Construction of a -based risk prediction model for vascular calcification detection in patients undergoing hemodialysis.

Vascular calcification (VC) is highly prevalent in patients undergoing hemodialysis, and is a significant contributor to the mortality rate. Therefore, biomarkers that can accurately predict the onset of VC are urgently required. Our study aimed to investigate serum levels in relation to VC and to develop a predictive model for VC in patients undergoing hemodialysis at the Beijing Friendship Hospital hemodialysis center between 1 January 2019 and 31 December 2020. The patients were categorized into two groups: VC and non-VC. Logistic regression (LR) models were used to examine the risk factors associated with VC. Additionally, we developed an -based nomogram based on the results of the multivariate LR analysis. A total of 138 patients under hemodialysis were investigated (age: 58.41 ± 13.22 years; 54 males). VC occurred in 79 (57.2%) patients. Multivariate LR analysis indicated that serum , age, and WBC count were independent risk factors for VC. A -based nomogram was developed by incorporating the following five predictors: age, dialysis vintage, predialysis nitrogen, WBC count, and . The receiver operating characteristic (ROC) curve had an area under the curve of 0.921, diagnostic threshold of 0.396, sensitivity of 0.722, and specificity of 0.932, indicating that this model had good discrimination. This study concluded that serum levels, age, and white blood cell (WBC) count are independent risk factors for VC. A nomogram constructed by integrating these risk factors can be used to predict the risk of VC in patients undergoing hemodialysis.