Predicting ultrasound-guided thermal ablation benefit in primary hyperparathyroidism

Objectives Ultrasound (US)-guided thermal ablation for primary hyperparathyroidism (PHPT) is a relatively novel minimally invasive treatment. The recurrence rate after ablation is between 10 and 15%. The characteristics of patients who can benefit from thermal ablation therapy are not clear yet. The aim of this research was to investigate the validity of a parathyroid hormone (PTH)-based classifier for stratifying patients with PHPT. Methods A total of 171 patients were screened, 148 (86.5%) of whom were eligible and were divided into development (n = 104) and external validation (n = 44) cohorts. The potential relationship between the PTH-based classifier and the cure rate of patients was initially assessed in the primary cohort and then validated in the external validation cohort. The nomogram was computed from the logistic regression model. Results A cut-off of PTH < 269.1 pg/mL or >= 269.1 pg/mL as the optimal prognostic threshold in the training cohort was generated to stratify the patients into low-risk and high-risk groups. Patients with PTH levels < 269.1 pg/mL in the training cohort had a higher cure rate than patients with PTH levels >= 269.1 pg/mL (p < 0.001). The PTH level remained the strongest predictor of the cure rate in all cohorts. Furthermore, a nomogram based on the PTH level was developed to predict the cure rate in the training cohort and it performed well in the external validation cohort (AUC: 0.816, 95%CI 0.703 to 0.930; AUC: 0.816, 95%CI 0.677 to 0.956). Conclusions The PTH-based classifier may help with individualised treatment planning for selecting patients who may benefit from thermal ablation.