Exploration Of Noninvasive Diagnostic Model For Hepatic Fibrosis In Children With Chronic Hepatitis B

Background: Although some noninvasive serum markers for liver fibrosis detection had been developed in adults with chronic hepatitis B(CHB),there were few corresponding validated markers for pediatric patients.Objective: To evaluate existing noninvasive methods and explore other potential parameters for the diagnosis of liver fibrosis in children with CHB.Methods: 57 children with CHB who underwent liver biopsy were retrospectively included.The inflammation grading and fibrosis staging of liver biopsy samples were evaluated with the Scheuer histological scoring system.Differences in clinical,blood,biochemical,and viral parameters among different liver fibrosis stages were analyzed,and correlation analysis was conducted simultaneously.Univariate and multivariate logistic regression analysis was conducted to explore new model for diagnosing of S≥ 2.Receiver operating characteristic(ROC)curve analysis was used to evaluate noninvasive tests,including the aspartate transaminase(AST)-to-platelet ratio index(APRI),FIB4 index and gamma-glutamyl transpeptidase(GGT)-to-platelet ratio(GPR),and new noninvasive parameters were further explored.Results: Histological analysis showed that necroinflammation and fibrosis were more severe in older or male children.Spearman rank correlation analysis indicated that fibrosis staging correlated positively with age and GGT,but negatively with ALP(r=0.4 P=0.002;r=0.35 P=0.008;and r=-0.322 P=0.014,respectively).Univariate and multivariate regression analysis indicated that age,GGT and ALP were independent factors of S≥2and then output an equivalent conversion model P=e^[0.289*age(y)+0.032*GGT(U/L)-0.031*ALP(U/L)+3.721].For diagnosis of significant fibrosis,the areas under the ROC curve(AUC)for FIB4 and the GPR were 0.802 and 0.705,while the APRI showed no statistical significance.The sensitivity(Se),specificity(Sp),positive predictive value(PPV),negative predictive value(NPV),positive likelihood ratio(LR+)and negative likelihood ratio(LR-)values were 90.5%,66.7%,61.3%,92.3%,2.7 and 0.14 for FIB4 and 47.6%,91.7%,77.0%,75.0%,5.7 and 0.57 for the GPR,respectively.The AUC of new diagnostic model reached 0.893 for diagnosing of S≥2,and the Se=81.0%,Sp=88.9%,PPV=80.9%,NPV=88.9%,LR+=7.3 and LR-=0.21,superior to FIB4 and GPR.Conclusion: The new noninvasive model composed of age,GGT and ALP owned a good diagnostic ability for S≥2,which is better than FIB4 and GPR models.This model is expected to be a good indicator for guiding treatment and follow-up in children with CHB.