Discussion

When deciding whether to implement contact precautions for a given organism, institutions must consider the endemicity of the organism in its catchment area, the ability of that organism to transmit between patients, and the effectiveness of contact precautions in preventing such transmission. To inform this process, infection preventionists and healthcare epidemiologists refer to consensus and evidence-based guidelines. Unfortunately, the limited guidelines specific to ESBL organisms are problematic.

In 2006, the CDC recommended contact precautions for all patients colonized or infected with multidrug-resistant organisms, irrespective of genus or species. Notably, none of the literature cited for this recommendation relates to ESBL organisms. ^{Reference Siegel, Rhinehart, Jackson and Chiarello1} The European Society of Clinical Microbiology and Infectious Diseases also recommends contact precautions for ESBL Enterobacterales. However, referenced studies evaluated contact precautions as part of larger infection prevention bundles or were undertaken in an outbreak setting; thus, conclusions about the effectiveness of contact precautions alone based on these data must be made with caution. ^{Reference Tacconelli, Cataldo and Dancer2} Although the Society of Healthcare Epidemiology of America more recently released guidelines describing when to discontinue contact precautions, they did not define when to initiate such precautions. ^{Reference Banach, Bearman and Barnden9}

In fact, literature exists to dispute the hypothesis that ESBL organisms spread rapidly between inpatients. Several studies have shown low rates of cross transmission of ESBL organisms from index to contact patient. ^{Reference Vehreschild, Hamprecht and Peterson4–Reference Tschudin-Sutter, Frei, Dangel, Stranden and Widmer6} In some cases, suspected transmissions were disproven using whole-genome sequencing. ^{Reference Vehreschild, Hamprecht and Peterson4,Reference Tschudin-Sutter, Frei, Dangel, Stranden and Widmer6} One potential explanation is that ESBL organisms, particularly ESBL E. coli, are not a frequent cause of environmental contamination. ^{Reference Valenza, Schulze and Friedrich5} Equally as important, contact precautions have not been shown to independently reduce the nosocomial spread of these organisms in the non-outbreak setting. ^{Reference Maechler, Schwab and Hansen3,Reference Vehreschild, Hamprecht and Peterson4}

Ultimately, we felt that the evidence was insufficient to support the continued use of contact precautions for ESBL organisms given the well-described negative impacts of contact precautions for both the patient and the institution. ^{Reference Sprague, Reynolds and Brindley10} Our decision was also influenced by our need to reserve our limited number of single-occupant rooms for emerging, highly transmissible organisms, most notably CRE and Candida auris.

Our study had several limitations. First, it was retrospective, raising the possibility of confounders. However, we found no difference in our ED ESBL prevalence rates between the study periods, suggesting that our community rates remained stable throughout. Although we detected a small but statistically significant improvement in hand hygiene compliance between study periods, this finding only underscores the importance of focusing on the basic tenets of infection prevention.

We calculated patient prevalence rates rather than infection/colonization incidence rates, which are a more direct measure of healthcare acquisition according to the NHSN. ⁸ A significant proportion of our patients seek care at multiple health systems; thus, we could not reliably capture prior ESBL infection or colonization to calculate incidence rates. Instead, we compared ESBL prevalence rates between cultures collected after hospital day 3, as an indicator of hospital acquisition, and we did not detect a difference between study periods. Lastly, our findings were consistent for both ESBL E. coli and Klebsiella spp, but our absolute number of Klebsiella isolates was relatively small. This finding underscores the need for additional research evaluating the impact of contact precautions on the transmissibility of non–E. coli ESBL spp.

In conclusion, discontinuing contact precautions was not associated with an increase in the prevalence of ESBL E. coli or Klebsiella spp in our hospital. As a result of our change in practice with respect to contact precautions for ESBL, MRSA, and VRE, we also generated a cost avoidance on materials. The coronavirus disease 2019 pandemic has only highlighted the significance of our study, underscoring the need to reserve single-occupant rooms and personal protective equipment for emerging highly infectious pathogens. Our findings add to the growing body of literature questioning the need for such precautions for ESBL organisms, particularly in institutions similar to ours, with limited availability of single-occupant rooms and prevalence of ESBL in our catchment area.