小肝癌合并微血管侵犯的术前危险因素分析及预测模型构建  

Preoperative risk factor analysis and prediction model establishment of microvascular invasion in small hepatocellular carcinoma

作　　者：高航 翁山耕[1] 张翔[1] 石铮[1] Gao Hang;Weng Shangeng;Zhang Xiang;Shi Zheng(Department of Hepatopancreatobiliary Surgery,the First Affiliated Hospital of Fujian Medical University,Fuzhou350004,China)

机构地区：[1]福建医科大学附属第一医院肝胆胰外科,福州350004

出　　处：《中华普通外科杂志》2019年第11期921-924,共4页Chinese Journal of General Surgery

摘　　要：目的分析小肝癌患者合并微血管侵犯(microvascular invasion,MVI)的相关术前危险因素并构建预测模型。方法回顾性分析2012年12月至2019年3月于福建医科大学附属第一医院157例行肝切除术的单个、肿瘤直径≤5 cm的肝细胞癌病例,其中合并MVI59例,无MVI98例。根据不同时间段将入组病例分为建模组(n=137)和验证组(n=20)。通过其术前特异性影像特征和临床资料,探讨小肝癌合并MVI的独立危险因素并构建预测模型。结果多因素Logistic回归分析显示:AFP、影像学测量肿瘤最大径、肿瘤边缘情况是小肝癌合并MVI的独立危险因素。根据上述因素构建术前预测模型:P=1/1+[e^(2.751-1.090X1-1.778X2-1.147X3)](X1-X3分别代表AFP、肿瘤最大径及肿瘤边缘情况的赋值)。模型的受试者工作特征曲线(receiver operating characteristic curve,ROC)下方面积的(area under curve,AUC)为0.786。结论基于AFP、影像学测量肿瘤最大径及肿瘤边缘情况等3项术前临床资料的小肝癌合并MVI术前预测模型,具有一定的临床实用价值。Objective To analyze relevant preoperative risk factors for microvascular invasion(MVI)in patients with small hepatocellular carcinoma and establish a prediction model.Methods A retrospective analysis of 157 cases of singlehepatocellular carcinoma with tumor diameter≤5 cm undergoing hepatectomy at the First Affiliated Hospital of Fujian Medical University from Dec 2012 to Mar 2019 was conducted.There were 59 cases with MVI and 98 cases without MVI.According to different time periods,the enrolled cases were divided into modeling group(n=137)and validation group(n=20).The independent risk factors of small hepatocellular carcinoma with MVI were discussed and a prediction model was established by using preoperative specific imaging characteristics and clinical data.Results Multivariate Logistic regression analysis showed that AFP,imaging tumor maximum diameter and tumor margin status were independent risk factors for small hepatocellular carcinoma complicated by microvascular invasion.Preoperative prediction model was established according to the above factors:P=1/1+[e^(2.751-1.090X1-1.778X2-1.147X3)].(X1-X3 represent the assignment of AFP,maximum tumor diameter and tumor margin state,respectively).The area under curve(AUC)and the receiver operating characteristic curve(ROC)of the model is 0.786.Conclusion Based on three preoperative clinical data,including AFP,tumor maximum diameter and tumor margin,the preoperative prediction model of small hepatocellular carcinoma complicated with MVI was successfully established with clinical value.

关 键 词：癌 肝细胞 微血管 预测模型 

分 类 号：R73[医药卫生—肿瘤]