Antibiograms generated in the United States support use of potentiated sulfonamides and intravenous amoxicillin–clavulanic acid when antimicrobials are needed in Escherichia coli canine pyometra
Abstract
Objective
Evaluate antimicrobial susceptibility patterns among bacterial pathogens isolated from canine pyometra cases and generate a cumulative antibiogram to inform evidence-based treatment recommendations.
Methods
This observational retrospective study included canine uterine samples submitted for antimicrobial susceptibility testing to Idexx Laboratories and the Cornell University Animal Health Diagnostic Center between 2007 and 2023 in the US. Susceptibility results were interpreted according to current Clinical and Laboratory Standards Institute criteria. Cumulative antibiograms were generated with an interactive app for analysis of diagnostic antimicrobial susceptibility test data (AMR Visualizer; RShiny).
Results
Across the 2 laboratories, 6,462 bacterial isolates were included. Escherichia coli was the predominant pathogen (71.7%). Among E coli isolates, poor susceptibility was observed to penicillins, β-lactam/β-lactamase inhibitor combinations, tetracyclines, and chloramphenicol. Excellent susceptibility was maintained to aminoglycosides, third-generation cephalosporins, potentiated sulfonamides, and imipenem. Most E coli isolates were susceptible-dose-dependent to fluoroquinolones and susceptible with increased exposure to intravenous amoxicillin–clavulanic acid.
Conclusions
Most canine pyometras are caused by E coli. The in vitro antimicrobial susceptibility patterns of these isolates contribute to understanding resistance trends and support antimicrobial stewardship efforts. Susceptibility should be interpreted in the context of regimens used for breakpoint development.
Clinical Relevance
When perioperative antimicrobial therapy of pyometras is appropriate, veterinarians can use antibiograms to assist in selecting an effective and appropriate antimicrobial before susceptibility test results are available. These antibiograms suggest that folate-pathway inhibitors could be effective empiric choices that align with antimicrobial stewardship principles. High-exposure intravenous amoxicillin–clavulanic acid could be an effective choice in regions where the product is available.
Keywords: pyometra; canine; antimicrobial resistance; antibiogram; Escherichia coli
Introduction
Pyometra is a common and serious bacterial infection of the uterus in intact female dogs, affecting an estimated 20% to 25% of them by middle-aged to older age, with a mean age at diagnosis of approximately 7 years.1,2 The disease develops during diestrus, when prolonged progesterone exposure induces endometrial proliferation, glandular secretion, and reduced myometrial contractility. This, paired with local immunosuppression, often predisposes dogs to cystic endometrial hyperplasia.3 These conditions facilitate ascending infection with opportunistic bacteria, most frequently Escherichia coli, which in pyometra are usually distinct from commensal intestinal strains and share virulence traits more commonly with urinary tract pathogens.4–6 Other organisms, including Staphylococcus spp, Streptococcus spp, Pseudomonas spp, Proteus spp, and Klebsiella spp, may also be isolated.2
Clinical manifestations of pyometra vary widely; common clinical signs include lethargy, dehydration, anorexia, hyporexia, polydipsia and polyuria, fever, and occasionally vomiting.7 While diagnosis is often straightforward in open-cervix pyometra with vaginal discharge easily noted, closed-cervix cases can present diagnostic challenges and carry a higher risk of systemic illness.2 Surgical ovariohysterectomy remains the primary definitive treatment for canine pyometra, as it resolves the acute infection and prevents recurrence.2 The procedure is typically classified as clean-contaminated due to transection of the infected uterus, which could increase the risk for surgical site infection (SSI) and other postoperative complications8; however, this classification can vary depending on intraoperative findings and surgeon interpretation.
Antimicrobial therapy is commonly administered perioperatively, yet consensus regarding optimal timing, duration, and agent selection is lacking, and practices vary among clinicans.7 Retrospective analyses show that antimicrobial regimens are often not aligned with bacterial susceptibility profiles.4 A survey8 found that antimicrobial use varied among Arizona-licensed veterinarians; 76% and 74% of clinicians consistently administered antibiotics preoperatively/intraoperatively and postoperatively, respectively, with substantial variability in the choice of drug class and treatment duration. In these cases, only 12% of the respondents regularly submitted a culture.
