# Environmental surveillance of bacteria in a new intensive care unit using plate sweeps

## TL;DR

This study used plate sweeps to monitor bacteria in a new ICU, finding changes in microbial communities and antibiotic resistance after patient introduction.

## Contribution

The study introduces plate sweeps as a cost-effective method for environmental bacterial surveillance in hospitals.

## Key findings

- Post-patient ICU samples showed different bacterial species compared to pre-patient samples.
- Post-patient samples had increased antimicrobial resistance genes, including resistance to fosfomycin, quinolones, and beta-lactams.
- Cluster analysis identified 17 possible transmission links, highlighting communal bathrooms as key transmission areas.

## Abstract

The hospital environment plays a critical role in the transmission of infectious diseases. Surveillance methods often rely on selective enrichment or deep metagenomic sequencing, which both have significant drawbacks in terms of community resolution and cost. Plate sweeps provide a practical moderate approach to cultivate a wide range of bacteria, capturing more diversity than a single colony pick without high sequencing costs. Here, we use this approach to characterize a newly built hospital intensive care unit (ICU) in Queensland, Australia. Between November 2023 and February 2024, we sampled 78 sites within an 8-bed private hospital ICU pre- and post-patient introduction to the environment. Samples were enriched on non-selective media before DNA was extracted from whole plate sweeps and sequenced using Illumina. We assessed species, antimicrobial resistance (AMR) genes, virulence genes and transmission across all samples and between the pre- and post-patient samples using Kraken2, AbritAMR and Tracs. While the rate of positive microbial growth within the ICU environment did not change significantly pre- and post-patient introduction, the post-patient microbiome consisted of largely different bacterial species; of 22 genera identified, only 3 genera were represented at both timepoints. Post-patient samples were enriched in AMR genes, including resistance to fosfomycin, quinolones and beta-lactams. Common genera identified post-patient were Pseudomonas, Delftia and Stenotrophomonas, often associated with areas of plumbing. Cluster analysis identified 17 possible transmission links from a single timepoint, highlighting several areas in the ICU (e.g. communal bathrooms) as key areas for transmission. We demonstrate the utility of plate sweeps as a means of economical non-selective environmental surveillance and highlight their ability to identify hotspots of transmission within a hospital ward that could be targeted by infection control prior to an outbreak of a more serious pathogen.