Background. Understanding drivers of multidrug resistance (MDR) and methicillin resistance, which have increased among canine staphylococcal isolates, is essential for guiding antimicrobial use practices. Therefore, the objective of this study was to identify predictors of MDR and methicillin resistance among Staphylococcus spp. commonly isolated from canine clinical specimens. Methods. This retrospective study used records of canine specimens submitted to the University of Tennessee College of Veterinary Medicine Clinical Bacteriology Laboratory for bacterial culture and antimicrobial susceptibility testing between 2006 and 2017. Records from 7,805 specimens positive for the following Staphylococcus species were included for analysis: Staphylococcus pseudintermedius, Staphylococcus au-reus, Staphylococcus coagulans (formerly Staphylococcus schleiferi subspecies coagulans), and Staphylococcus schleiferi (formerly S. schleiferi subsp. schleiferi). Generalized linear regression models were fit using generalized estimating equations (GEE) to identify predictors of MDR (defined as resistance to three or more antimicrobial classes) and methicillin resistance among these isolates. Results. Multidrug resistance (42.1%) and methicillin resistance (31.8%) were relatively common. Isolates from skeletal (joint and bone) specimens had the highest levels of MDR (51.3%) and methicillin resistance (43.6%), followed by cutaneous specimens (45.8% multidrug-resistant, 37.1% methicillin resistant). Staphylococcus species, spec-imen site, and clinical setting were significant (p < 0.01) predictors of both outcomes. Compared to S. pseudintermedius, S. schleiferi had higher odds of methicillin resistance, while S. coagulans and S. schleiferi had lower odds of MDR. The odds of both MDR and methicillin resistance for isolates from hospital patient specimens were significantly higher than those from referral patients for urine/bladder and otic specimens. Odds of MDR among isolates from skeletal specimens of hospital patients were also higher than those of referral patients. Conclusions. Staphylococcus isolates in this study had substantial levels of MDR and methicillin resistance. Differences in the odds of these outcomes between referral??and hospital patient isolates did not persist for all specimen sites, which may reflect differences in diagnostic testing and antimicrobial use practices with respect to body site or system. Judicious antimicrobial use, informed by culture and susceptibility testing, is important to limit treatment failures and curb selection pressure.
