基于超声图像和临床病理特征的早期浸润性乳腺癌腋窝淋巴结转移的诊断模型  

Diagnostic model of axillary lymph node metastasis of early invasive breast cancer based on ultrasound image and clinicopathological characteristics

作　　者：陆俊名 金华[1] 纪宗萍[1] LU Jun-ming;JIN Hua;JI Zong-ping(Department of Ultrasound,Zhongshan Hospital of Traditional Chinese Medicine,Zhongshan 528400,China)

机构地区：[1]广东省中山市中医院超声科,中山市528400

出　　处：《广西医学》2022年第4期380-384,共5页Guangxi Medical Journal

基　　金：广东省中山市社会公益与基础研究(2020SYF03)。

摘　　要：目的基于超声图像特征和临床病理特征构建早期浸润性乳腺癌腋窝淋巴结转移的诊断模型。方法纳入经术后病理确诊的156例早期浸润性乳腺癌患者,所有患者术前均接受超声检查。根据手术病理结果评估腋窝淋巴结转移的发生情况,将患者分为转移组47例和未转移组109例。比较两组超声图像特征和临床病理特征。采用多因素Logistic回归模型分析超声图像特征及临床病理特征与乳腺癌腋窝淋巴结转移的相关性,并建立诊断模型。通过受试者工作特征(ROC)曲线评估各个危险因素和诊断模型对乳腺癌患者发生腋窝淋巴结转移的诊断价值。结果两组患者的肿瘤大小、有无边缘毛刺征、皮质厚度、血管分级、组织学分级、肿瘤位置差异均有统计学意义(均P<0.05);有边缘毛刺征、皮质厚度≥3 mm、血管分级为Ⅲ级、组织学分级为Ⅱ级或Ⅲ级与患者发生腋窝淋巴结转移相关(均P<0.05)。皮质厚度、边缘毛刺征、血管分级、组织学分级诊断乳腺癌患者发生腋窝淋巴结转移的曲线下面积(AUC)为0.617~0.738,其中皮质厚度的AUC大于其他指标(均P<0.05)。基于多因素Logistic回归分析结果建立的诊断模型为P=1/(1+e^(0.524×边缘毛刺征+0.710×皮质厚度+0.582×血管分级+0.619×组织学分级-0.302)),该模型诊断乳腺癌患者发生腋窝淋巴结转移的AUC为0.836。结论有边缘毛刺征、皮质厚度≥3 mm、血管分级为Ⅲ级、组织学分级为Ⅱ级或Ⅲ级的早期浸润性乳腺癌患者存在腋窝淋巴结转移的可能性增大,基于上述超声图像及临床病理特征建立的模型对早期浸润性乳腺癌患者发生腋窝淋巴结转移的诊断价值较高。Objective To establish a diagnostic model for axillary lymph node metastasis of early invasive breast cancer based on ultrasound image and clinicopathological characteristics.Methods A total of 156 patients pathologically identified with early invasive breast cancer after the operation were enrolled.All patients undertook an ultrasound examination before the operation.According to the incidence of axillary lymph node metastasis assessed by surgical pathology,the patients were divided into metastasis group(n=47)and non-metastasis group(n=109).The ultrasound image and clinicopathological characteristics were compared between the two groups.The multivariate Logistic regression model was employed to analyze the correlations of ultrasound image and clinicopathological characteristics with axillary lymph node metastasis of breast cancer,and to establish a diagnostic model.The diagnostic values of various risk factors and the diagnostic model for the occurrence of axillary lymph node metastasis in patients with breast cancer were assessed by the receiver operating characteristic(ROC)curve.Results There were statistically significant differences between the two groups in the tumor size,incidence of marginal burr sign,cortical thickness,vascular grade,histological grade,and tumor location(all P<0.05);the incidence of marginal burr sign,cortical thickness not less than 3 mm,vascular gradeⅢ,and histological gradeⅡorⅢcorrelated with the occurrence of axillary lymph node metastasis in the patients(all P<0.05).For diagnosing the occurrence of axillary lymph node metastasis in patients with breast cancer,the areas under the curve(AUC)of cortical thickness,marginal burr sign,vascular grade,and histological grade ranged from 0.617 to 0.738,wherein cortical thickness had a greater AUC than the other indices(all P<0.05).The diagnostic model established based on the results of multivariate Logistic regression analysis was presented as follows:P=1/(1+e^(0.524×marginal burr sign+0.710×cortical thickness+0.582×vascular grade+0

关 键 词：超声图像特征 临床病理特征 乳腺癌 腋窝淋巴结转移 诊断模型 

分 类 号：R445.1[医药卫生—影像医学与核医学]