ABSTRACT
ObjectiveTo investigate the value of aspartate aminotransferase-to-platelet ratio index (APRI), fibrosis-4 (FIB-4) score, and gamma-glutamyl transpeptidase-to-platelet ratio (GPR) in diagnosis of liver inflammation grade in patients with chronic hepatitis B (CHB). MethodsA total of 545 patients with CHB who underwent percutaneous liver biopsy and routine laboratory examinations during hospitalization in Shanghai Public Health Clinical Center Affiliated to Fudan University from October 2016 to October 2019 were enrolled. Inflammation grade (G) was determined according to the Scheuer scoring system, and APRI, FIB-4, and GPR were calculated based on related clinical indicators. The t-test was used for comparison of normally distributed continuous data between two groups, and the Mann-Whitney U test was used for comparison of non-normally distributed continuous data between two groups; the chi-square test was used for comparison of categorical data between two groups. A Spearman correlation analysis was used to investigate the correlation between two variables. The receiver operating characteristic (ROC) curve was used to evaluate the diagnostic performance of the three serum noninvasive diagnostic models in determining liver inflammation grade, and the Delong test was used for comparison of the area under the ROC curve (AUC). ResultsAmong the 545 patients, 224 had grade G0-1 liver inflammation, 209 had grade G2 liver inflammation, and 112 had grade G3 liver inflammation. The Spearman correlation analysis showed that APRI, FIB-4, and GPR were positively correlated with liver inflammation grade (r=0.611, 0.470, and 0.563, all P＜0.001). APRI, FIB-4, and GPR had an AUC of 0.820, 0.719, and 0782, respectively, in the diagnosis of G≥2 liver inflammation, with optimal cut-off values of 0.53, 1.48, and 0.20, respectively; for the diagnosis of G≥2 liver inflammation, GPR had a better performance than FIB-4 (P=0.01) and a slightly lower performance than APRI (P=0.048). The stratified analysis based on alanine aminotransferase (ALT) level showed that in the ≤1×upper limit of normal (ULN) group, the (1-2)×ULN group, and the (2-5)×ULN group, APRI had an AUC of 0.847, 0.786, and 0.724, respectively, in the diagnosis of G≥2 liver inflammation, FIB-4 had an AUC of 0.777, 0.729, and 0.626, respectively, and GPR had an AUC of 0.801, 0.781, and 0.607, respectively; the subgroup analysis showed that GPR had a similar diagnostic performance to APRI and FIB-4 in all ALT stratification groups except the (2-5)×ULN group, in which GPR had a lower diagnostic performance than APRI (P=0.042). APRI, FIB-4, and GPR had an AUC of 0.791, 0.725, and 0.801, respectively, in the diagnosis of G≥3 liver inflammation, with optimal cut-off values of 0.66, 1.49, and 0.25, respectively; in the diagnosis of G≥3 liver inflammation, GPR had a similar diagnostic performance to APRI and a better diagnostic performance than FIB-4 (P=0.006). The stratified analysis based on ALT level showed that in the ≤1×ULN group, the (1-2)×ULN group, and the (2-5)×ULN group, APRI had an AUC of 0.900, 0.742, and 0.693, respectively, in the diagnosis of G≥3 liver inflammation, FIB-4 had an AUC of 0.874, 0.683, and 0.644, respectively, and GPR had an AUC of 0.890, 0.805, and 0.668, respectively. The subgroup analysis showed that GPR had a similar diagnostic performance to APRI and FIB-4 in all ALT stratification groups except the (1-2)×ULN group, in which GPR had a better diagnostic performance than FIB-4(P=0.015). ConclusionAPRI, FIB-4, and GPR may accurately diagnose liver inflammation grade in CHB patients, which helps to monitor the progression of CHB and determine the timing of antiviral therapy.