首页|肝硬化并发肺部感染预测模型的构建和验证

肝硬化并发肺部感染预测模型的构建和验证

扫码查看
目的:建立肝硬化并发肺部感染的危险因素数学模型并进行验证分析.方法:选取2023年7月-2024年3月在我院消化内科实施治疗的200例肝硬化患者,其中43例患者并发肺部感染(感染组),另外157例患者未发生肺部感染(非感染组),采用单因素和多因素数学模型分析两组患者的一般资料、病情资料、实验室指标等与肺部感染发生的相关因素,并绘制列线图预测模型.结果:感染组患者的年龄大于非感染组,合并慢阻肺的占比高于非感染组,差异均有统计学意义(P<0.05);感染组患者的白细胞测定值、肝肾综合征患者占比、自发性腹膜炎患者占比、合并腹水患者占比、上消化道出血患者占比、电解质失衡患者占比、低蛋白血症患者占比均高于非感染组患者,差异均有统计学意义(P<0.05);logistic回归模型结果显示:合并慢阻肺、合并肝肾综合征、合并自发性腹膜炎、合并腹水、伴有上消化道出血、发生电解质失衡、出现低蛋白血症的患者肺部感染的发生风险明显增高(P<0.05);列线图模型具有较好的判断价值(C指数=0.801,95%CI:0.656~0.883),列线图模型预测肝硬化患者发生肺部感染的ROC曲线下面积AUC值为0.886,95%CI:0.805~0.931,预测临界值总分为267分.结论:肝硬化并发肺部感染的危险因素较多,建立各项因素的列线图模型用于预测患者的肺部感染发生风险具有较高的临床实用价值.
Construction and Validation of a Predictive Model for Concurrent Pulmonary Infection in Liver Cirrhosis
Objective:Explore the mathematical model of risk factors for pulmonary infection in patients with liver cirrhosis and conduct validation analysis.Methods:Select 200 patients with liver cirrhosis who underwent treatment in the department of Gastroenterology at our hospital from July 2023 to March 2024.Among them,43 patients had concurrent pulmonary infec-tions(infection group),and another 157 patients did not have pulmonary infections(non infection group).Use univariate and multivariate mathematical models to analyze the general data,disease data,laboratory indicators,and other related factors of the two groups of patients with pulmonary infections,and draw a column chart prediction model.Results:The age of patients in the infected group was higher than that in the non infected group,and the proportion of patients with chronic obstructive pulmonary disease in the infected group was higher than that in the non infected group,with statistical significance(P<0.05).The white blood cell test values,proportion of patients with hepatorenal syndrome,proportion of patients with sponta-neous peritonitis,proportion of patients with concurrent ascites,proportion of patients with upper gastrointestinal bleeding,proportion of patients with electrolyte imbalance,and proportion of patients with hypoalbuminemia in the infection group were all higher than those in the non infection group,and the differences were statistically significant(P<0.05).The results of the logistic regression model showed that patients with chronic obstructive pulmonary disease,hepatorenal syndrome,spontaneous peritonitis,ascites,upper gastrointestinal bleeding,electrolyte imbalance,and hypoalbuminemia had a significantly higher risk of pulmonary infection(P<0.05).The column chart model has good diagnostic value(C-index=0.801,95%CI:0.656~0.883).The AUC value of the area under the ROC curve predicted by the column chart model for pulmonary infec-tion in patients with liver cirrhosis is 0.886,95%CI:0.805~0.931,and the total predicted critical score is 267 points.Conclusion:There are many risk factors for concurrent pulmonary infection in liver cirrhosis,and establishing a column chart model for each factor has high clinical practical value in predicting the risk of pulmonary infection in these patients.

Liver cirrhosisPulmonary infectionRisk factorsColumn chartForecast

关芬海、林彩战、陈明倩

展开 >

广东省阳江市阳东区人民医院 529500

肝硬化 肺部感染 危险因素 列线图 预测

2025

医学理论与实践
河北省预防医学会 河北省药学会

医学理论与实践

影响因子:0.746
ISSN:1001-7585
年,卷(期):2025.38(1)