Background: Continuous fluid infusions delivered between therapies by piggy-back systems avoid disconnection and reconnection of central venous catheters (CVCs), thereby reducing opportunities for line contamination. However, the impact of continuous versus intermittent infusions on central line-associated bloodstream infections (CLABSIs) is unknown.

Aim: To investigate the effect of temporary infusion interruption and line disconnection, with or without use of a 70% isopropyl alcohol cap (IPA-C) on CLABSI rates in haematology patients.

Methods: Quasi-experimental study in two haemato-oncology units. At baseline (P1, September 2020 to August 2021), continuous intravenous piggy-back infusions were mandatory. In a first intervention phase (P2, September 2021 to August 2022), infusion disconnections were implemented with use of a 70% isopropyl alcohol cap (IPA-C) for passive decontamination. In a second intervention phase (P3, September 2022 to August 2023), infusion disconnections continued without the use of IPA-C. Rates of CLABSI were compared across the three intervention periods using segmented Poisson regression.

Findings: A total of 11,039 catheter-days across 764 CVCs and 16,226 patient-days were included. Twenty-one CLABSIs were recorded across all intervention periods. Compared with P1, incidence rate ratios (IRRs) for CLABSI did not significantly change in P2 (IRR 0.76 (95% CI 0.27-2.15)) and P3 (IRR 0.79 (95% CI 0.28-2.22)). No CVCs were removed due to occlusion during the study period. Five of 21 CLABSIs were polymicrobial, and coagulase-negative staphylococci were isolated in 19/21 cases (90%).

Conclusion: Interruption of continuous infusions in haemato-oncology patients with a CVC was not associated with a substantial change in CLABSI rates, whether or not an IPA-C was used.