liver fibrosis accompanied with hepatocellular carcinoma (HCC). Methods We enrolled HBV-infected patients with liver cancer who had received hepatectomy. The accuracy of APRI and FIB-4 for diagnosing liver fibrosis was assessed based on their sensitivity, specificity, diagnostic efficiency, positive predictive value (PPV), negative predictive value (NPV), kappa (κ) value and area under the receiver-operating characteristic curve (AUC). Results Finally 2176 patients were included, with 1682 retrospective subjects and 494 prospective subjects. APRI (rs = 0.310) and FIB-4 (rs = 0.278) were positively correlated with liver fibrosis. And χ2 analysis demonstrated that APRI and FIB-4 values correlated with different levels of liver fibrosis with all P values less than 0.01. The AUC values for APRI and FIB-4 were 0.685 and 0.626 (P = 0.73) for predicting significant fibrosis, 0.681 and 0.648 (P = 0.81) for differentiation of advanced fibrosis and 0.676 and 0.652 (P = 0.77) for diagnosing cirrhosis. Conclusion APRI and FIB-4 correlate with liver fibrosis. However these two models have low accuracy for predicting HBV-related liver fibrosis in HCC patients.