Construction and validation of a predictive model for hypocalcemia after parathyroidectomy in patients with secondary hyperparathyroidism.

Objective: We aimed to construct and validate a predictive model for the risk of hypocalcemia following parathyroidectomy (PTX) for the treatment of secondary(renal) hyperparathyroidism (SHPT). Methods: Information regarding patients with SHPT who underwent PTX between January 2019 and April 2022 was collected retrospectively. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for hypocalcemia following PTX and to construct predictive models. The areas under the receiver operating characteristic curve (AUC), the calibration curve, and the clinical decision curve (decision curve analysis, DCA) were used to assess the discrimination, calibration, and level of clinical benefit obtained using the predictive models. Results: We studied 238 patients who were randomly allocated in a 7:3 ratio to a training group (n=166) and a test group (n=72). Univariate and multivariate logistic regression analyses were performed, in which three variables (the circulating parathyroid hormone (PTH) and Ca concentrations, and alkaline phosphatase (ALP) activity) were interrogated for possible roles as independent risk factors for hypocalcemia in patients with SHPT who undergo PTX, and used to construct predictive models. The AUCs for the constructed models were high for both the training (0.903) and test (0.948) groups. The calibration curve showed good agreement between the incidence of postoperative hypocalcemia estimated using the predictive model and the actual incidence. The DCA curve indicated that the predictive model performed well. Conclusion: A predictive model constructed using a combination of preoperative PTH, Ca, and ALP may represent a useful means of identifying patients with SHPT at high risk of developing hypocalcemia following PTX in clinical practice.