肝癌TACE术后急性肝功能恶化的危险因素及风险列线图模型的构建与验证  

Analysis of risk factors for acute liver function deterioration after liver cancer intervention surgery and construction and validation of a risk column chart model

作　　者：卢双动 王会哲 王谦[3] LU Shuang-dong;WANG Hui-zhe;WANG Qian(Emergency Department,The Second Central Hospital of Baoding,Baoding Hebei,072750,China;不详)

机构地区：[1]保定市第二中心医院急诊科,河北保定072750 [2]保定市第二中心医院消化科 [3]保定市第二中心医院放射科

出　　处：《中西医结合肝病杂志》2024年第10期884-888,共5页Chinese Journal of Integrated Traditional and Western Medicine on Liver Diseases

基　　金：保定市科学计划项目(No.2341ZF037)。

摘　　要：目的:探讨肝癌肝动脉化疗栓塞(TACE)术后急性肝功能恶化(ALFD)的危险因素,构建预测术后ALFD发生的列线图模型,并进行验证。方法:回顾性分析2020年1月至2022年12月开展TACE术的163例肝癌患者的临床资料,根据术后是否发生ALFD分成ALFD组和良好组。对患者一般资料、TACE资料、术前实验室检查资料进行单因素分析,采用多因素Logistic分析TACE术后发生ALFD的危险因素。据此构建TACE术后发生ALFD的预测模型,并利用ROC曲线下面积(AUC)、校准曲线验证其效果。结果:163例患者中,TACE术后发生ALFD 44例(26.99%)。两组患者术前Child-Pugh分级、合并门脉癌栓情况、TACE次数、术前Alb、术前NLR、术前APRI比较,差异有统计学意义(P<0.05);Logistic回归分析显示,ALFD发生的危险因素主要有Child-Pugh B级、合并门脉癌栓、TACE次数>2次、术前NLR≥3.81、术前高APRI;内部验证显示,AUC为0.837(95%CI=0.773~0.902),模型预测ALFD发生的概率高度吻合实际概率,拟合优度HL检验χ^(2)=11.305,P=0.185。结论:根据术前Child-Pugh分级、是否合并门脉癌栓、TACE次数、术前NLR、术前APRI构建的列线图模型可对肝癌介入术后ALFD予以有效预测。Objective:Investigating the risk factors for acute liver function deterioration(ALFD) following transcatheter arterial chemoembolization(TACE) for hepatocellular carcinoma,developing a nomogram model to predict the occurrence of postoperative ALFD,and validating its accuracy.Methods:A retrospective analysis was conducted on the data of 163 liver cancer patients who underwent interventional surgery in our hospital from January 2020 to December 2022.They were separated into ALFD group and good group based on whether ALFD occurred after surgery.General information,disease and interventional procedure data,and preoperative laboratory examination data were collected for univariate analysis,and multivariate analysis was applied to analyze the risk factors for postoperative ALFD after TACE.Based on this,a prediction model for postoperative ALFD in TACE was constructed,and its effectiveness was verified using the area under the ROC curve(AUC) and calibration curve.Results:Among 163 patients,44 cases developed ALFD after interventional surgery,with an incidence rate of 26.99%.There were differences between the two groups in preoperative Child-Pugh grading,concurrent portal vein cancer thrombus,number of interventional treatments,preoperative Alb,preoperative NLR,and preoperative APRI(P<0.05).Logistic regression analysis showed that the main risk factors for the occurrence of ALFD included Child-Pugh grade B,concurrent portal vein tumor thrombus,interventional therapy frequency>2 times,preoperative NLR≥3.81,and high preoperative APRI;internal validation shows that the AUC was 0.837(95% CI:0.773-0.902),and the probability of ALFD predicted by the model was highly consistent with the actual probability,the goodness of fit HL test showed χ~2=11.305,and P=0.185.Conclusion:A column chart model constructed based on preoperative Child-Pugh grading,presence or absence of portal vein tumor thrombus,number of interventional treatments,preoperative NLR,and preoperative APRI can effectively predict postoperative ALFD in liver cance

关 键 词：肝癌 肝动脉化疗栓塞术 急性肝功能恶化 危险因素 列线图模型 

分 类 号：R735.7[医药卫生—肿瘤]