[A nomogram for predicting the risk of postoperative symptomatic hypocalcemia following parathyroidectomy in patients with primary hyperparathyroidism]

Objective: To construct and validate a nomogram to predict the risk of symptomatic hypocalcemia after parathyroidectomy in patients with primary hyperparathyroidism (PHPT). Methods: This is a retrospective case series study. The clinical data of 360 patients who underwent parathyroidectomy for PHPT at the Department of General Surgery, Tianjin Medical University General Hospital, from July 2014 to July 2024 were retrospectively collected. There were 76 male and 284 female patients, with an age (M(IQR))of 57 (16) years (range: 17 to 89 years). Of these patients, 66 developed postoperative symptomatic hypocalcemia, whereas 294 did not. The collected variables included sex, age, preoperative symptoms, laboratory parameters, pathological findings of the resected parathyroid glands, and the type of neck exploration. Binary logistic regression analysis was performed to evaluate the associations between postoperative symptomatic hypocalcemia and these variables. A nomogram was constructed to predict the risk of symptomatic hypocalcemia and was internally validated using bootstrap resampling. Results: Univariable analysis showed that age (P<0.01), preoperative parathyroid hormone (PTH) level (P<0.01), preoperative alkaline phosphatase (ALP) level (P<0.01), 25-hydroxyvitamin D deficiency or not (P<0.01), the presence of preoperative clinical symptoms or not (P<0.01), and corrected serum calcium level on postoperative day 1 (P=0.031) were associated with postoperative symptomatic hypocalcemia in patients with PHPT. Multivariable logistic regression analysis showed that age (OR=0.967, 95%CI: 0.938 to 0.997, P=0.030), 25-hydroxyvitamin D deficiency or not (OR=2.527, 95%CI: 1.109 to 5.853, P=0.020), preoperative PTH level (OR=1.008, 95%CI: 1.001 to 1.021, P=0.047), and the presence of preoperative clinical symptoms or not(OR=2.598, 95%CI: 1.019 to 6.658, P=0.049) were independently associated with postoperative symptomatic hypocalcemia in patients with PHPT. These four factors were incorporated into the nomogram. The nomogram showed moderate discrimination, with an area under the receiver operating characteristic curve of 0.718 (95%CI: 0.644 to 0.791). The calibration curve showed good agreement between the predicted and observed probabilities of postoperative symptomatic hypocalcemia. Conclusion: The nomogram constructed using age, preoperative symptoms, preoperative PTH level, and 25-hydroxyvitamin D level has potential clinical utility for predicting the risk of postoperative symptomatic hypocalcemia in patients with PHPT and may assist clinicians in individualized risk stratification and diagnostic and treatment decision-making.