儿童肝移植非计划性术中低体温风险预测模型的构建及验证  

Construction and validation of a risk prediction model for pediatric liver transplantation with unplanned intraoperative hypothermia

作　　者：曹义 彭玉娜[2] 胡芳[3] 朱敏[4] 张华[2] 韩旭[2] 彭丽清 CAO Yi;PENG Yuna;HU Fang;ZHU Min;ZHANG Hua;HAN Xu;PENG Liqing(Operating Room,The First Central Clinical School,Tianjin Medical University,Tianjin 300192,China;Operating Room,Tianjin First Central Hospital,Tianjin 300192,China;Department of Nursing Administration,Tianjin Medical University General Hospital,Tianjin 300070,China;Departmentof Anesthesiology,Tianjin First Central Hospital,Tianjin 300192,China;Operating Room,Tian-jin Huanghe Hospital,Tianjin 300110,China)

机构地区：[1]天津医科大学一中心临床学院手术室,天津300192 [2]天津市第一中心医院手术室,天津300192 [3]天津医科大学总医院护理部,天津300070 [4]天津市第一中心医院麻醉科,天津300192 [5]天津市黄河医院手术室,天津300110

出　　处：《天津医科大学学报》2024年第1期6-10,共5页Journal of Tianjin Medical University

基　　金：天津市护理学会面上科研项目(TJHLKY20180112);天津市第一中心医院科研项目(院CF201822)。

摘　　要：目的:分析儿童肝移植非计划性术中低体温的影响因素,并构建风险预测模型。方法:选取天津市第一中心医院器官移植中心2021年10月—2023年9月儿童肝移植手术397例,将其分为建模组(317例)和验证组(80例)。采用Logistic回归分析探讨儿童肝移植非计划性术中低体温的影响因素,并构建列线图模型。结果:多因素回归分析结果显示,无肝期持续时间(OR=1.894,95%CI:1.124~3.193)、终末期肝病评分(OR=1.032,95%CI:1.006~1.059)、移植物与受者体重比(GRWR)评分(OR=1.310,95%CI:1.042~1.648)和术前血小板值(OR=2.237,95%CI:1.310~3.820)是儿童肝移植非计划性术中低体温的独立影响因素(均P<0.05)。预测模型ROC曲线下面积(AUC)值为0.773(95%CI:0.710~0.836),灵敏度为0.722、特异度为0.712,约登指数为0.434,H-L拟合优度检验显示拟合优度良好(χ^(2)=3.904,P=0.886)。验证组ROC的AUC值为0.754(95%CI:0.648~0.861),经验证预测模型总准确率为73.75%。结论:筛选的4项预测因素为儿童肝移植非计划性术中低体温的特异性因素,构建的风险预测模型有较好的预测效能。Objective:To explore the risk factors of unplanned intraoperative hypothermia(UPH)in pediatric liver transplantation and construct the nomogram prediction model.Methods:A total of 397 children undergoing liver transplantation admitted to the organ transplantation center of Tianjin First Central Hospital from October 2021 to September 2023 were selected.The patients were divided into modeling group(317 cases)and validation group(80 cases).Logistic regression analysis was used to explore the factors influencing UPH in children with liver transplantation,and a nomogram model was constructed.Results:Multivariate Logistic regression analysis showed that duration of the anhepatic phase(OR=1.894,95%CI:1.124-3.193),end-stage liver disease score(OR=1.032,95%CI:1.006-1.059),graft-to-recipient weight ratio(GRWR)score(OR=1.310,95%CI:1.042-1.648),pre-operative platelet value(OR=2.237,95%CI:1.310-3.820)were the predictors of UPH in pediatric liver transplantation(all P<0.05).The area under the ROC curve(AUC)of the nomogram model was 0.773(95%CI:0.710-0.836),the sensitivity was 0.722,the specificity was 0.712 and the Youden index of the model was 0.434.H-L goodness of fit test corresponded to good model calibration(χ^(2)=3.904,P=0.886).The AUC of the validation group was 0.754(95%CI:0.648-0.861)and the total accuracy of the verified model was 73.75%.Conclusion:Four screened predictors are specific factors for UPH in pediatric liver transplantation,and the constructed risk prediction model has good prediction efficiency.

关 键 词：肝移植 儿童 非计划性术中低体温 风险因素 预测模型 

分 类 号：R617[医药卫生—外科学] R473.72[医药卫生—临床医学]