【摘要】 目的:分析术前导致甲状腺乳头状癌(papillary carcinoma of thyroid, PCT)患者颈部淋巴结转移(lymph node metastasis, LNM)的风险因素，构建列线图模型并评估模型效能。方法:纳入2020年4月至2024年6月于陕西省肿瘤医院经首次手术的210例患者，随机分为训练集150例，测试集60例，回顾性分析其临床资料、超声特征、术前血清甲状腺球蛋白(preoperative serum thyroglobulin, PS-Tg)及BRAF V600E基因突变情况，根据术后的病理结果分为淋巴结转移组与淋巴结未转移组，使用单、多因素logistic回归分析影响PTC患者颈部LNM的风险因素并构建列线图模型，采用受试者工作特征(receiver operating characteristic, ROC)曲线、校准曲线、决策曲线分析法(decision ccurve analysis, DCA)评估模型的诊断效能及临床净收益。结果:多因素Logistic回归分析显示，患者在年龄(OR=0.96,95%CI:0.92～0.99,P=0.01)、性别(OR=2.21,95%CI:0.86～5.69,P=0.099)、肿瘤多灶性(OR=5.05,95%CI:1.87～13.67,P=0.001)、肿瘤分布(OR=0.16,95%CI:0.03～1.01,P=0.051)方面差异有统计学意义。纳入上述指标构建列线图预测模型，并在训练集与验证集中验证，训练集与验证集ROC曲线的曲线下面积(area under curve, AUC)分别为0.748(95%CI:0.670～0.825)和0.635(95%CI:0.489～0.782),模型区分度良好，校准曲线与理想曲线基本符合，DCA曲线显示其临床净收益较高。结论:基于多项指标联合的列线图可对术前PTC患者颈部LNM风险进行预测，为临床诊治及患者的个体化指导提供一定帮助。更多还原

【Abstract】 Objective:To analyze the risk factors of cervical lymph node metastasis(LNM)in patients with papillary carcinoma of thyroid(PCT)before operation, and to construct a nomogram model and evaluate the efficacy of the model.Methods:A total of 210 patients who underwent first surgery in Shaanxi Cancer Hospital from April 2020 to June 2024 were enrolled and randomly divided into training set(150 cases)and validation set(60 cases).The clinical data, ultrasound features, preoperative serum thyroglobulin(PS-Tg)and BRAF V600E gene mutation were retrospectively analyzed.According to the postoperative pathological results, the patients were divided into lymph node metastasis group and lymph node non-metastasis group.Univariate and multivariate logistic regression were used to analyze the risk factors affecting cervical LNM in PTC patients and construct a nomogram model.Receiver operating characteristic(ROC)curve, calibration curve, and decision curve analysis(DCA)were used to evaluate the diagnostic efficacy and net clinical benefit of the model.Results:Multivariate logistic regression analysis showed that age(OR=0.96,95%CI:0.92～0.99,P=0.01),gender(OR=2.21,95%CI:0.86～5.69,P=0.099),tumor multifocality(OR=5.05,95%CI:1.87～13.67,P=0.001),nodule distribution(OR=0.16,95%CI:0.03～1.01,P=0.051).The above indicators were included to construct a nomogram prediction model, which was verified in the training set and validation set. The area under curve(AUC)of ROC curve in the training set and validation set were 0. 748(95%CI:0. 670 ～ 0. 825)and 0. 635(95% CI:0. 489 ～ 0. 782). The model had good discrimination,and the calibration curve was basically consistent with the ideal curve.Conclusion:The nomogram based on multiple indicators can predict the risk of cervical LNM in preoperative PTC patients,which can provide certain help for clinical diagnosis and treatment and individualized guidance for patients.更多还原