Analysis of risk factors and development of a risk prediction model for postoperative hypocalcemia in differentiated thyroid cancer.

OBJECTIVE

To systematically analyze the risk factors of hypocalcemia following surgery for differentiated thyroid cancer (DTC), and to develop and validate a high-precision Nomogram-based prediction model, so as to provide a basis for accurate clinical prevention and management.

METHODS

This retrospective analysis included 597 DTC patients admitted between March 2019 and January 2025 (training set: n=353; validation set: n=133: external validation set: n=111). Patient features (age, sex, body mass index, diabetes history, etc.), surgical factors (thyroidectomy extent, lymph node dissection, etc.), pathological characteristics (capsular invasion, Tumor, Node, Metastasis [TNM] staging, etc.), and postoperative biochemical indicators (intact parathyroid hormone [iPTH] and blood calcium) were collected. Independent risk factors were screened by univariate and multivariate logistic regression. A Nomogram was constructed based on these factors, and its predictive performance was evaluated using the area under the receiver operating characteristic (ROC) curves (AUC), calibration plots, and decision curve analysis (DCA), with comparisons made to postoperative iPTH-based predictions.

RESULTS

Multivariate logistic regression identified the following as independent predictors of hypocalcemia: diabetes history (OR=3.132, P=0.006), bilateral thyroidectomy (OR=2.142, P=0.023), lateral compartment lymph node dissection (OR=2.011, P=0.037), capsular invasion (OR=3.196, P<0.001), surgical time (OR=10.843, P<0.001), and intraoperative bleeding (OR=7.493, P<0.001). The Nomogram model exhibited excellent discriminatory ability across the training (AUC=0.888), validation (AUC=0.866), and external validation sets (AUC=0.913). Calibration curves and DCA demonstrated that the Nomogram had high prediction consistency and clinical net benefits (peak net benefits: 56.94%, 62.40%, and 63.90%, respectively). Moreover, the model significantly outperformed iPTH-based predictions in both the training (P=0.019) and external validation cohorts (P=0.042).

CONCLUSIONS

Diabetes history, bilateral thyroidectomy, lateral lymph node dissection, capsular invasion, prolonged surgical time (≥82.5 min), and increased intraoperative bleeding (≥25.5 mL) are significant risk factors for postoperative hypocalcemia in DTC patients. The Nomogram model, integrating these factors, outperforms iPTH-based predictions and offers a precise tool for preoperative risk assessment and postoperative management to reduce hypocalcemia and improve patient outcomes.