In 1 Swedish study,9 all observed SSIs occurred in dogs without pre- or intraoperative antibiotics, while none developed in those that received antibiotics, suggesting a protective effect. In contrast, some authors emphasized a more selective approach, recommending antimicrobial therapy be tailored to clinical severity and infection risk, rather than administered uniformly to all cases.2,8,9 Additionally, a recent randomized controlled trial10 demonstrated that placebo was noninferior to postoperative treatment with a folate-pathway inhibitor for SSI prevention when all dogs received a single perioperative antimicrobial dose. Together, these findings suggest that while antimicrobial therapy remains important in the management of canine pyometra, uncertainty persists regarding the most appropriate agents and the timing and duration of treatment. Characterizing susceptibility patterns of pyometra-associated pathogens may contribute to surveillance-level understanding of antimicrobial resistance trends and support the development and refinement of cumulative antibiograms, which in turn may help inform empiric antimicrobial selection and stewardship-guided perioperative treatment protocols.
According to a survey of veterinarians conducted in Arizona, amoxicillin–clavulanate potassium was the most frequently administered antimicrobial agent (86%) for pyometras, followed by enrofloxacin (52%), cefazolin (37%), and ampicillin-sulbactam (22%).8 Some respondents reported the use of multiple antimicrobial agents, reflecting variability in prescribing practices. International guidelines differ in their recommendations for antimicrobial therapy in canine pyometra. Scandinavian guidelines describe combination antimicrobial therapy primarily in the context of medical (nonsurgical) management, whereas in cases undergoing ovariohysterectomy, antimicrobial use is generally reserved for dogs with moderate to severe systemic illness or sepsis.7,11–13 Reported drug recommendations across regions vary, with enrofloxacin and amoxicillin-clavulanate used in some guidelines, while others emphasize potentiated folate-pathway inhibitors or ampicillin as alternative options.7,11–13
Rising concerns about antimicrobial resistance (AMR) and regional variability in pathogen susceptibility patterns underscore the need for localized evidence-based treatment protocols.14 Antibiograms are a valuable tool for guiding empiric therapy, tracking changes in AMR, and informing stewardship efforts in veterinary medicine. Antibiograms include organism names, the number of isolates analyzed, the antimicrobial agents tested, and the percentage of each organism interpreted as susceptible to each listed antimicrobial agent based on the standardized clinical breakpoints.15 Clinical breakpoints are thresholds used to interpret antimicrobial susceptibility test results, translating in vitro MIC values or disk diffusion zones to the in vivo likelihood of therapeutic effectiveness.16 Isolates classified as susceptible are expected to respond favorably to treatment with the corresponding antimicrobials, whereas isolates categorized as resistant are unlikely to achieve successful clinical outcomes. The objective of this study was to identify predominant canine pyometra pathogens and their resistance profiles, inform empiric perioperative antimicrobial use, support antimicrobial stewardship, and contribute to the broader understanding of resistance trends in canine pyometras.
Methods
Study design and microbiological procedures
Two collections of historic pyometra bacterial culture and susceptibility data were used: isolates from the Cornell University Animal Health Diagnostic Center (AHDC) between 2007 and 2023 and at IDEXX Laboratories between 2018 and 2023. Animal Health Diagnostic Center isolates were primarily from New York state and the northeastern US, whereas IDEXX isolates were submitted from across the US and Canada. At both laboratories, isolates were cultured with standard bacteriological methods for culture and organism identification as previously described.14,17,18 All AHDC procedures were performed under standardized operating procedures in a laboratory accredited by the American Association of Veterinary Laboratory Diagnosticians, and testing was conducted by IDEXX Laboratories, operated in accordance with international standard ISO/IEC 17025 for laboratory technical competence.
Antimicrobial susceptibility testing of AHDC isolates was performed with the broth microdilution method, as previously described,17 and interpreted according to Clinical and Laboratory Standards Institute (CLSI) guidelines (Table 1). Between 2007 and 2016, Thermo Scientific Sensititre MIC susceptibility plates (COMPAN1F and COMPAN2F panels) were used; after 2016, the COMPGN1F panel was adopted. All MICs were reinterpreted according to the most current CLSI guidelines.19 Samples processed by IDEXX between 2018 and 2023 were subjected to comparable microbiological culture methods based on client submissions specifying the sampling site and clinical history (uterine and pyometra cases). Antimicrobial susceptibility testing at IDEXX was conducted with the VITEK 2 automated system (bioMérieux) via the VITEK 2 GN-98 card for gram-negative isolates and GP-81 or GP-82 cards for gram-positive isolates20 in accordance with manufacturer instructions, and MICs were interpreted with current CLSI breakpoints.19
