Construction of predictive model for early allograft dysfunction after liver transplantation
10.16016/j.2097-0927.202312009
- VernacularTitle:肝移植术后早期移植器官功能不全的预测模型构建
- Author:
Xin LI
1
;
Xinglin YI
;
Yan CHEN
;
Xin DENG
;
Xiangfeng LIU
;
Xianzhe LIU
;
Ying JIANG
;
Guanlei LIU
;
Chunmei CHEN
;
Fang QIU
;
Jianteng GU
Author Information
1. 400038 重庆,陆军军医大学(第三军医大学)第一附属医院麻醉科
- Keywords:
liver transplantation;
early allograft dysfunction;
prediction model;
risk factors
- From:
Journal of Army Medical University
2024;46(7):746-752
- CountryChina
- Language:Chinese
-
Abstract:
Objective To analyze the factors related to early allograft dysfunction(EAD)after liver transplantation and to construct a predictive model.Methods A total of 375 patients who underwent liver transplantation in our hospital from December 2008 to December 2021 were collected,including 90 patients with EAD and 266 patients without EAD.Thirty items of baseline data for the 2 groups were compared and analyzed.Aftergrouping in a ratio of 7∶3,univariate and multivariate logistic regression analyses were used in the training set to evaluate the factors related to EAD and construct a nomogram.Receiver operating characteristic(ROC)curve,decision curve analysis(DCA),sensitivity,specificity,positive predictive value,negative predictive value,Kappa value and other indicators were used to evaluate the model performance.Results The incidence of EAD after liver transplantation was 24%.Multivariate logistic regression analysis showed that preoperative tumor recurrence history(OR=3.15,95%CI:1.28～7.77,P=0.013)and operation time(OR=1.22,95%CI:1.04～1.42,P=0.015)were related to the occurrence of EAD after surgery.After predicting the outcome according to the cut-off point of 0.519 identified by the Youden index,the model performance in the both training set and validation set was acceptable.DCA suggested the model has good clinical applicability.Conclusion The risk factors for EAD after liver transplantation are preoperative tumor recurrence history and operation time,and the established model has predictive effect on prognosis.
