Independent Risk Factors And Noninvasive Model Establishment Of Liver Histological Changes In HBeAg-Positive Chronic HBV Infection Patients With Normal Alanine Aminotransferase

BackgroundHepatitis B virus(HBV)infection is still a global public health problem.It is estimated that there are about 250 million hepatitis B surface antigen(HBsAg)carriers worldwide and 70 million chronic HBV infection individuals in China.The Chinese guidelines for the prevention and treatment of chronic hepatitis B(2019 edition)recommend that people with HBV DNA positive,alanine aminotransferase(ALT)greater than the upper limit of normal(ULN)should be treated with antiviral drugs.For HBeAg-positive CHB patients with normal ALT,it was previously believed that these patients were in the "immune tolerance" stage with no significant liver inflammation and fibrosis.However,recent studies have shown that significant histological changes[G≥2 and/or S≥2]are still present in chronic HBV infected patients with normal or mildly elevated ALT,and these patients are also at higher risk of developing cirrhosis,hepatocellular carcinoma(HCC),and end-stage liver diseases.Therefore,it is of great significance to accurately evaluate the degree of liver lesions in HBeAg-positive CHB patients with normal ALT.Liver biopsy is the gold standard for evaluating liver histological changes,but liver biopsy is an invasive procedure with potential bleeding risk and sampling error.Therefore,the clinical application of liver biopsy is restricted.In recent years,a variety of noninvasive methods for liver tissue assessment,such as serum noninvasive models and transient elastography(TE),have become the main alternative methods for evaluating liver fibrosis.However,the noninvasive models mentioned above are mainly used to evaluate the stage of liver fibrosis,and their diagnostic performances for liver inflammation are limited.At present,there is a lack of noninvasive model to evaluate liver inflammation and fibrosis simultaneously in HBeAg-positive chronic HBV infection patients with normal ALT.ObjectiveTo investigate the independent risk factors for significant liver histological changes in HBeAg-positive chronic HBV infection patients with normal ALT,then to develop and validate a noninvasive model for predicting liver histology.Methods1.A total of 251 patients with chronic HBV infection who were treated in the first affiliated hospital of Zhengzhou university from May 2016 to November 2020 were retrospectively enrolled.All patients were treatment-naive and randomly assigned to a training cohort(n=156)or a validation cohort(n=95).2.According to the results of liver biopsies,the enrolled patients were divided into two groups:the significant liver lesion group(G>2 and/or S>2)and no significant liver lesion group(G<2 and S<2).Logistic regression was used to explore the independent risk factors for liver histological changes in the training cohort.3.Pearson correlation analysis was used to investigate the correlation between the new noninvasive model and patients’ age,HBsAg quantification and liver stiffness measurement(LSM).Spearman correlation analysis was further used to explore the correlation between the new noninvasive model and liver inflammation grade and fibrosis stage.4.The area under receiver operating characteristic(ROC)curve was used to evaluate the predictive performance of the noninvasive model.Results1.According to the Scheuer score system,88 of 251 patients had significant liver lesions.To be specific,34 patients(21.8%)in the training cohort and 25 patients(26.3%)in the validation cohort,respectively had moderate to severe liver inflammation;48 patients in the training cohort(30.8%)and 30 patients(31.6%)in the validation cohort respectively had significant liver fibrosis;55 patients(35.3%)in the training cohort and 33 patients(34.7%)in the validation cohort,respectively had significant liver lesions.The clinical parameters were comparable between the training cohort and the validation cohort(all P>0.05).2.Multivariate analysis showed that age(P=0.016),HBsAg(P=0.002)and LSM(P<0.001)were independent risk factors for significant liver histological changes.Consequently,a new noninvasive model,which was composed of age,HBsAg and LSM,was developed to predict the presence of liver histology.3.Pearson correlation analysis showed that the new noninvasive model was positively correlated with age(r=0.61,P<0.001),negatively correlated with HBsAg quantification(r=-0.66,P<0.001),and positively correlated with LSM(r=0.78,P<0.001).Similarly,spearman correlation analysis showed that the new noninvasive model had a positive correlation with liver inflammation grade(r=0.53,P<0.001)and liver fibrosis stage(r=0.62,P<0.001).4.The area under ROC curve(AUC)of the new noninvasive model was 0.861 in the training cohort,which was significantly better than that of aspartate aminotransferase to platelet ratio index(APRI)(AUC=0.759,P=0.011),fibrosis 4 score(FIB-4)(AUC=0.766,P=0.012)and LSM(AUC=0.808,P=0.042).In the validation cohort,the predictive performance of the noninvasive model was 0.872,which was also significantly better than that of APRI(AUC=0.765,P=0.029),FIB-4(AUC=0.782,P=0.035)and LSM(AUC=0.798,P=0.045).ConclusionAge,HBsAg and LSM are independent risk factors for liver histological changes in HBeAg-positive chronic HBV infection patients with normal ALT.The new noninvasive model constructed by the three parameters is better than APRI,FIB-4 and LSM in predicting liver histological lesions.