Breaking the culture habit: Complementing culture-based veterinary diagnostics with metagenomic data -A case study of feline and canine skin infections.

backgroundSkin infections have been described as the primary cause for veterinary small animal practice visits, frequently requiring topical and systemic antibiotics. These infections often represent secondary complications of underlying pathologies, that can lead to recurrent infections and multiple antibiotic exposures. This creates selection pressure toward antibiotic resistance at the intersection of skin, bloodstream, and shared human-animal environments. This case study integrates Veterinary Diagnostic Laboratory (VDL) aerobic culture results with metagenomic (MGX) data to evaluate the combined utility of these approaches in advancing One Health veterinary diagnostics. Simultaneous reporting of culture-recovered pathogens alongside infection microbiomes and resistomes could strengthen pathogen epidemiology, illuminate polymicrobial etiologies, and inform antimicrobial stewardship.

resultsOne feline and eight canine skin swabs were analyzed with aerobic culture and traditional antimicrobial susceptibility testing (AST) and compared with MGX profiles. VDL aerobic culture and AST identified Staphylococcus aureus, S. pseudintermedius, S. schleiferi, methicillin resistant (MR) S. schleiferi (MRSS), MR S. pseudintermedius (MRSP) and Pseudomonas aeruginosa. MGX data detected the identical bacterial pathogens and identified methicillin resistance genes (mecA, mecI, mecR1) in samples where AST had confirmed MRSP and MRSS. MGX data also detected mec genes in samples without culture confirmed MR phenotypes as well as describing multi-domain microbiota (bacteria, fungi, protists, viruses, phages), antimicrobial resistance genes (ARGs), plasmids, and metabolic features associated with the skin infection samples.

conclusionsMGX data detected the identical VDL recovered pathogens and genes that confer the AMR phenotypes recovered by VDL AST. MGX data also detected additional uncultured pathogens, ARGs, multi-domain microbiota, mobile AMR elements, and metabolic features. Future applications for these methods used simultaneously could support monitoring programs, advance pathogen epidemiology, inform treatment strategy, advance judicious antimicrobial administration, and provide data for machine learning (ML) models to improve precision veterinary diagnosis and treatment.