无创性指标预测丙型肝炎肝硬化患者食管静脉曲张  

Prediction of esophageal varices in patients with hepatitis C-associated liver cirrhosis by non-invasive markers

作　　者：高丽丽[1] 张亦瑾[1] 高学松[1] 李洪杰[1] 刘楠[1] 高萍[1] 段雪飞[1] GAO Li-li;ZHANG Yi-jin;GAO Xue-song;LI Hong-jie;LIU Nan;GAO Ping;DUAN Xue-fei(Department of General Medicine,Beijing Ditan Hospital,Capital Medical University,Beoing 100015,China)

机构地区：[1]首都医科大学附属北京地坛医院综合科,北京100015

出　　处：《中国肝脏病杂志（电子版）》2018年第3期73-76,共4页Chinese Journal of Liver Diseases：Electronic Version

基　　金：新发突发传染病研究北京市重点实验室开放课题(DTKF201602);北京市医管局培育计划(PX2017029)

摘　　要：目的筛选无创性预测丙型肝炎肝硬化患者食管静脉曲张的指标,建立Logistic回归模型,并探讨该模型的应用价值。方法回顾性分析2005年1月至2016年1月首都医科大学附属北京地坛医院收治的110例丙型肝炎肝硬化患者的临床资料,根据胃镜结果将患者分为无食管静脉曲张组和食管静脉曲张组。采用非条件Logistic回归模型分析患者的临床特点、超声检查结果和生物化学指标,筛选无创预测食管静脉曲张的指标,在此基础上建立回归模型。另选取2016年2月至2018年2月就诊于首都医科大学附属北京地坛医院的丙型肝炎肝硬化初治患者,随访1年,验证该模型的阳性预测值及阴性预测值。结果单因素分析表明,无食管静脉曲张组和食管静脉曲张组患者的白蛋白、凝血酶原时间、血小板和脾长差异有统计学意义(P <0.05)。多因素非条件Logistic回归分析表明,脾长、血小板及凝血酶原时间可有效预测食管静脉曲张。Logistic回归模型为Y=-1.053+1.644×凝血酶原时间(s)+2.030×脾长(cm)-1.573×血小板(×109/L)。该模型对食管静脉曲张的预测灵敏度为83.55%,特异度为87.04%,阴性预测值为87.5%,阳性预测值为87.5%。结论血小板、脾长和凝血酶原时间可预测丙型肝炎肝硬化患者食管静脉曲张的存在。Objective To investigate non-invasive predictors of esophageal varices （EV） in patients with hepatitis C-associated liver cirrhosis, to establish a Logistic regression model and to explore the application value of the model. Methods Clinical data of 110 patients with hepatitis C-associated cirrhosis in Beijing Ditan Hospital, Capital Medical University from January 2005 to January 2016 were retrospectively analyzed. The patients were divided into esophageal varices （EV） group and non-EV group according to gastroscopy results. The non- conditional Logistic regression model was used to analyze the clinical characteristics, ultrasound results and biochemical indexes of the patients, the non-invasive indexes for predicting esophageal varices were screened, and the regression model was established on the basis of this model. Results Univariate analysis showed that there were significant differences in albumin, prothrombin time, platelet and spleen length between patients with and without esophageal varices （P 〈 0.05）. Multivariate conditional Logistic regression analysis showed that splenic length, platelet and prothrombin time could effectively predict esophageal varices. The logistic regression model was Y = -1.053 ＋ 1.644 × prothrombin time （s） ＋ 2.030 × spleen length （cm） - 1.573 × platelets （× 10 9 /L）. The sensitivity was 83.55%, the specificity was 87.04%, the negative predictive value was 87.5% and the positive predictive value was 87.5%. Conclusions Platelet, splenic length and prothrombin time can predict the presence of esophageal varices in patients with hepatitis C-associated cirrhosis.

关 键 词：丙型肝炎肝硬化 食管静脉曲张 无创性 

分 类 号：R575.2[医药卫生—消化系统] R512.63[医药卫生—内科学]