Antimicrobial susceptibility of Salmonella and Escherichia coli from equids sampled in the NAHMS 2015-16 equine study and association of management factors with resistance.

Abstract: Several studies have investigated antimicrobial resistance (AMR) in Salmonella spp. and Escherichia coli isolated from hospitalized horses, but studies conducted on community-based populations of equids are limited. The factors associated with AMR in these bacteria in the general horse population are not well understood. The primary objective of our study was to estimate the prevalence of Salmonella and describe antimicrobial susceptibility of Salmonella and E. coli from equids across the United States. The second objective was to identify associations between health management and biosecurity practices and AMR. Fecal samples submitted from 1357 equids on 199 operations were tested for Salmonella, identifying 27 positive samples with 29 isolates belonging to 18 serotypes. Fecal sample and operation-level prevalence of Salmonella was 2.0% (27/1357) and 7.0% (14/199), respectively. Most (25/29) isolates were pan-susceptible while four isolates exhibited resistance, three of which were multidrug resistant. Of the 721 samples cultured for E. coli, 85% (613/721) were positive. Eighty-six percent of the E. coli isolates recovered were pan-susceptible (529/612). Ten isolates were intermediate to one antimicrobial drug and susceptible to all others. Seventy-three E. coli isolates (11.9%, SE=1.3) were resistant to one or more antimicrobials, corresponding to a 33.0% (64/194) operation-level prevalence. Resistance to sulfonamide drugs was most common with 63 isolates (10.3%) resistant to sulfisoxazole, 57 of which (9.3%) were resistant to trimethoprim-sulfamethoxazole. MDR in E. coli was rare (1.8%, SE=0.5). Univariate and multivariable regression were used to evaluate associations between health management and biosecurity questionnaire items and AMR in E. coli. The outcome modeled was resistance to any of the 14 tested antimicrobials. Depending on the operation type, operations with greater than 20 resident equids were significantly associated with resistance. In addition, performance operations were significantly associated with resistance when compared to farm/ranch operations. Operations with feed containers that prevent fecal contamination and those that had treated any equids for illness or injury were associated with a lower AMR. The study results suggest that equids in the general population appear to pose low risk of shedding antimicrobial resistant strains of Salmonella and E. coli, and therefore low transmission potential to other equids, animals, humans, or the environment. However, it is prudent to practice good hand hygiene to prevent spread of Salmonella as well as AMR, and to protect both animal and human health. Despite study limitations, potential management factors that may influence prevalence and prevent spread of AMR shed by equids were identified.