To address the clinical bottleneck in preoperative assessment of central lymph node metastasis (CLNM) in patients with C-TIRADS (Chinese Thyroid Imaging Reporting And Data System) 4 papillary thyroid carcinoma (PTC), this study aims to develop a preoperative predictive model capable of quantifying individual metastasis risk through integrated multifactorial analysis, thereby guiding more precise surgical planning. A retrospective analysis was conducted on the clinical-pathological characteristics and ultrasound features of 196 patients with C-TIRADS 4 PTC who underwent thyroid surgery at the Lequn Campus of the First Hospital of Jilin University between January 2021 and December 2023. Univariate and multivariate binary logistic regression analyses were employed to identify independent risk factors for CLNM. The diagnostic performance of the predictive model was evaluated, and a nomogram was developed to forecast the risk of thyroid cancer metastasis. This study included a total of 196 patients, with a CLNM incidence rate of 47.45%. Multivariate logistic regression analysis revealed that age < 55 years ( P = .024), multiple nodules ( P = .033), nodular borders with poor definition ( P = .001), microcalcifications ( P = .002), very hypoechoic ( P = .041), and capsular contact ( P < .001) were independent risk factors for CLNM. The established logistic regression model for diagnosing CLNM yielded an AUC of 0.791, with sensitivity and specificity of 62.4% and 82.5%, respectively. This study developed a nomogram for predicting thyroid cancer metastasis risk based on retrospective data, integrating 6 independent predictors: patient age, number of lesions, margin characteristic, microcalcification, very hypoechoic, and capsular contact. The predictive model demonstrated good performance for predicting CLNM.
Hu et al. (Fri,) studied this question.