Skip to main navigation Skip to search Skip to main content

Water-free care in Dutch intensive care unit patient rooms: impact on Gram-negative bacteria detections in routine patient care

  • S. A.M. van Kessel*
  • , A. F. Schoffelen
  • , K. H.S. van Son
  • , D. W. Notermans
  • , J. A. Severin
  • , F. Bakhshi-Raiez
  • , F. Velthuis
  • , M. Schipper
  • , D. Dongelmans
  • , A. Verbon
  • , C. C.H. Wielders
  • , the ISIS-AR study group
  • *Corresponding author for this work
  • National Institute of Public Health and the Environment
  • University of Amsterdam
  • National Intensive Care Evaluation (NICE) Foundation
  • Utrecht University
  • IJsselland Ziekenhuis
  • Reinier de Graaf Groep
  • Deventer Ziekenhuis
  • Albert Schweitzer Ziekenhuis
  • Admiraal de Ruyter Hospital
  • Haga Hospital

Research output: Contribution to journalArticleAcademicpeer-review

23 Downloads (Pure)

Abstract

Background: Patients in intensive care units (ICUs) are at an increased risk of healthcare-associated infections with Gram-negative bacteria (GNB), for which sinks in patient rooms are known reservoirs. We investigated the association between water-free care practices and the incidence of GNB detections in Dutch ICUs in non-outbreak settings. Methods: We performed a retrospective ecological study (2018–2022) using data from the Infectious diseases Surveillance Information System-Antibiotic Resistance (ISIS-AR), the National Intensive Care Evaluation registry and a questionnaire on water-free care. Detections (colonisation and infections) of seven bacteria groups (Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, Acinetobacter spp., all Enterobacterales, extended-spectrum beta-lactamase-producing Enterobacterales [ESBL-E] and carbapenemase-producing Enterobacterales) were analysed at ICU-year level. Incidence rate ratios (IRRs) were calculated for water-free vs. non-water-free ICUs, adjusted for ICU and patient characteristics. Findings: Data from 37 ICUs were analysed, 22 ICU-years in the water-free group and 131 in the non-water-free group. Water-free ICUs were larger, with more surgery admissions and mechanically ventilated patients. For all bacteria, adjusted IRRs were close to 1 with broad 95% confidence intervals (CIs), ranging from 0.82 (95% CI: 0.44–1.52) for ESBL-E to 1.39 (95% CI: 0.69–2.84) for Acinetobacter spp. Sensitivity analyses showed similar results. Conclusion: Although positive effects of water-free care on GNB detection rates have been described in single ICUs, these findings were not reflected in this Dutch multi-centre study. Possible explanations are low infection prevalence, high prevention standards, widespread usage of selective decontamination and insufficient power to detect small differences. Evidence for benefits of water-free care in non-outbreak settings remains limited, highlighting the importance of future research in different ICU settings.

Original languageEnglish
Pages (from-to)9-16
Number of pages8
JournalJournal of Hospital Infection
Volume170
DOIs
Publication statusPublished - Apr 2026

Bibliographical note

Publisher Copyright: © 2026 The Authors

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 3 - Good Health and Well-being
    SDG 3 Good Health and Well-being

Fingerprint

Dive into the research topics of 'Water-free care in Dutch intensive care unit patient rooms: impact on Gram-negative bacteria detections in routine patient care'. Together they form a unique fingerprint.

Cite this