BMC Infect Dis. 2026 Feb 14. doi: 10.1186/s12879-026-12837-2. Online ahead of print.
ABSTRACT
BACKGROUND: Central venous catheterization is a routine procedure in clinical practice, but it can result in severe and costly central line-associated bloodstream infections (CLABSIs), the risk of which may vary by insertion sites. This study sought to evaluate and compare the incidence of CLABSI in critically ill patients between internal jugular and subclavian sites, and to explore the correlation between insertion site and risk of CLABSI.
METHODS: This research encompassed critically ill patients with Central venous catheters (CVCs) placed in the internal jugular or subclavian site from ICUs of 22 public tertiary hospitals throughout China, from September 4, 2023, to February 29, 2024. The incidence of CLABSI was calculated and compared between the two groups using Poisson regression. The association between insertion site and CLABSI risk was assessed using univariable and multivariable Cox proportional hazards regression models, with additional adjustments for center-effects using a frailty model.
RESULTS: Among 2,379 patients analyzed, the internal jugular site was the predominant insertion site (68.3%, n = 1,626) compared to the subclavian site (31.7%, n = 753). Over 18,359 total CVC days, 19 CLABSIs were documented. The incidence rate was 0.66 and 1.22 per 1000 CVC days for the subclavian and internal jugular groups, respectively (Incidence rate ratio [IRR] = 1.86, 95% CI [0.67-6.62]; P = 0.27). In the primary analysis, no statistically significant association was found between insertion site and CLABSI risk, either in the unadjusted model (Hazard ratio [HR] = 1.83, 95% CI [0.61-5.53]) or after multivariable adjustment and accounting for between-center heterogeneity (adjusted HR = 2.07, 95% CI [0.65-6.62]).
CONCLUSIONS: In this large multicenter cohort from China, the overall CLABSI incidence was low. We observed no statistically significant difference in CLABSI risk between internal jugular and subclavian CVC insertion sites after adjustment for confounders and center effects. However, the findings were inconclusive due to the limited number of events and the potential for residual confounding inherent in the observational design. These results highlight the need for larger, sufficiently powered studies to provide definitive evidence on the comparative infection risk of CVC insertion sites.
CLINICAL TRIAL NUMBER: Not applicable.
PMID:41691144 | DOI:10.1186/s12879-026-12837-2
