The predictive value of different scoring models on short-term outcome in patients with hepatitis B-related acute-on-chronic liver failure undergoing liver transplantation
10.3760/cma.j.cn113884-20200813-00429
- VernacularTitle:不同评分模型对乙肝相关慢加急性肝功能衰竭肝移植术后早期预后的预测作用
- Author:
Qikun ZHANG
1
;
Menglong WANG
Author Information
1. 首都医科大学附属北京佑安医院普外中心 100069
- Keywords:
Liver transplantation;
Scoring model;
Hepatitis B-related acute-on-chronic liver failure;
Short-term outcome
- From:
Chinese Journal of Hepatobiliary Surgery
2021;27(6):438-444
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To compare the prognostic accuracy of 16 pre-transplant scoring models in predicting the post-transplant short-term outcome of patients with hepatitis B-related acute-on-chronic liver failure (HBACLF), and to explore an efficient predictive model.Methods:A retrospective analysis of the clinical data of HBACLF patients who underwent liver transplantation at the Liver Transplant Center of Beijing Youan Hospital from August 2004 to September 2014. Score of 16 models (CTP, UNOS-MELD, Updated-MELD, Integrated-MELD, MELD-Na, MLED Na, CLIF-SOFA, CLIF-OFs, CLIF-C ACLFs, CLIF-C ADs, Refit MELD, Refit MELD Na, MELD-AS, Zheng's Risk, UKELD, MESO) was based on time-dependent operation characteristic curve, and the area under the curve (AUC) was calculated to evaluate the prediction accuracy of 3-month survival after transplantation. Selection of univariate factors associated with postoperative short-term mortality was performed, and then 16 scoring models one by one with statistically significant mortality-related factors were entered into LASSO regression (Least Absolute Shrinkage and Selection Operator regression) to confirm the independent variables. Finally, a predictive model was constructed by Cox regression.Results:A total of 135 patients were included in this study, including 106 males and 29 females, aged (45.0±10.5) years old. Among the 16 scoring models, the AUC of MELD-Na and CLIF-SOFA were more than 0.7 in early survival prediction after liver transplant. The MELD-Na was confirmed as an independent predictive variable in the final model with univariate and LASSO regression multivariate selection analysis ( HR=1.0481, 95% CI: 1.0136-1.0838, P<0.05). The model was constructed by MELD-Na and combined with other clinical parameters (female, systemic infection, placement of T tube during operation) could better predict the early survival after liver transplant. The overall C-index of the final model was 0.886, and the C-index at 3-month after liver transplant was 0.844 through internal validation (Bootstrap). Conclusion:Compared with other scoring models, MELD-Na and CLIF-SOFA were better for early survival prediction after liver transplantation for patients with HBACLF. The constructed predictive model based on MELD-Na was superior than single MELD-Na or CLIF-SOFA in prognostic assessment and case selection.
