医学信息2025，Vol.38Issue(9)：27-32,6.DOI:10.3969/j.issn.1006-1959.2025.09.005
肝硬化合并消化道出血患者行内镜治疗后再出血预测模型的建立与验证
Establishment and Validation of a Predictive Model for Recurrent Bleeding in Liver Cirrhosis Patients with Gastrointestinal Bleeding After Endoscopic Treatment
摘要
Abstract
Objective To establish a predictive model for recurrent bleeding in liver cirrhosis patients with upper gastrointestinal bleeding after endoscopic treatment,and to verify the predictive value of this model.Methods A total of 352 liver cirrhosis patients with gastrointestinal rebleeding admitted to our hospital from January 2017 to December 2021 were included as the study subjects.They were divided into a non bleeding group of 199 cases and a bleeding group of 153 cases according to whether the patients experienced gastrointestinal rebleeding within 2 years after endoscopic treatment.Univariate and multivariate logistic regression were used to analyze the influencing factors of rebleeding after endoscopic treatment in patients with liver cirrhosis complicated with upper gastrointestinal bleeding.A column chart model was constructed based on the independent predictive factors.ROC curves and decision curves were drawn,and the predictive efficiency and net return of independent predictive factors and joint prediction were analyzed.Results Univariate analysis showed that there were significant differences in BMI,portal vein diameter,Child-Pugh classification,degree of esophageal and gastric varices,Hp infection,portal vein thrombosis and bleeding volume between the two groups(P＜0.05).Multivariate Logistic regression analysis showed that portal vein diameter,Child-Pugh classification,degree of esophagogastric varices,Hp infection and bleeding volume were independent predictors of rebleeding after endoscopic treatment in patients with cirrhosis and gastrointestinal bleeding(P＜0.05).A column chart model was established:P=1/(1+e-x),X=0.002×bleeding volume(ml)+1×Child-Pugh grade(A=0,B=0.890,C=1.440)+1×Gsophageal and gastric varices(Mild=0,Moderate=1.886,Severe=2.773)+1.470×Portal vein diameter(cm)+2.574×Hp infection(Yes=1,No=0)-1.895.The result of goodness of fit(H-L)test showed that the predicted value of the risk of rebleeding after endoscopic treatment in liver cirrhosis patients with gastrointestinal bleeding was in good agreement with the actual observed value(x2=0.305,P=0.815).ROC analysis results showed that the area under the curve(AUC)of predicting rebleeding in liver cirrhosis patients with gastrointestinal bleeding was 0.863(95%CI:0.824-0.902)after endoscopic treatment.The decision curve analysis results showed that independent predictive factors and combination predicting rebleeding in liver cirrhosis patients with gastrointestinal bleeding after endoscopic treatment had good net returns.Conclusion The column chart model constructed in this study has good predictive performance and net return rate,and can be used to accurately predict the risk of rebleeding in liver cirrhosis patients with gastrointestinal bleeding after endoscopic treatment.关键词
肝硬化/消化道出血/内镜治疗/预测模型Key words
Cirrhosis/Gastrointestinal bleeding/Endoscopic therapy/Prediction model分类
临床医学引用本文复制引用
汤信鹏,冯娟,范晓棠..肝硬化合并消化道出血患者行内镜治疗后再出血预测模型的建立与验证[J].医学信息,2025,38(9):27-32,6.基金项目
新疆维吾尔自治区自然科学基金资助项目(编号:2022D01C234) （编号:2022D01C234）
