Background: The aim of this study was to evaluate the risk of catheter-associated bloodstream infection (CABSI) among different catheter types using a large prospective database in the neonatal intensive care unit (NICU) of a tertiary care centre in Switzerland.

Methods: We included all neonates admitted to the NICU with at least one central intravascular catheter inserted between January 2017 and December 2020. We used marginal Cox model to determine the risk of CABSI among different catheter types.

Results: A total of 574 neonates and 1103 intravascular catheters were included in the study: 581 venous umbilical catheters, 198 arterial umbilical catheters and 324 peripherally inserted central catheters (PICCs). We identified 17, four and four CABSIs in neonates with venous umbilical catheters, arterial umbilical catheters and PICCs, respectively. The risk of CABSI increased after two days of umbilical catheter maintenance. Using univariable Cox models, and adjusting for sex and gestational age, we observed a similar CABSI risk between venous and arterial umbilical catheters (HR 0.57; 95% CI 0.16e2.08). Birth weight was associated with CABSI, with higher weight being protective (HR 0.37, 95% CI 0.16e0.81).

Conclusions: Strategies aimed at reducing umbilical catheter dwell time, particularly in low and very low birth weight neonates, may be effective in decreasing the incidence of CABSI in this population.