Development of a risk prediction model for catheter-related bloodstream infection in hemodialysis patients based on identified risk factors

Le résumé fourni par la source

OBJECTIVE: To analyze the risk factors for catheter-related bloodstream infection (CRBSI) in patients with deep vein indwelling hemodialysis catheters and construct a corresponding risk prediction model. METHODS: Data were retrospectively collected from 154 patients who underwent temporary hemodialysis via an indwelling central venous catheter (CVC) between January 2023 and December 2024. One hundred patients who received hemodialysis via a Gracz arteriovenous fistula during the same period were also selected to analyze the impact of the dialysis method. Additionally, 58 maintenance hemodialysis patients with CVCs admitted from January to December 2022 were enrolled as a validation group. The incidence of CVC-related bloodstream infections was recorded. The risk factors of deep venous CRBSI were determined by multivariate Logistic regression analysis, and the risk prediction model was constructed and verified. RESULTS: Univariate analysis identified diabetes mellitus, catheter indwelling time, femoral vein catheterization, age, and serum albumin level as significant influencing factors for CVC-associated bloodstream infection (P < 0.05). Multivariate logistic regression analysis confirmed that diabetes mellitus, prolonged indwelling time, and femoral vein placement were independent risk factors for CRBSI in maintenance hemodialysis patients (P < 0.05). The ROC curve for the constructed risk prediction model showed an area under the curve (AUC) of 0.783. The overall prediction accuracy of the model was 90.00%. CONCLUSION: Diabetes mellitus, prolonged catheter indwelling time, and femoral vein catheterization are independent risk factors for CRBSI in maintenance hemodialysis patients. The developed risk prediction model demonstrates good predictive performance.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.