Predicting lymph node metastasis in papillary thyroid carcinoma with Hashimoto’s thyroiditis using regression and network analysis

Abstract

Abstract The comprehensive study of the relationship between lymph node metastasis (LNM) and its associated factors in patients with concurrent papillary thyroid carcinoma (PTC) and Hashimoto’s thyroiditis (HT) remains insufficient. Building upon the initial investigation of factors associated with LNM in patients with concurrent PTC and HT, we further analyzed the complex relationships between different severity indicators of LNM and these associated factors. This study included patients confirmed PTC with HT who underwent total thyroidectomy at Xiangya Hospital, from January 2020 to December 2021. A total of 271 patients from 2020 were used as the training set, and 300 patients from 2021 as the validation set. Univariate analysis and regression modeling were used to identify key factors associated with LNM. Model reliability was assessed using the area under the receiver operating characteristic curve (AUC). Network analysis was employed to explore associations between LNM severity and its related factors. The regression model indicated that age, calcification, free triiodothyronine (FT3), and tumor maximum diameter (TMD) are independent factors for LNM. The severity model showed free thyroxine (FT4) and hemoglobin (Hb) are independent protective factors for the region and quantity of LNM, respectively, while TMD is an independent risk factor for both. Network analysis revealed TMD has a closer relationship with LNM severity compared to other associated factors. This study innovatively combined regression models and network analysis to investigate factors related to LNM in patients with PTC and HT, providing a theoretical basis for predicting preoperative LNM in future clinical practice.

Keywords

• Papillary thyroid carcinoma (PTC)
• Hashimoto’s Thyroiditis (HT)
• Lymph Node Metastasis (LNM)
• Biochemical markers
• Regression modeling
• Network Analysis