Antimicrobial-resistant organism carriage and metagenomic characterization of the microbiome and resistome in veterinary workers.

BACKGROUND

Veterinary personnel work in environments with high exposure to antimicrobials and antimicrobial-resistant pathogens. However, the extent to which these occupational exposures influence their carriage of antimicrobial-resistant (AMR) organisms and the composition of their microbiome and resistome remains poorly understood. The objective of this study was to characterize AMR pathogen carriage and microbiome/resistome composition in veterinary personnel and to assess whether workplace setting and patient-related exposures were associated with detectable differences in these outcomes.

METHODS

Personnel from a large academic veterinary hospital and a convenience sample of private-practice small-animal veterinary hospitals provided nasal swabs, stool samples, and exposure survey data; environmental samples were also collected from the academic hospital. Nasal swabs were cultured for methicillin-resistant Staphylococcus aureus, and stool samples were cultured for extended spectrum cephalosporin resistance (ESCR) Escherichia coli. Samples also underwent shotgun metagenomic sequencing or 16S rRNA gene sequencing for taxonomic and antimicrobial resistance gene profiling. Alpha/beta diversity were compared across groups, beta diversity by Permutational Multivariate Analysis of Variance, linear models tested differential abundance, and logistic regression was used to assess covariates associated with MRSA and ESCR carriage. A subset of academic-hospital participants also submitted additional samples after time away (1-2 weeks) from the hospital.

RESULTS

Among academic-hospital personnel (n = 38), MRSA and ESCR-E. coli carriage prevalence were 18.4% and 11.1%, respectively. In private-practice personnel (n = 8), carriage was 12.5% and 25%, respectively. Within the academic hospital, MRSA carriage was associated with administering antibiotics to patients. Microbial composition differed modestly by worksite (R2 = 0.02) when assessed using unweighted UniFrac metrics only, suggesting subtle site-related differences in low-abundance phylogenetically distinct taxa. Global resistome composition did not differ significantly by site or clinical exposure, although a small number of individual AMR ontologies and broad AMR gene families varied across groups, including those associated with commonly-administered oral antibiotics at the hospital. Paired on/off-clinic analyses did not show a clear directional shift after time away from the hospital.

CONCLUSIONS

Veterinary personnel carried MRSA at a notable frequency, while workplace effects on the microbiome and resistome were subtle, with limited evidence for large-scale shifts in community structure by site or patient exposure.