LAUSR

Objectives To develop a nomogram based on postoperative clinical and ultrasound findings to quantify the probability of central compartment lymph node metastases (CLNM). Methods A total of 952 patients with histologically confirmed papillary thyroid carcinoma (PTC) were included in this retrospective study and assigned to three groups based on sex and age. The strongest predictors for CLNM were selected according to ultrasound imaging features, and an ultrasound (US) signature was constructed. By incorporating clinical characteristics, a predictive model presented as a nomogram was developed, and its performance was assessed with respect to calibration, discrimination and clinical usefulness. Results Predictors contained in the nomogram included US signature, US-reported LN status and age. The US signature was constructed with tumour size and microcalcification. The nomogram showed excellent calibration in the training dataset, with an AUC of 0.826 (95% CI, 0.765–0.887) for male patients, 0.818 (95% CI, 0.746–0.890) for young females and 0.808 (95% CI, 0.757–0.859) for elder females. For male and young female patients, application of the nomogram to the validation cohort revealed good discrimination, with AUCs of 0.813 (95% CI, 0.722–0.904) and 0.814 (95% CI, 0.712–0.915), respectively. Conversely, for elderly female patients, the nomogram failed to show good performance with an AUC of 0.742 (95% CI, 0.661–0.823). Conclusion This ultrasound-based nomogram may serve as a useful clinical tool to provide valuable information for treatment decisions, especially for male and younger female patients. Key Points • Age, gender, US-reported LN status and US signature were the strongest predictors of CLNM in PTC patients and informed the development of a predictive nomogram. • Microcalcification was the strongest predictor in the US signature, as CLMN was identified in approximately 92% of patients characterised by diffuse microcalcification. • Stratified by sex and age, this nomogram achieved good performance in predicting CLNM, especially in male and young female patients. This prediction tool may be useful as an imaging marker for identifying CLNM preoperatively in PTC patients and as a guide for personalised treatment.