探究超声联合肿瘤标记物对cN0期甲状腺微小乳头状癌(PTMC)颈中央区淋巴结转移(CLNM)的临床预测价值。

回顾性选取2018年5月至2020年4月行预防性中央区淋巴结清扫术的187例cN0-PTMC患者分为CLNM转移阳性组(+)和阴性组(-)。采用SPSS 19.0进行数据分析，计量资料采用(±s)表示，计数资料采用n(%)表示，P<0.05为检验标准。经单因素及Logistic回归多因素分析CLNM转移独立危险因素，建立联合预测模型，绘制ROC曲线，计算AUC值，评估预测模型诊断效能。

187例cN0-PTMC患者CLNM转移率为36.4%(68/187)。经单、多因素分析显示，肿瘤直径>7 mm、BRAF^V600E基因突变阳性、多癌灶及包膜侵犯是cN0-PTMC患者发现CLNM的独立危险因素(P<0.05)。通过建立联合预测模型，当Y取0.722时诊断价值最高，灵敏度为88.9%，特异度为83.3%。Hosmer-Lemeshow拟合优度检验预测模型的P值为0.579，该模型预测CLNM的结果与术后病理确诊的一致性良好，Kappa＝0.721，P＝0.000。

超声特征联合肿瘤分子标记物预测cN0-PTMC发生CLNM具有良好的诊断灵敏度及特异度，可为临床术式的选择提供指导依据。

To investigate the clinical predictive value of ultrasound combined with tumor markers for lymph node metastasis (CLNM) in the central cervical region of cN0 thyroid micropapillary carcinoma (PTMC).

A retrospective study of 187 patients with CN0-PTMC who underwent preventive central lymph node dissection from May 2018 to April 2020 were retrospectively selected and divided into two groups: CLNM metastatic positive group (+ ) and the NegativeEGATIVE group (-). SPSS 19.0 was used for data analysis., Mmeasurement data were expressed as (±s), counting data as n(%), and P<0.05 was the test standard. The singleindependent risk factor s of CLNM transfer were analyzed by univariate and Logistic regression factor were used to analyze the independent risk factors of CLNM transfer. s, Tthe joint prediction model was established, the ROC curve was drawn, the AUC value was calculated, and the diagnostic efficiency of the prediction model was evaluated.

The CLNM metastasis rate of 187 patientsCLNM in 187with CN0-PTMC patients was 36.4%(68/187). Single and multivariate analysis showed that tumor diameter of > 7 mm, positive BRAF^V600E mutation, multiple foci and capsule invasion were independent risk factors for CLNM detection in patients with cn0-ptmc (P<0.05). By establishing the joint prediction model, when Y is 0.722, the diagnostic value is the highest, the sensitivity is 88.9%, and the specificity is 83.3%. The Predictive model of Hosmer-Lemeshow goodness of fit test showeshad a P value of 0.579, and the indicating that the predicted results of CLNM by the model were in good agreement with postoperative pathological diagnosis, Kappa=0.721, P=0.000.

Ultrasonographic features combined with tumor molecular markers canto predict the occurrence of CLNM of CN0-PTMC have with good diagnostic sensitivity and specificity, which can provide guidance for the selection of clinical operation.