Combined Conventional Ultrasound and Contrast‐Enhanced Computed Tomography for Cervical Lymph Node Metastasis Prediction in Papillary Thyroid Carcinoma

Abstract

Objectives-This study aimed to evaluate conventional ultrasound (US) combined with contrast-enhanced computed tomography (CT) of the neck to predict central lymph node metastasis (CLNM) in clinical lymphnegative patients with papillary thyroid carcinoma (PTC), establish a simple preoperative risk-scoring model, and validate its effectiveness in a two-center dataset.Methods-A total of 423 patients with PTC preoperatively evaluated by US and contrast-enhanced CT were included in the modeling group, and 102 patients from two hospitals were enrolled in the validation group. Independent predictive factors were determined using multivariate logistic regression analysis. Diagnostic performance was evaluated using receiver operating characteristic curve analysis.Results-The independent predictive factors for CLNM were age ≤45 years (odds ratio [OR] = 3.950), nodule presence in the non-upper pole (OR = 2.385), nodule size >12.5 mm (OR = 2.130), Thyroid Imaging Reporting and Data System score ≥9 (OR = 2.857), normalized enhancement CT value ≥0.75 (OR = 3.132), central enhancement (OR = 0.222), and capsular invasion (OR = 3.478). The area under the curve (AUC) of the model was 0.790 (95% confidence interval [CI]: 0.747-0.834), and the sensitivity and specificity were 70.4% and 73.9%, respectively. The AUC in the validation group was 0.827 (95% CI: 0.747-0.907), and the sensitivity and specificity were 88.9% and 63.2%, respectively.Conclusions-We found conventional US combined with contrast-enhanced CT of the neck to be useful in predicting CLNM preoperatively and established a simple risk-scoring model that might help surgeons with appropriate surgical plans and prognostic evaluation.

A LASSO-based model to predict central lymph node metastasis in preoperative patients with cN0 papillary thyroid cancer

“…Park et al ( 33 ) aimed to develop a radiomics signature using US images of the primary tumor to preoperatively predict LNM in patients with conventional PTC. Zhao et al ( 34 ) used independent predictive factors, followed by multivariate logistic regression, to evaluate risk factors by using ROC curve analysis. Most of these articles included the characteristics of US-related factors and some clinical features.…”

Multimodal Diverse Granularity Fusion Network based on US and CT Images for Lymph Node Metastasis Prediction of Thyroid Carcinoma

“…Considering that US and CT provide complementary information and are widely utilized in thyroid cancer diagnosis, there's a great potential to improve the performance by integrating US and CT images through a multimodal approach for predicting LNM status. For instance, Zhao et al developed a multivariate logistic regression multimodal model for predicting central LNM status by incorporating clinical factors, US-derived diagnostic features, and CT measurements, achieving an AUC of 0.827 [17].…”

“…Considering that US and CT provide complementary information and are widely utilized in thyroid cancer diagnosis, there’s a great potential to improve the performance by integrating US and CT images through a multimodal approach for predicting LNM status. For instance, Zhao et al developed a multivariate logistic regression multimodal model for predicting central LNM status by incorporating clinical factors, US-derived diagnostic features, and CT measurements, achieving an AUC of 0.827[17]. Nevertheless, the study did not directly compare the multimodal method’s performance with unimodal methods, besides, it utilized a simplified model that overlooked the interaction between the two modalities, leaving the full potential of multimodal fusion approach unexplored.…”