Antimicrobial resistance in ovine bacteria: A sheep in wolf’s clothing?

• DOI: 10.1371/journal.pone.0238708
• Publication Date: 2020-09-03T17:45:56Z
• Authors: Nuno Silva; Clare J. Phythian; Carol Currie; Riccardo Tassi; Keith T. Ballingall; Giada Magro; Tom N. McNeilly; Ruth N. Zadoks
• Keywords: Flock

Background To monitor the prevalence of antimicrobial resistance (AMR), methods for interpretation susceptibility phenotypes bacteria are needed. Reference limits to declare generally based on or dominated by data from human bacterial isolates and may not reflect clinical relevance wild type (WT) populations in livestock other hosts. Methods We compared observed AMR using standard bespoke interpretations breakpoints epidemiological cut-offs (ECOFF) gram positive (Staphylococcus aureus) negative (Escherichia coli) sheep as exemplars. Isolates were obtained a cross-sectional study three lowland flocks Scotland, longitudinal one flock Norway. S. aureus (n = 101) was predominantly isolated milk mammary glands whilst E. coli 103) mostly faecal samples. Disc diffusion testing used determine inhibition zone diameters, which interpreted either ECOFF, distinguish population with acquired mutational compound interest (non-wild type). Standard ECOFF values considered well sheep-specific calculated Normalized Resistance Interpretation (NRI) methodology. Results The measured low, e.g. 4.0% penicillin aureus. Estimation ECOFFs hampered lack relevant reference values. In addition, ECOFFS, data, bisected normal distribution diameters several compounds our analysis isolates. This contravenes recommendations setting NRI methodology lead high apparent prevalence. Using NRI, showed non-wild less than 4% across 13 compounds, ca. 13% amoxicillin ampicillin, while 3% 12 tetracyclines sulfamethoxazole-trimethoprim. Conclusion is highly dependent criteria. industry want establish cut-off monitoring avoid use developed host species. latter could resistance, including critically important classes such 4th generation cephalosporins carbapenems, suggesting an problem that actually exist.