To explore the value of combined use of ultrasonographic indicators in predicting central lymph node metastasis in PTC.

Clinical data of PTC patients (n=446) with thyroid surgery and central neck dissection were collected retrospectively from the First Affiliated Hospital of Shihezi University School of Medicine from March 2013 to May 2020, involving sex, age, and ultrasonographic features 〔diameter of nodule, number of nodules, location of single and multiple nodules, internal composition, internal echo, boundary, margin, micro calcification, anteroposterior and transverse diameter (A/T) ratio, posterior echo attenuation, capsular invasion〕. Postoperative pathological results confirmed that 159 cases (35.65%) with central lymph node metastasis and other 287 (64.35%) without. Unconditional stepwise binary Logistic regression was used to explore the factors associated with central lymph node metastasis. Receiver operating characteristic (ROC) analysis was performed to estimate the predictive value of each indicator for central lymph node metastasis.

Patients with and without central lymph node metastasis had significant differences in sex ratio, distribution of age, thyroid nodule diameter, margin conditions, and A/T ratio, as well as prevalence of micro calcification and capsular invasion (P<0.05) . Binary Logistic regression analysis showed that male 〔OR=1.727, 95%CI (1.059, 2.816) 〕, <45 years old 〔OR=2.690, 95%CI (1.728, 4.187) 〕, diameter of the thyroid nodule >10 mm 〔OR=2.385, 95%CI (1.544, 3.684) 〕, and capsular invasion 〔OR=1.773, 95%CI (1.153, 2.724) 〕 were independently associated with increased risk of central lymph node metastasis (P<0.05) . And the Logistic regression algorithm was: Logit (P) =-1.627+0.546×sex +0.989×age+0.869× (thyroid nodule diameter) +0.572× (capsular invasion) . In predicting central lymph node metastasis, the AUC of combination of four indicators in the algorithm was 0.695〔95%CI (0.650, 0.738) 〕, with 67.92% sensitivity, and 61.67% specificity, and 0.32 optimal cut-off value. The AUC of the combination of four indicators was larger than that of sex, age, thyroid nodule diameter, or capsular invasion alone (Z=4.137, 3.682, 3.070, 3.679, P<0.05) .

Being male, age <45 years old, thyroid nodule diameter>10 mm, and capsular invasion were independent risk factors for central lymph node metastasis in PTC, and the value of combined use of them had a higher predictive value for central lymph node metastasis.

探讨超声检查指标联合预测PTC颈部CLNM的价值。

回顾性收集2013年3月至2020年5月于石河子大学医学院第一附属医院行甲状腺手术及颈部中央区淋巴结清扫的446例PTC患者的临床资料，包括性别、年龄及超声检查指标（结节直径、结节数目、单发位置、多发部位、内部成分、内部回声、边界、边缘、微钙化、纵横比、后方回声衰减、被膜侵犯）。依据患者术后病理结果将其分为发生颈部CLNM 159例（35.65%），未发生颈部CLNM 287例（64.35%）。采用二元逐步非条件Logistic回归分析探究PTC患者发生颈部CLNM的影响因素。绘制各指标预测PTC患者发生颈部CLNM的受试者工作特征（ROC）曲线。

不同性别、年龄、结节直径、边缘、微钙化、纵横比、被膜侵犯PTC患者颈部CLNM发生率比较，差异有统计学意义（P<0.05）。二元Logistic回归分析结果显示，男性〔OR=1.727，95%CI（1.059，2.816）〕、年龄<45岁〔OR=2.690，95%CI（1.728，4.187）〕、结节直径>10 mm〔OR=2.385，95%CI（1.544，3.684）〕、被膜侵犯〔OR=1.773，95%CI（1.153，2.724）〕是PTC患者发生颈部CLNM的独立危险因素（P<0.05），回归方程：Logit（P）=-1.627+0.546×性别+0.989×年龄+0.869×结节直径+0.572×被膜侵犯。Logit（P）联合指标预测PTC患者发生颈部CLNM的灵敏度为67.92%、特异度为61.67%（最佳截断值为0.32），ROC曲线下面积（AUC）为0.695〔95%CI（0.650，0.738）〕；联合指标预测PTC患者发生颈部CLNM的AUC分别大于性别、年龄、结节直径、被膜侵犯单独预测（Z值分别为4.137、3.682、3.070、3.679，P值均<0.05）。

男性、年龄<45岁、结节直径>10 mm、被膜侵犯是PTC患者发生颈部CLNM的独立危险因素，且上述4种指标联合预测PTC患者发生颈部CLNM的价值较高。