The effects of a prospective sink environmental hygiene intervention on Pseudomonas aeruginosa and Stenotrophomonas maltophilia burden in hospital sinks

Abstract

Background: Opportunistic premise plumbing pathogens (OPPPs) can establish reservoirs in hospital plumbing and cause healthcare associated infections (HAIs). There is currently no widely accepted protocol for sink drain cleaning to reduce OPPP burden.

Methods: We implemented a sink cleaning intervention in 12 intensive care unit (ICU) rooms. At low frequency (1×/week) and high frequency (5×/week) intervals, we wiped sink surfaces with 10% bleach wipes and pumped a foamed preacid disinfectant into sink drains. We also maintained untreated rooms (0×/week). We used E-swabs to sample sink drains and surrounding surfaces during one baseline, two intervention, and two post-intervention periods over 23 months. Samples were selectively cultured for bacterial growth and antimicrobial resistant organism (ARO) isolation. We conducted whole-genome sequencing (WGS) on Pseudomonas spp. and Stenotrophomonas spp. isolates to track impacts on reservoirs over time. We also collected and analysed clinical isolates from patients occupying the study rooms and information about HAIs that occurred.

Findings: The intervention reduced the proportion of sink drains yielding Gram-negative bacteria by up to 85% (95% CI: 56-114%) in high frequency rooms versus the baseline period, but this was not significant in low frequency rooms. It also reduced the proportion of sink drains yielding Pseudomonas spp. and Stenotrophomonas spp. by up to 100% (95% CI: 79-121%) and 95% (95% CI: 65-125%) versus the baseline period in high frequency rooms and up to 71% (95% CI: 50-92%, p < 0.001) and 66% (95% CI: 27-105%, p < 0.05) in low frequency rooms, respectively. WGS showed strains of Pseudomonas aeruginosa and Stenotrophomonas maltophilia that colonised sink drains for over 3 years across two studies. Following the intervention periods, P. aeruginosa reservoirs were replaced with new strains, while S. maltophilia reservoirs returned with the same strains.

Interpretation: This environmental hygiene intervention may be effective in reducing the burden of OPPPs in hospital sinks.

Funding: Agency for Healthcare Research and Quality (R01HS027621), National Institute of Allergy and Infectious Diseases (U01AI123394, 1K23AI137321), Barnes-Jewish Hospital Foundation (5102), Washington University Institute of Clinical and Translational Sciences (4462) from the National Center for Advancing Translational Sciences (UL1TR002345).

Keywords: Antimicrobial resistance; Healthcare-associated infections; Infection prevention; Sink drains; Whole-genome sequencing.

Fig. 1

Study design and sample processing overview. a) E-swabs were collected from the raised counter, surrounding counter, faucet, sink bowl, and sink drain at every sampling timepoint. Water was not collected at every sampling timepoint. b) This study included 18 rooms, with eight from the stem cell transplant intensive care unit (SCTO-ICU) and ten from the surgical intensive care unit (SICU). c) The intervention was applied at low (1×/week) and high (5×/week) frequencies. d) This study covered 5 periods of intervention or non-intervention and included 31 sample collection timepoints.

Fig. 2

Interventions reduce total bacterial growth and specific ARO growth. Average weekly proportion of sink drains in each intervention group that yielded viable cultivable growth on a) BAP (n = 528 plates from sink drains cultured) and b) MAC plates (n = 550 plates from sink drains cultured). Significant changes in weekly proportion of sink drains yielding growth by period versus the baseline period shown. c) Weekly counts per sink drain of unique AROs (n = 984 AROs collected from sink drains). Average weekly counts across all rooms in each week were calculated, and significant changes in weekly counts by period versus the baseline period are shown. Proportion of rooms yielding d) Pseudomonas spp. and e) Stenotrophomonas spp. from patient room sink drain. Significant changes in weekly proportion of sink drains yielding growth by period versus the baseline period shown (t-test with BH correction).∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001.

Fig. 3

Strain tracking of P. aeruginosa and S. maltophilia reveals long term sink colonisation. Tracking of core SNP-based strain groups from this study and a previous study in the same ward (HM Study¹²) for a) P. aeruginosa (n = 264) and b) S. maltophilia (n = 139). Grey boxes mark rooms where samples were not collected in the previous study. Interventions are labelled in the second and fourth periods of this study. SICU = surgical intensive care unit, HC = housekeeping closet, SU = soiled utility room, SCTO-ICU = stem cell transplant intensive care unit.

Fig. 4

Variation in strain diversity of P. aeruginosa and S. maltophilia across study periods. Number of unique strain groups found each week in each intervention group for a) P. aeruginosa (n = 49 total strain groups) and b) S. maltophilia (n = 12 total strain groups). Isolates not included in strain groups counted as their own unique strains. Lines represent rolling mean across 4 collection timepoints. Significant decreases in mean number of unique strain groups versus the baseline period shown (permutation test with BH correction). Count of sink surface isolates that were c) P. aeruginosa or d) S. maltophilia strains collected from the same room an earlier period (recurring, blue) or not seen yet (new, red) in that room. Significant changes in proportions of recurring and new strains versus untreated rooms during the same period shown (Fisher's exact test with BH correction). ∗: p < 0.05, ∗∗: p < 0.01), ∗∗∗: p < 0.001.