Predictive value of preoperative bone turnover markers for early postoperative severe hypocalcemia in patients with maintenance hemodialysis and secondary hyperparathyroidism

OBJECTIVE: To evaluate the independent predictive value of preoperative bone turnover markers (BTMs)-particularly P1NP, N-MID osteocalcin, and β-CTX-for early severe hypocalcemia (SH) after parathyroidectomy (PTX) in maintenance hemodialysis (MHD) patients with secondary hyperparathyroidism (SHPT). METHODS: This retrospective cohort study included 196 MHD patients undergoing PTX for refractory SHPT. Patients were categorized into SH and non-SH groups based on postoperative serum calcium < 1.875 mmol/L or severe symptoms. Baseline clinical data and preoperative BTMs (ALP, P1NP, N-MID osteocalcin, and β-CTX) were analyzed using multivariate logistic regression to identify independent risk factors. Predictive performance was evaluated via ROC curves. RESULTS: Early postoperative SH occurred in 101 patients (51.5%). Preoperative ALP, P1NP, and β-CTX were significantly higher in the SH group, whereas N-MID osteocalcin and patient age were significantly lower (P < 0.05). Multivariate analysis confirmed that younger age (OR = 0.950, P = 0.019), lower preoperative serum calcium (P < 0.001), elevated preoperative ALP (OR = 1.006, P = 0.002), and extremely high β-CTX (OR = 3.029, P = 0.001) were robust independent predictors of SH. The statistical weight of P1NP was absorbed by collinearity in the full model. The optimal ROC cutoffs were 238 U/L for ALP (AUC = 0.879) and 5.97 ng/mL for β-CTX (AUC = 0.707). Strikingly, the combined model (age + preoperative calcium + ALP + β-CTX) improved the predictive AUC to 0.918. CONCLUSION: Preoperative BTMs, especially the intense coupling of ALP and β-CTX, along with patient baseline age and calcium, are powerful independent predictors of early SH following PTX. Utilizing a multidimensional prediction model can accurately identify high-risk patients.