(1) Background: As specialparameters in predicting significant hepatitis activity of hepatitis B e antigen (HBeAg)-positive chronic hepatitis B virus (HBV) infection, the quantitative standard of HBV DNA has not been agreed and that of hepatitis B surface antigen(HBsAg) has not been formed. Our objective is to evaluate the validity of HBsAg and HBV DNA in predicting the significant hepatitis activity of HBeAg-positive patients. (2) Methods: A population of 516 patients with HBeAg-positive chronic HBV infection was enrolled. Serum ALT was measured using an Abbott Architect c16000 autoanalyzer; diagnoses of liver pathological grade and stage referred to the Scheuer standard. Three levels of significant hepatitis activity were preset, which were successively "ALT ≥ 20 IU/L or Grade > G1 or Stage > S1", "ALT ≥ 30 IU/L or Grade > G1 or Stage > S1" and "ALT ≥ 40 IU/L or Grade > G1 or Stage > S1". (3) Results: A subpopulation of 288 patients with possible high HBV replication was selected based on locally weighted scatterplot smoothing regression curves between ALT and HBsAg, HBeAg and HBV DNA. In the subpopulation with possible high HBV replication, areas under receiver operating characteristic curves of HBsAg for predicting the three levels of significant hepatitis activity were successively 0.868, 0.839 and 0.789, which were all significantly greater than those of HBV DNA, as those were successively 0.553, 0.550 and 0.574 (p = 0.0002, p < 0.0001 and p < 0.0001). With the standard of HBsAg ≤ 4.699 log10 IU/mL, the sensitivity and specificity of HBsAg for predicting the three levels of significant hepatitis activity were successively 75.81% and 81.82%, 79.23% and 78.57% and 80.82% and 67.44%. (4) Conclusion: Quantitative HBsAg instead of HBV DNA is valuable in predicting significant hepatitis activity of HBeAg-positive chronic HBV infection.
Keywords: chronic hepatitis B virus infection; hepatitis B surface antigen; hepatitis B virus DNA; natural history; significant hepatitis activity.