Linchuang shenzangbing zazhi (Jul 2025)
Construction and validation of a nomogram in predicting the risk of hypocalcemia in maintenance hemodialysis patients
Abstract
ObjectiveTo investigate the occurrence of hypocalcemia in maintenance hemodialysis patients and its influencing factors， and to construct a nomogram to predict the risk of hypocalcemia in this population.MethodsA total of 426 maintenance hemodialysis patients admitted to the Blood Purification Center of the Fourth Affiliated Hospital of Nanjing Medical University from January 2023 to March 2024 were selected as the research subjects. They were divided into the training and validation sets at a ratio of 7∶3. The current situation and influencing factors for hypocalcemia in the patients were statistically analyzed. A nomogram was created by logistic regression to predict the risk of hypocalcemia in maintenance hemodialysis patients.ResultsThe general clinical data and information on influencing factors among 298 cases of maintenance hemodialysis patients in the training set were included. They had an average age of 61.25±7.41 （44-82） years. Among them， 133 （44.63%） cases were female and 165 （55.37%） cases were male. Of the 298 elderly maintenance hemodialysis patients included， 101 （33.89%） developed hypocalcemia. Univariate analysis showed that gender， body mass index （BMI）， dialysis age， combined thyroid disease， albumin， blood creatinine， blood phosphorus value， parathyroid hormone （PTH）， nutritional status， and taking compound α-ketoacid tablets were independent risk factors for hypocalcemia in maintenance hemodialysis patients （P<0.05）. Logistic regression analysis showed that combined thyroid disease， blood creatinine>800 μmol/L， blood phosphorus>1.78 mmol/L， and PTH >150 ng/L were risk factors for hypocalcemia in maintenance hemodialysis patients （P<0.05）， and male and taking compound α-ketoacid tablets were protective factors for hypocalcemia in maintenance hemodialysis patients （P<0.05）. Among 128 patients in the validation set， the area under the curve （AUC） of the nomogram in predicting the risk of hypocalcemia in maintenance hemodialysis patients was 0.812 （95%CI：0.804-0.852）， with a sensitivity of 88.72%， and a specificity of 71.90%.ConclusionThe onset of hypocalcemia in maintenance hemodialysis patients can be influenced by the gender， comorbidity with thyroid disease， blood creatinine， blood phosphorus value， PTH， and taking compound α-ketoacid tablets. In the future， trajectory studies should be conducted to optimize the nomogram， thus providing a reliable tool for clinical nurses to dynamically predict the occurrence of hypocalcemia in maintenance hemodialysis patients， with the aim of reducing hypocalcemia and improving the quality of life of patients.
