Objective: To develop and validate a nomogram model for predicting central venous catheter-related infections (CRI) in patients with maintenance hemodialysis (MHD).
Methods: MHD patients with central venous catheters (CVCs) visiting the outpatient hemodialysis (HD) center of Xuzhou Medical University Affiliated Hospital from January 2020 to December 2023 were retrospectively selected through a HD monitoring system. Patient data were collected, and the patients were divided into training and validation sets in a 7:3 ratio. The training set was used to establish the model, which was verified using the validation set. Multiple logistic regression analysis was performed to identify risk factors for central venous CRI and develop a nomogram prediction model.
Results: A total of 300 MHD patients were enrolled. Multivariate analysis showed that catheter duration, catheter site, catheter reinsertion, history of catheter infection, diabetes, and albumin <35 g/L were risk factors for central venous CRI. The area under the receiver operating characteristic (ROC) curve (AUC) for the training set was 0.902 (95% confidence interval (CI) = 0.862-0.941), with a sensitivity of 85.7%, specificity of 80%, and a Youden index of 65.7%, and that for the validation set was 0.826 (95% CI = 0.726-0.905), with a sensitivity of 80.5%, specificity of 77.9%, and a Youden index of 58.4%. The model demonstrated good discrimination and calibration (Hosmer-Lemeshow goodness-of-fit test statistics: training set: χ2 = 4.709, p = 0.788; validation set: χ2 = 7.171, p = 0.518).
Conclusion: This study identified six risk factors associated with central venous CRI in MHD patients. This predictive model demonstrates good prognostic performance and can be used by clinicians to screen for high-risk patients with central venous CRI, thereby enabling the early implementation of risk management strategies.
Keywords: Maintenance hemodialysis; central venous catheter-related infections; nomogram; prediction model; risk factor.