Herd-level association between antimicrobial use and antimicrobial resistance in bovine mastitis Staphylococcus aureus isolates on Canadian dairy farms.

Surveillance of antimicrobial use and resistance is needed to manage antimicrobial resistance in bacteria. In this study, data were collected on antimicrobial use and resistance in Staphylococcus aureus (n=562), isolated from intramammary infections and (sub)clinical mastitis cases on 89 dairy farms in 4 regions of Canada [Alberta, Ontario, Québec, and the Maritime Provinces (Prince Edward Island, Nova Scotia, and New Brunswick)]. Dairy producers were asked to deposit empty drug containers into specially provided receptacles, and antimicrobial drug use rate was calculated to quantify antimicrobial use. Minimum inhibitory concentrations were determined using the Sensititer bovine mastitis plate system (TREK Diagnostic Systems Inc., Cleveland, OH), containing antimicrobials commonly used for mastitis treatment and control. Multivariable logistic regression models were built to determine herd-level risk factors of penicillin, ampicillin, pirlimycin, penicillin-novobiocin combination, tetracycline and sulfadimethoxine resistance in Staph. aureus isolates. Intramammary administration of the penicillin-novobiocin combination for dry cow therapy was associated with penicillin and ampicillin resistance [odds ratio (OR): 2.17 and 3.10, respectively]. Systemic administration of penicillin was associated with penicillin resistance (OR: 1.63). Intramammary administration of pirlimycin for lactating cow mastitis treatment was associated with pirlimycin resistance as well (OR: 2.07). Average herd parity was associated with ampicillin and tetracycline resistance (OR: 3.88 and 0.02, respectively). Average herd size was also associated with tetracycline resistance (OR: 1.02). Dairy herds in the Maritime region had higher odds of penicillin and lower odds of ampicillin resistance than dairy herds in Québec (OR: 2.18 and 0.19, respectively). Alberta dairy herds had lower odds of ampicillin and sulfadimethoxine resistance than dairy herds in Québec (OR: 0.04 and 0.08, respectively). Ontario dairy herds had lower odds of tetracycline and sulfadimethoxine resistance than dairy herds in Québec (OR: 0.05 and 0.33, respectively). Herd-level use of certain antimicrobials administered for mastitis treatment and control, such as intramammary penicillin and pirlimycin as well as systemically administered penicillin and florfenicol, was positively associated with antimicrobial resistance in bovine mastitis pathogens in the field conditions. Differences in antimicrobial resistance outcomes across 4 regions of Canada were observed.