Epidemiology of the virulence and antimicrobial resistance of Escherichia coli from canine and feline clinical isolates.

Abstract

Escherichia coli resistance impacts the veterinary professional because of therapeutic failure and the public health significance. The lack of national monitoring programs in small animal practices limits evidence-based empirical antimicrobial choices. The objectives of the resistance studies were to establish the current status of E. coli susceptibility to routinely-selected antimicrobials and to assess veterinary prescribing behaviors. Clinical canine or feline E. coli isolates (n=3172) were collected from six geographic regions in the US between May 2008 and January 2013. Minimum inhibitory concentrations (MIC) of E. coli isolates were determined for 17 antimicrobials. Population MIC distributions were bimodal with the second mode above the resistant breakpoint for all drugs except for gentamicin, amikacin, and meropenem. The highest percentage of isolates were resistant to doxycyline (100%), and cephalothin (98%). None of isolates was susceptible to all drug tested; 46% were single-drug resistant (SDR), and 52% multi-drug resistant (MDR). Most antimicrobial prescriptions were amoxicillin-clavulanic acid (25%) at 14.9±5.4 mg/kg q12h and enrofloxacin (22%) at 5.6±2.8 mg/kg q24h used for treatment of urinary tract infections (UTI). Our findings reported that most antimicrobial prescriptions were inappropriate for the drug selected, and the dosing regimen. Uropathogenic E. coli (UPEC) isolates acquire virulence genes that encode virulence factors (VF) necessary for colonization of the urinary tract. VFs may offer targets to alter a method of antimicrobial therapy. The objectives of virulence studies were to investigate the associations among severity of UTI, antimicrobial resistance, and virulence within UPEC. Virulence profiles of UPEC differ significantly with severity of infections and resistance patterns. Most severe disease isolates were phylogroups B2 (50%), and ABU were phylogroup D (63%). MDR isolates exhibited shifts in phylogenetic distribution (A, B1>D, B2), compared to SDR isolates (B2, D>B1, A). These results suggest that as the resistance level increases, virulence decreases. The expression levels of virulence genes were computed by principal component and linear discriminant analyses to classify severity of UTI as ABU and non-ABU. The model estimated the non-ABU with a 19% error rate. These findings suggest that the model can be useful in practices by identifying those isolates that do not need antimicrobial therapy.