基于术前常规超声及超声造影特征的列线图模型预测甲状腺乳头状癌颈部淋巴结转移的临床价值  

Clinical value of nomogram model based on preoperative conventional ultrasound and contrast-enhanced ultrasound features in predicting cervical lymph node metastasis of papillary thyroid cancer

作　　者：张宇[1] 黄九平 范莉芳 汪珺莉 ZHANG Yu;HUANG Jiuping;FAN Lifang;WANG Junli(Department of Ultrasound Medicine,Wuhu Hospital Affiliated to East China Normal University,Anhui 241000,China)

机构地区：[1]华东师范大学附属芜湖医院,芜湖市第二人民医院超声医学科,安徽省芜湖市241000 [2]北京大学第三医院超声医学科 [3]皖南医学院医学影像学院

出　　处：《临床超声医学杂志》2023年第8期639-644,共6页Journal of Clinical Ultrasound in Medicine

摘　　要：目的 探讨基于术前常规超声及超声造影特征的列线图模型预测甲状腺乳头状癌(PTC)颈部淋巴结转移的临床价值。方法 选取我院及北京大学第三医院经病理确诊的PTC患者216例,其中颈部淋巴结转移者54例(CLNM组),非转移者162例(NCLNM组),比较两组术前临床资料、常规超声及超声造影特征;应用单因素、多因素Logistic回归分析筛选预测PTC颈部淋巴结转移的独立危险因素。基于独立危险因素构建联合预测模型,并以列线图形式展示;绘制受试者工作特征(ROC)曲线分析模型的预测效能。采用Hosmer-Lemeshow检验及校准曲线评价模型的拟合优度与校准度;临床决策曲线评价模型的临床适用性;5折交叉验证法评价模型的稳定性。结果 两组性别、穿刺手感、多灶性、最大径、微钙化、血流信号、增强均匀性、增强模式及浅方被膜连续性比较差异均有统计学意义(均P<0.05);单因素、多因素Logistic回归分析显示,性别、多灶性、最大径、血流信号、增强均匀性、增强模式、浅方被膜连续性均为预测PTC颈部淋巴结转移的独立危险因素(OR=2.792、3.482、2.730、2.581、3.096、3.651、3.807,均P<0.05)。ROC曲线分析显示,列线图模型预测PTC颈部淋巴结转移的曲线下面积(AUC)为0.822;Hosmer-Lemeshow检验显示模型拟合较好(χ2=8.298,P=0.405);校准曲线显示预测值与实测值的一致性较高;临床决策曲线分析显示,列线图模型在所有模型中临床适用性最佳。5折交叉验证法的AUC分别为0.871(95%可信区间:0.781~0.961)、0.900(95%可信区间:0.840~0.959)、0.707(95%可信区间:0.581~0.834)、0.678(95%可信区间:0.556~0.800)、0.762(95%可信区间:0.654~0.870),平均AUC为0.784。结论 基于术前常规超声及超声造影特征的列线图模型能够有效预测PTC颈部淋巴结转移。Objective To investigate the clinical value of nomogram model based on preoperative conventional ultrasound and contrast-enhanced ultrasound features in predicting cervical lymph node metastasis of papillary thyroid carcinoma(PTC).MethodsA total of 216 patients with PTC confirmed by pathology in our hospital and Peking University Third Hospital were selected,including 54 patients with cervical lymph node metastasis(CLNM group)and 162 patients without metastasis(NCLNM group).Preoperative clinical data,conventional ultrasound and contrast-enhanced ultrasound features between the two groups were compared.Univariate analysis and multivariate Logistic regression analysis were used to screen independent risk factors for cervical lymph node metastasis of PTC.A prediction model based on independent risk factors was constructed,and display it in the form of nomogram.Receiver operating characteristic(ROC) curve was drawn to analysis the predictive performance of the model,Hosmer-Lemeshow test and calibration curve were drawn to evaluate the goodness of fit and calibration of the model.The clinical applicability of the model was evaluated by clinical decision curve.The 5-fold crossvalidation were used to evaluate the stability of the model.ResultsThere were significant differences in gender,puncture feel,multifocality,maximum diameter,microcalcification,blood flow signal,enhancement uniformity,enhancement mode and superficial capsule between the two groups(all P<0.05).Univariate analysis and multivariate Logistic regression analysis showed that gender,multifocality,maximum diameter,blood flow signal,enhancement uniformity,enhancement mode and superficial capsule were all independent risk factors for screening cervical lymph node metastasis of PTC(OR=2.792,3.482,2.730,2.581,3.096,3.651,3.807,all P<0.05).ROC curve analysis showed that the area under the curve(AUC)of the nomogram model in predicting cervical lymph node metastasis of PTC was 0.822.Hosmer-Lemeshow test showed that the model fitted well(χ2=8.298,P=0.405).The cal

关 键 词：超声检查 造影剂 乳头状癌 甲状腺 颈部淋巴结转移 列线图 

分 类 号：R445.1[医药卫生—影像医学与核医学] R736.1[医药卫生—诊断学]