基于超声、钼靶及病理分级的联合诊断模型鉴别乳腺导管原位癌伴浸润成分的价值  

Value of a combined diagnosis model based on ultrasound,mammography and pathological grading in the identification of ductal carcinoma in situ with invasive components

作　　者：孙晓岚[1] 戴宇萍[2] 章骏[3] 陈真[2] SUN Xiaolan;DAI Yuping;ZHANG Jun;CHEN Zhen(Department of Ultrasound,Shidong Hospital,Yangpu District,Shanghai 200438,China)

机构地区：[1]上海市杨浦区市东医院超声科,200438 [2]上海市杨浦区市东医院病理科,200438 [3]上海市杨浦区市东医院外二科,200438

出　　处：《临床超声医学杂志》2021年第8期586-590,共5页Journal of Clinical Ultrasound in Medicine

基　　金：上海市卫生和计划生育委员会科研课题基金(20164Y0276)。

摘　　要：目的建立基于常规超声及声辐射力脉冲弹性成像(ARFI)联合钼靶、病理分级的联合诊断模型,探讨其鉴别乳腺导管原位癌(DCIS)伴浸润成分的临床价值。方法选取我院经病理证实的DCIS患者112例,其中无浸润成分79例(单纯DCIS组),伴浸润成分33例(DCIS-I组)。比较两组患者的病理结果、钼靶、超声特征及剪切波速度(SWV)的差异。应用Logistic回归建立多指标联合诊断DCIS伴浸润成分的模型;绘制受试者工作特征曲线分析各指标单独及联合鉴别DCIS伴浸润成分的诊断效能。结果DCIS-I组病理等级为高级的比例(24.2%)、钼靶检查中微钙化伴肿物比例(33.3%)、肿瘤直径[(3.72±0.83)cm]及SWV[(7.03±2.11)m/s]均高于单纯DCIS组[6.3%、10.1%、(2.07±0.62)cm、(5.98±1.43)m/s],差异均有统计学意义(均P<0.05);且均为鉴别DCIS伴浸润成分的独立预测因子(OR=6.744、4.332、1.356、1.621,均P<0.05)。建立多指标联合诊断DCIS伴浸润成分的模型:Logit(P)=-11.205+0.483*SWV+0.305*肿瘤直径+1.466*微钙化伴肿物+1.909*病理等级。该模型鉴别DCIS伴浸润成分的曲线下面积为0.923,显著高于病理等级、微钙化伴肿物、直径、SWV单独鉴别的曲线下面积(0.710、0.616、0.728、0.826),差异均有统计学意义(均P<0.05)。结论基于常规超声、ARFI联合钼靶及病理分级建立的联合诊断模型能够准确鉴别DCIS是否伴浸润成分。ObjectiveTo establish a combined diagnosis model based on ultrasound,mammography and pathological grading,and to explore its value in the identification of ductal carcinoma in situ(DCIS)with invasive components.MethodsA total of 112 patients with DCIS confirmed pathologically in our hospital were enrolled,79 cases without invasive components(DCIS group),and 33 cases with invasive components(DCIS-I group).The differences in pathological results,mammography,ultrasonic characteristics,and shear wave velocity(SWV)between the two groups were compared.Logistic regression was used to establish a combined diagnosis model of DCIS with invasive components.The receiver operating characteristic curve were drawn to analyze the value of each predictor alone and in combination in identifying DCIS with invasive components.ResultsIn the DCIS-I group,the proportion of advanced pathological grades(24.2%),the proportion of microcalcifications with tumors in mammography(33.3%),tumor diameter measured by ultrasound[(3.72±0.83)cm]and SWV[(7.03±2.11)m/s]were higher than those in the DCIS group[6.3%,10.1%,(2.07±0.62)cm,and(5.98±1.43)m/s].There were significant differences(all P<0.05).They were the independent predictors for identifying DCIS with invasive components(OR=6.744,4.332,1.356,1.621,all P<0.05).Logit(P)=-11.205+0.483*SWV+0.305*diameter+1.466*microcalcification with tumors+1.909*pathological grade.The area under the curve(AUC)of the combined diagnostic model for identifying DCIS with invasive components was 0.923,which was significantly higher than that of the AUC by each index alone(pathological grade:0.710,microcalcification with tumor:0.616,diameter:0.728,SWV:0.826,all P<0.05).Conclusion ultrasound,ARFI combined with mammography and pathological grading can accurately identify DCIS with invasive components.

关 键 词：声辐射力脉冲弹性成像 超声检查 钼靶 乳腺导管原位癌 浸润成分 

分 类 号：R445.1[医药卫生—影像医学与核医学] R737.9[医药卫生—诊断学]