Department of Thyroid and Breast Surgery, General Hospital of Western Theater Command of Chinese People's Liberation Army, No.270 Rongdu Avenue, Chengdu, Sichuan, 610083, China.
Department of Thyroid and Breast Surgery, General Hospital of Western Theater Command of Chinese People's Liberation Army, No.270 Rongdu Avenue, Chengdu, Sichuan, 610083, China.
Eur J Surg Oncol. 2022 Jun;48(6):1272-1279. doi: 10.1016/j.ejso.2022.04.001. Epub 2022 Apr 6.
Preoperative status of central lymph nodes is a key determinant of the initial surgical extent for papillary thyroid carcinoma (PTC). We aimed to develop and validate a nomogram based on preoperative clinical characteristics and ultrasound features to predict central lymph node status in patients with clinically lymph node-negative (cN0) T1/T2 PTC.
This retrospective study included 729 patients with cN0T1/T2 PTC who were treated between January 2015 and March 2020. Based on the ratio of 6:4, 431 patients who underwent surgeries relatively earlier comprised the training set to develop the nomogram, while the other 298 who underwent surgeries relatively later comprised validation set to validate the performance of nomogram. Least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression were used to identify predictors of central lymph node metastasis (CLNM). These variables were used to construct a nomogram for predicting the risk of CLNM. The predictive performance, discriminative ability, calibration, and clinical utility of the nomogram model were evaluated in both sets.
A total of 313 (42.9%) PTC patients were identified with CLNM. On multivariate logistic regression analyses, malegender, younger age, larger maximum diameter, multifocality, capsular invasion, infiltrative margins, intra-nodular vascularity, and aspect ratio >1 were independent risk factors for CLNM. Nomogram integrating these 8 factors showed excellent discrimination in the training [area under the curve (AUC): 0.788] and validation (AUC: 0.829) sets, and obtained well-fitted calibration curves. The cut-off value of this nomogram was 0.410 (∼245 points). Decision curve analysis confirmed the clinical utility of the nomogram.
The CLNM-predicting nomogram can facilitate stratification of cN0T1/T2 PTC patients. Prophylactic central neck lymph node dissection can be considered for those with high nomogram scores.
中央区淋巴结的术前状态是甲状腺乳头状癌(PTC)初始手术范围的关键决定因素。我们旨在开发和验证一个基于术前临床特征和超声特征的列线图,以预测临床淋巴结阴性(cN0)T1/T2 PTC 患者的中央区淋巴结状态。
本回顾性研究纳入了 2015 年 1 月至 2020 年 3 月期间接受治疗的 729 例 cN0T1/T2 PTC 患者。根据 6:4 的比例,431 例较早接受手术的患者纳入训练集以开发列线图,而另外 298 例较晚接受手术的患者纳入验证集以验证列线图的性能。最小绝对收缩和选择算子(LASSO)回归和多变量逻辑回归用于识别中央淋巴结转移(CLNM)的预测因子。这些变量用于构建预测 CLNM 风险的列线图。在两个集合中评估了列线图模型的预测性能、判别能力、校准和临床实用性。
共 313 例(42.9%)PTC 患者存在 CLNM。多变量逻辑回归分析显示,男性、年龄较小、最大直径较大、多灶性、包膜侵犯、浸润性边缘、结节内血管和纵横比>1 是 CLNM 的独立危险因素。整合这 8 个因素的列线图在训练集[曲线下面积(AUC):0.788]和验证集(AUC:0.829)中表现出良好的判别能力,并获得了拟合良好的校准曲线。该列线图的截断值为 0.410(约 245 分)。决策曲线分析证实了该列线图的临床实用性。
该 CLNM 预测列线图有助于分层 cN0T1/T2 PTC 患者。对于那些列线图评分较高的患者,可以考虑预防性中央颈部淋巴结清扫。