Predicting Cervical Lymph Node Metastasis Using Preoperative Multiparameter Data Based on Imaging, Serology, and Clinical Features in Unifocal Papillary Thyroid Carcinoma

Objective: To evaluate the predictive performance of preoperative multiparameter data, including computed tomography (CT), ultrasound, clinical features, and serology, for central cervical lymph node metastasis (CLNM) in unifocal papillary thyroid carcinoma (PTC) and to develop a nomogram model for predicting CLNM preoperatively. Methods: Data were collected from 340 consecutive patients with pathologically confirmed unifocal PTC who underwent radical thyroidectomy between January 2019 and December 2023. Radiologists with varying levels of experience analyzed the multiparameter data, including clinical features, CT, ultrasound, and thyroid function tests. The correlation between these characteristics and CLNM was assessed using univariate analysis, and independent risk factors (IRFs) were identified via forward selection in multivariate logistic regression analysis. The patients were randomly divided into training and validation cohorts at a 7:3 ratio. A nomogram model for predicting CLNM was established using IRFs from the training cohort and tested on the validation cohort. The predictive performance of the nomogram was evaluated using receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). Results: A total of 20 characteristics, including two clinical features, four CT features, seven ultrasound features, and seven thyroid hormone levels, were recorded and analyzed. Age ≤55 years, male sex, maximum lesion diameter >10 mm (ultrasound), capsular contact >0 (CT), and irregular tumor margin (ultrasound) were identified as IRFs for CLNM. The area under the curve of the nomogram ROCs in the training and validation cohorts was 0.815 (95% CI: 0.761–0.870) and 0.747 (95% CI: 0.646–0.848), respectively. The calibration curve and DCA showed excellent utility of the nomogram for predicting CLNM. Conclusions: The multiparameter characteristics of unifocal PTC can effectively predict CLNM. The constructed nomogram can assist in clinical decision-making and benefit patients with PTC.