引用本文:杨 雪,徐 斌,刘燕敏,高 文,陈 杰,单 晶,任美欣,刘 丹,边新渠,韩 莹.原发性胆汁性胆管炎患者无创肝硬化诊断敏感指标的筛选[J].中国临床新医学,2021,14(8):762-766.
【打印本页】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 1892次   下载 1223 本文二维码信息
码上扫一扫!
分享到: 微信 更多
原发性胆汁性胆管炎患者无创肝硬化诊断敏感指标的筛选
杨 雪,徐 斌,刘燕敏,高 文,陈 杰,单 晶,任美欣,刘 丹,边新渠,韩 莹
100069 北京,首都医科大学附属北京佑安医院肝病中心二科(杨 雪,徐 斌,刘燕敏,高 文,韩 莹),慢病管理中心(陈 杰),急诊科(单 晶),感染与免疫医学科(任美欣),肝病中心一科(刘 丹),重症医学科(边新渠)
摘要:
[摘要] 目的 筛选原发性胆汁性胆管炎(PBC)患者无创肝硬化诊断敏感指标。方法 收集2011—2020年在首都医科大学附属北京佑安医院住院确诊为PBC的患者380例。常规电子胃镜、血常规、凝血、肝功能及B超检查,瞬时弹性扫描仪(FibroScan)测定肝脏硬度,根据肝穿及临床诊断分为肝硬化组166例和非肝硬化组214例,比较分析两组指标的差异。随机按8∶2比例分为模型组和验证组,通过模型组筛选出敏感指标,应用随机森林法建立无创肝硬化诊断模型,然后使用验证组患者验证其诊断效果。结果 (1)筛选出红细胞比容(HCT)、红细胞计数、肝脏硬度、胆碱酯酶(CHE)、尿酸、天冬氨酸氨基转氨酶-丙氨酸转氨酶比值(AAR)、胆固醇(CHO)、血小板计数、凝血酶原活动度9个对PBC肝硬化影响效力最高的指标,采用随机森林方法建立模型,ROC曲线下面积为89.6%。9个指标中HCT影响权重最高。(2)肝硬化组HCT[(29.8±7.1)%]较非肝硬化组的(35.4±6.6)%明显降低,差异有统计学意义(P<0.001)。结论 HCT可以作为PBC患者无创肝硬化诊断模型中重要的指标之一,它无创且极易获得,检测难度低,外推性好,可能使部分患者避免肝活检。
关键词:  原发性胆汁性胆管炎  无创肝硬化模型  红细胞比容
DOI:10.3969/j.issn.1674-3806.2021.08.06
分类号:R 575.7
基金项目:北京市医院管理局消化内科学科协同发展中心项目(编号:XXT25);2018年度佑安肝病艾滋病基金资助院内中青年人才孵育项目(编号:YNKTTS20180109)
Screening of sensitive indicators for diagnosis of cirrhosis by a noninvasive method in patients with primary biliary cholangitis
YANG Xue, XU Bin, LIU Yan-min, et al.
Second Department of Liver Disease Center, Beijing Youan Hospital, Capital Medical University, Beijing 100069, China
Abstract:
[Abstract] Objective To screen the sensitive indicators for the diagnosis of cirrhosis by a noninvasive method in patients with primary biliary cholangitis(PBC). Methods A total of 380 patients diagnosed with PBC admitted to Beijing Youan Hospital, Capital Medical University from 2011 to 2020 were collected.The patients underwent routine electronic gastroscopy, blood routine, blood coagulation, liver function and B-ultrasound examinations, and the patients′ liver hardness was detected by FibroScan. According to the results of liver puncture and clinical diagnosis, the patients were divided into cirrhosis group(166 cases) and non-cirrhosis group(214 cases). The differences in the indicators between the two groups were compared and analyzed. The patients were randomly divided into the model group and the validation group according to a ratio of 8∶2. The sensitive indicators were screened out from the model group, and a noninvasive diagnostic model of liver cirrhosis was established by using random forest method, and then the efficacy of the diagnostic model was validated by using the patients in the validation group. Results (1)A total of 9 indicators with the highest impact effectiveness on PBC cirrhosis were screened, including hematocrit(HCT), erythrocyte count, liver hardness, cholinesterase(CHE), uric acid, ratio of aspartate aminotransferase to alanine aminotransferase(AAR), cholesterol(CHO), platelet count and prothrombin activity. The model was established by using random forest method. The area under the receiver operator characteristic(ROC) curve was 89.6%. Among the 9 indicators, HCT had the highest influence weight. (2)The HCT in the cirrhosis group[(29.8±7.1)%] was significantly lower than that in the non-cirrhosis group[(35.4±6.6)%], and the difference was statistically significant(P<0.001). Conclusion HCT can be used as one of the important indicators in the noninvasive diagnostic model of cirrhosis in PBC patients. It is noninvasive and easy to obtain, with low difficulty of detection and good extrapolation, and may make some patients avoid liver biopsy.
Key words:  Primary biliary cholangitis(PBC)  Noninvasive cirrhosis model  Hematocrit(HCT)