摘要

Background: Hepatitis B virus (HBV) infection is a serious public health problem worldwide. This study aimed to investigate the relationship between serum alpha-fetoprotein (AFP) levels and pathological stages of liver biopsy in patients with chronic hepatitis B (CHB). Methods: The study included 619 patients who were diagnosed with CHB from March 2005 to December 2011. AFP levels were measured by electrochemiluminescence. Liver biopsy samples were classified into five levels of inflammation (G) and fibrosis (S) stages, according to the Chinese guidelines for prevention and treatment of viral hepatitis. Two multivariable ordinal regression models were performed to determine associations between AFP, GGT, and APRI (AST/PLT ratio) and stages of inflammation and fibrosis. Results: Significant positive and moderate correlations were shown between AFP levels and inflammation stages and between AFP levels and fibrosis stages (rho = 0.436 and 0.404, p < 0.001). Median values of AFP at liver fibrosis stages S0-1, S2, S3, and S4 were 3.0, 3.4, 5.4, and 11.3 ng/ml, respectively, and median APRI (AST/PLT ratio) was 0.41. Receiver operating characteristic (ROC) curve analyses revealed that the areas under the curves (AUCs) were 0.685, 0.727, and 0.755 (all p < 0.001) for judging inflammation stages of G >= 2, G >= 3, G = 4 by AFP; and 0.691, 0.717, and 0.718 (all p < 0.001) for judging fibrosis stages of S >= 2, S >= 3, and S = 4 by AFP. APRI levels showed significant positive and moderate correlations with inflammation stages (p = 0.445, p < 0.001). AST, GGT, and APRI levels showed significant positive but very weak to weak correlations with fibrosis stages (p = 0.137, 0.237, 0.281, p < 0.001). Conclusions: Serum AFP levels increased as pathological levels of inflammation and fibrosis increased in CHB patients. Our data showed the clinical significance of serum AFP levels in diagnosing liver inflammation and fibrosis. Assessment of liver pathology may be improved by creating a predictive mathematical model by which AFP levels with other biomarkers.