Related Experiment Video
Updated: Aug 15, 2025
Computer-Aided Three-Dimensional Visualization in the Treatment of Locally Advanced Thyroid Cancer
Published on: June 9, 2023
A model based on clinical data and multi-modal ultrasound for predicting cervical lymph node metastasis in patients
Bin Wang1, Qing Cao1, Xin-Wu Cui2
1Department of Medical Ultrasound, Yueyang Central Hospital, Yueyang, China.
Objective:
The aim of this study was to explore diagnostic performance based on clinical characteristics, conventional ultrasound, Angio PLUS (AP), shear wave elastography (SWE), and contrast-enhanced ultrasound (CEUS) for the preoperative evaluation of cervical lymph node metastasis (CLNM) in patients with papillary thyroid carcinoma (PTC) and to find a reliable predictive model for evaluating CLNM.
Materials And Methods:
A total of 206 thyroid nodules in 206 patients were included. AP, SWE, and CEUS were performed for all thyroid nodules. Univariate analysis and multivariate logistic regression analysis were performed to ascertain the independent risk factors. The sensitivity, specificity, and the area under the curve (AUC) of independent risk factors and the diagnostic model were compared.
Results:
Sex, age, nodule size, multifocality, contact extent with adjacent thyroid capsule, Emax, and capsule integrity at CEUS were independent risk predictors for CLNM in patients with PTC. A predictive model was established based on the following multivariate logistic regression: Logit (p) = -2.382 + 1.452 × Sex - 1.064 × Age + 1.338 × Size + 1.663 × multifocality + 1.606 × contact extent with adjacent thyroid capsule + 1.717 × Emax + 1.409 × capsule integrity at CEUS. The AUC of the predictive model was 0.887 (95% CI: 0.841-0.933), which was significantly higher than using independent risk predictors alone.
Conclusion:
Our study found that male presence, age < 45 years, size ≥ 10 mm, multifocality, contact extent with adjacent thyroid capsule > 25%, Emax ≥ 48.4, and interrupted capsule at CEUS were independent risk predictors for CLNM in patients with PTC. We developed a diagnostic model for predicting CLNM, which could be a potentially useful and accurate method for clinicians; it might be beneficial to surgical decision-making and patient management and for improving prognosis.
