Prediction of Extrathyroidal Extension of Papillary Thyroid Carcinoma via Nomogram Model Based on Conventional Ultrasound and Contrast-Enhanced Ultrasound Features
10.3969/j.issn.1005-5185.2025.08.006
- VernacularTitle:基于常规超声及超声造影特征构建列线图模型预测甲状腺乳头状癌被膜外侵犯
- Author:
Yiming CHENG
1
;
Yu LIN
;
Xu LI
;
Taohua GOU
;
Yan ZHANG
;
Yukun LUO
Author Information
1. 川北医学院,四川 南充 637000;解放军总医院第一医学中心超声诊断科,北京 100853
- Publication Type:Journal Article
- Keywords:
Thyroid cancer,papillary;
Ultrasonography;
Extranodal extension;
Nomograms;
Contrast media;
Forecasting
- From:
Chinese Journal of Medical Imaging
2025;33(8):827-833,847
- CountryChina
- Language:Chinese
-
Abstract:
Purpose To construct a nomogram model based on conventional ultrasound(US)and contrast-enhanced ultrasound(CEUS)features for predicting extrathyroidal extension(ETE)of papillary thyroid carcinoma,and to evaluate its diagnostic performance.Materials and Methods A retrospective analysis of clinical and ultrasound data from 715 papillary thyroid carcinoma patients in the First Medical Center of Chinese PLA General Hospital from January 2017 to December 2022 was conducted.The patients were divided into two groups based on the presence or absence of ETE.Univariate and Multivariate analyses was performed to identify independent risk factors associated with ETE.Three models were established:clinical,clinical+US and clinical+US+CEUS.The nomogram of the best model was constructed and validated.Results The model based on clinical+US+CEUS features performed the best,the area under the curve was 0.885.Multivariate analysis indicated that older age(OR=1.029,95%CI 1.011-1.047),higher body mass index(OR=1.108,95%CI 1.049-1.171),capsular contact<25%(OR=4.716,95%CI 2.079-10.701),capsular contact 25%-50%(OR=21.320,95%CI 8.240-55.160),capsular contact>50%(OR=24.045,95%CI 6.792-85.126),nodules adjacent to the lateral side(OR=4.265,95%CI 1.366-13.318),nodules adjacent to the medial side(OR=6.416,95%CI 2.067-19.920)and interrupted capsular enhancement(OR=6.044,95%CI 3.588-10.180)were independent risk factors for ETE(all P<0.05).Decision curve analysis and clinical impact curve indicated high net benefit and strong clinical utility of the model.Ten-fold cross-validation showed good model stability.Conclusion The nomogram model constructed based on US and CEUS features demonstrates good predictive performance and holds significant clinical utility.
