부산대학교 도서관

ABSTRACTThe purpose of this study was to develop an effective and non-invasive nomogram for evaluating liver obvious inflammation in untreated HBeAg positive patients with chronic hepatitis B virus (HBV) infection. A nomogram was established on a model cohort of 292 treatment-naïve HBeAg positive patients with normal alanine aminotransferase (ALT ≤40 U/L) at Beijing Ditan Hospital from January 2008 to March 2018. Then the nomogram was prospectively validated in a cohort of 88 patients from July 2019 to May 2021. Calibration curves and Concordance index were used to evaluate the accuracy of prediction and identification performance of the model. In untreated HBeAg positive chronic hepatitis B virus infection patients with normal ALT, the formula for predicting liver inflammation was Logit (P) =-0.91-0.41×log10 (qHBeAg)+0.11×AST-0.01×PLT. The nomogram had C-index of 0.751 (95% CI, 0.688–0.815), indicating a good consistency between prediction and real observation on the model cohort. The validation cohort confirmed its good performance. In this study, liver inflammation nomograms based on HBeAg, AST, and PLT were established and verified in treatment-naïve HBeAg positive chronic HBV patients with normal ALT.