Association of patient clinical and gut microbiota features with vancomycin-resistant enterococci environmental contamination in nursing homes: a retrospective observational study

Abstract

Background: Preventing transmission is crucial for reducing infections with multidrug-resistant organisms (MDROs) in nursing homes. To identify resident characteristics associated with MDRO spread, we investigated associations between patient characteristics and contamination of their proximate room surfaces with vancomycin-resistant enterococci (VRE).

Methods: In this retrospective observational study, we used demographic and clinical data (including data on comorbidities, physical independence, catheter use within the past 30 days, and antibiotic exposure within the past 30 days) and surveillance cultures of patient body sites and room surfaces at enrolment and during weekly follow-up visits within the first month, and monthly thereafter (up to 6 months), in six US nursing homes collected in a previous clinical trial (September, 2016, to August, 2018). We did 16S rRNA gene sequencing on perirectal surveillance swabs to investigate the association between the gut microbiota and the culture status of participants and their rooms.

Findings: We included 245 participants (mean age 72·5 years [SD 13·6]; 111 [45%] were men, 134 [55%] were women, 132 [54%] were non-Hispanic white, and 112 [46%] were African American). We collected 2802 participant samples and 5592 environmental samples. At baseline, VRE colonisation was present in 49 (20%) participants, with environmental surfaces being contaminated in 36 (73%) of these patients. Hand contamination among VRE-colonised participants was more common in those with environmental contamination compared with those without (50 [51%] of 99 vs seven [13%] of 55; p<0·0001). We found a correlation between hand contamination and both groin and perirectal colonisation and contamination of various high-touch room surfaces (Cohen's κ 0·43). We found participant microbiota composition to be associated with antibiotic receipt within the past 30 days (high-risk antibiotics p=0·011 and low-risk antibiotics p=0·0004) and participant VRE colonisation status, but not environmental contamination among VRE-colonised participants (participant only vs uncolonised p=0·071, both participant and environment vs uncolonised p=0·025, and participant only vs participant and environment p=0·29). Multivariable analysis to identify independent factors associated with VRE-colonised participants contaminating their environment identified antibiotic exposure (adjusted odds ratio 2·75 [95% CI 1·22-6·16]) and male sex (2·75 [1·24-6·08]) as being associated with increased risk of environmental contamination, and physical dependence as being associated with a reduced risk of environmental contamination (0·91 [0·83-0·99]).

Interpretation: Our data support antibiotic use and interaction with proximal surfaces by physically independent nursing home residents as under-appreciated drivers of environmental contamination among VRE-colonised residents. Integrating resident hand-hygiene education and antimicrobial stewardship will strengthen efforts to reduce MDROs in nursing homes.

Funding: US Centers for Disease Control and Prevention, National Institute of Health, Canadian Institutes of Health Research, and University of Michigan.

Trial registration: ClinicalTrials.gov NCT02909946.