Development and performance evaluation of a risk assessment model for cervical lymph node metastasis in patients with Hashimoto's thyroiditis coexisting with papillary thyroid carcinoma.

Background: The incidence of coexisting Hashimoto's thyroiditis (HT) and papillary thyroid carcinoma (PTC) is rising. The risk of cervical lymph node metastasis (CLNM) in HT-PTC patients remains controversial, and overlapping ultrasonographic features of lymph nodes complicate accurate diagnosis. Therefore, this study aimed to construct a risk assessment model for CLNM in patients with concurrent HT and PTC, and to evaluate its clinical application value. Methods: A total of 261 patients with postoperative pathologically confirmed PTC coexisting with HT who underwent thyroid surgery at the Affiliated Hospital of Guizhou Medical University from January 2018 to December 2023 were retrospectively enrolled. Univariate and multivariate logistic regression analyses were first performed to identify independent predictors of CLNM. Variables with P<0.05 in the multivariate analysis were incorporated into model development. Four models were constructed: based on tumor nodule ultrasound features (TN-US), lymph node ultrasound features (LN-US), their combination (TN-LN-US), and a full-feature model integrating clinical, laboratory, and ultrasound parameters (Clin-US). Receiver operating characteristic (ROC) curves were plotted to evaluate diagnostic performance, and the DeLong test was used to compare the area under the ROC curve (AUC) among the models. Internal validation was conducted using the bootstrap method, and model performance was further assessed using calibration curves, decision curve analysis (DCA), and clinical impact curve (CIC). Results: Among the 261 patients with PTC coexisting with HT, univariate and multivariate logistic regression analyses identified age, alkaline phosphatase, primary tumor microcalcifications, primary tumor vascularity, lymph node morphology, intranodal cortical echogenicity, and lymph node microcalcifications as independent risk factors for CLNM (P<0.05). The Clin-US model demonstrated the best diagnostic performance (AUC =0.842, bootstrap 95% confidence interval: 0.792-0.885). Bootstrap internal validation indicated stable model parameters. Calibration curves showed good model fit, with predicted probabilities closely matching actual incidence. DCA and CIC further confirmed favorable clinical applicability of the model. Conclusions: We developed a Clin-US model that combines clinical data and ultrasound for predicting lymph node metastasis in PTC patients with concurrent HT. The model demonstrated promising diagnostic performance and may assist clinicians in preoperative risk stratification, facilitating individualized treatment decisions for this population.