Use of standardized nasal and skin decolonization to reduce rates of bacteremia in patients undergoing extracorporeal membrane oxygenation.

Abstract

Objective: We aimed to determine if implementation of universal nasal decolonization with daily chlorhexidine bathing will decrease blood stream infections (BSI) in patients undergoing extracorporeal membrane oxygenation (ECMO).

Design: Retrospective cohort study.

Setting: Tertiary care facility.

Patients: Patients placed on ECMO from January 1, 2017 to December 31, 2023.

Intervention: Daily bathing with 4% chlorhexidine soap and universal mupirocin nasal decolonization were initiated for all ECMO patients May 2021. The primary outcome was rate of ECMO-attributable positive blood cultures. Zero-inflated Poisson regression analysis was performed to estimate rate ratios (RRs) for the association between decolonization with BSI rates.

Results: A total of 776 patients met inclusion criteria during the study period, 425 (55%) preimplementation and 351 (45%) post-implementation. Following implementation of decolonization, the overall incidence rate of BSI increased nonsignificantly from 10.7 to 14.0 infections per 1000 ECMO days (aRR 1.09, 95% CI 0.74-1.59). For gram-positive cocci (GPC) pathogens, a nonsignificant 40% increased rate was observed in the post-implementation period (RR 1.40, 95% CI 0.89-2.21), due mostly to a significant increase in the crude rate of Enterococcus BSI (RR 1.89, 95% CI 1.01-3.55). Excluding Enterococcus resulted in a nonsignificant 28% decreased rate (aRR 0.72, 95% CI 0.39-1.36) due to a nonsignificant 55% decreased rate of MRSA (aRR 0.45, 95% CI 0.18-3.58).

Conclusions: Implementation of a universal decolonization protocol did not significantly reduce rates of certain BSIs, including MRSA and other gram-positive pathogens. Although nonsignificant, reduction in BSI rates in this patient population has important implications on surveillance metrics, such as MRSA, and in the future, hospital-onset bacteremia.