肝硬化糖代谢异常筛查模型的构建  

Construction of a Screening Model for Cirrhosis with Abnormal Glucose Metabolism

作　　者：李雪勤 朱炳喜 Li Xueqin;Zhu Bingxi(Xuzhou Medical University Graduate School,Jiangsu 221000,China)

机构地区：[1]徐州医科大学研究生学院,221000 [2]徐州医科大学附属医院消化内科,221000

出　　处：《医学研究杂志》2020年第3期141-145,共5页Journal of Medical Research

摘　　要：目的 研究肝硬化患者发生糖代谢异常的相关因素,并构建肝硬化患者并发糖代谢异常的筛查模型.方法 采用回顾性病例对照的研究方法,收集了2017年1月~2018年12月徐州医科大学附属医院收治的572例肝硬化患者的临床病例资料(其中以肝硬化患者糖代谢异常者为实验组,肝硬化患者糖代谢正常者为对照组).通过Logistic回归分析,构建肝硬化糖代谢异常的相关筛查模型,并最终绘制筛查模型及各个指标的受试者工作特征(ROC)曲线,计算曲线下面积.结果 Logistic回归分析表明年龄、肝病病程、饮酒史、高胆固醇水平、高甘油三酯水平、肝功能C级、丙肝病毒感染、高血压是肝硬化患者并发糖代谢异常的独立危险因素.基于危险因素构建的模型对于筛查肝硬化糖代谢异常的敏感度为63.5%,特异性为77.6%,准确度为71.3%.ROC曲线下面积(AUG)为0.763(0.724 ~0.803),优于其他任意单一指标.结论 构建的筛查模型有较高的敏感度及特异性,有助于早期预防和发现肝硬化糖代谢异常的发生,为临床医生的诊疗提供一定的参考价值.Objective To investigate the related factors that caused the abnormal glucose metabolism in patients with eirhosis and to construct a screening model of abnormal glucose metabolism in patients with cirrhosis.Methods The retrospective case-control method was used to collet clinical case data of 572 patients with cirrhosis in the Affiliated Hospital of Xuzhou Medical University from January 2017 to December 2018(The patients with cirrhosis and abnormal glucose metabolism were set as the experimental group,and the patients with cirrhosis and normal glucose metabolism were set as the control group).A screening model for cirrhosis and abnormal glucose metab-olism was built using the Logistic regression analysis.The screening model and the receiver operating characteristic curve(ROC curve)of studied factors were plotted.In addition,the area under the curve was calculated.Results Logistic regression analysis showed that age,course of liver disease,drinking history,high cholesterol level,high triglyceride level,grade C of liver function,infection of hepatitis C virus and hypertension were independent risk factors for cirrhosis with abnormal glucose metabolism.The sensitivity of the model based on risk factors for screening cirrhosis and abnormal glucose metabolism was 63.5%,the specificity was 77.6%,and the aceuracy was 71.3%.The area under the ROC curve was 0.763(0.724-0.803),which was superior to any other single indicator.Conclusion The constructed logistic regression model had high sensitivity and specificity,which is helpful for early prevention and detection of eirrhosis with abnormal glucose metabolism and can provide the certain reference for clinical diagnosis and treatment.

关 键 词：肝硬化 糖代谢异常 危险因素 Logistic回归分析模型 

分 类 号：R575.[医药卫生—消化系统]