Diagnostic accuracy of different noninvasive scores for detecting advanced fibrosis in chronic hepatitis B.

Abstract

The liver biopsy is the gold standard for determining the level of fibrosis in chronic hepatitis B infection (CHBI). Nonetheless, it is possible to predict liver fibrosis through some noninvasive methods such as noninvasive scoring (NIS) of some serum biomarkers obtained from routine blood tests. We aimed to evaluate the diagnostic accuracy of nine NIS for detecting advanced fibrosis in CHBI. We reviewed the hospital records of CHBI cases with liver biopsy between January 2011 and December 2016 retrospectively. Using Ishak scoring method, we classified fibrosis stage 1-2 as mild and 3-6 as advanced fibrosis. We calculated the NIS by considering the age, platelet count, aspartate aminotransferase, alanine aminotransferase, gamma-glutamyltransferase, platelet, and international normalized ratio values at the time of the biopsy. The mean age of 202 patients was 37.69± 11.33 years. In cases with advanced fibrosis, the age, gammaglutamyltransferase, and international normalized ratio values were higher and platelet count was lower (P < 0.05). Mean platelet volume was not different between the two groups (P = 0.499). The median values of γ-glutamyl peptidase-platelet ratio (GPR), FibroQ, Goteborg University Cirrhosis Index, fibrosis-4 (FIB-4), aspartate aminotransferase-platelet ratio index, age-platelet index, and King scoring were significantly higher in the advanced fibrosis group. The highest area under the curve value was in GPR [AUC = 0.731 (0.639-0.788); P = 0.000] in the receiver operating characteristic curve analysis. Cirrhosis Discriminant Score and Aspartate aminotransferase-to-alanine aminotransferase ratio tests were not valuable in detecting advanced fibrosis. FIB-4 had the highest (0.678) diagnostic accuracy rate. We found that the calculation of NIS before liver biopsy, especially GPR and FIB-4, may be useful for predicting advanced fibrosis in cases with CHBI.