Colonisation of equine patients and the associated clinical environment with extended-spectrum β-lactamase-producing Enterobacteriaceae at an equine clinic

Introduction Extended-spectrum β-lactamase-producing Enterobacteriaceae (ESBL-E) pose a serious threat to human and animal health due to their clinical importance. It has been recognised that companion animals, such as horses attending equine clinics, can be colonised with ESBL-E. However, the temporal dynamic of ESBL-E carriage and the role of the environment is often unknown. Here, we aimed to assess the prevalence of ESBL-E and their susceptibility to other clinically relevant antibiotics among equine patients upon admission, during the hospital stay and on release at an equine clinic and the associated with their stay clinical environment. Additionally, we investigated whether there were any associations between various relevant patient factors and the ESBL-E colonisation status. Materials and Methods Faecal samples of 40 horses attending and staying in the equine hospital of Veterinary Medicine in Vienna and matching clinical environment swabs were collected upon admission (T1), at 48h (T2) and at 72h/release (T3) and then plated on selective agar (MacConkey agar supplemented with 2 µg/ml cefotaxime). Species identification was performed using MALDI-TOF MS. VITEK2 was used for antimicrobial susceptibility testing against aminopenicillins, 3rd generation cephalosporins, aminoglycosides, quinolones, tetracyclines, sulfonamides and carbapenems. Fisher’s exact test (for categorical variables) and Wilcoxon rank-sum test (for continuous variables) were used to assess the association between the ESBL-E colonisation status and various patient factors such as age, prior antibiotic treatment and surgery. Results We observed a drastic increase in ESBL-E prevalence among the 40 horses: from 2.5% upon admission to 25% at T2 (48h) and 42.5% at T3 72h/ release, while the prevalence in the associated clinical environment was initially considerably higher throughout but nonetheless followed the same trend of a progressive increase: 22,5% at T1 and T2, and 35% at T3. Over 90% of isolates from horses and the environment displayed resistance to ≥3 antimicrobial classes and could be classified as multi-drug resistant. Concerningly, resistance to imipenem of the carbapenem antimicrobial class was found in two faecal samples from animals, and an intermediate level of resistance to imipenem was found in one faecal sample. We found a moderate statistical association between ESBL carriage and general antimicrobial use, as well as a younger age, but not surgery. Conclusions Our data show a progressive increase in ESBL-E prevalence among horses and the associated clinical environment over time, suggesting that hospitalisation may play a role in ESBL-E acquisition, with general antibiotic use being statistically associated with ESBL-E positivity. Worryingly, the majority of ESBL-E isolates have an MDR phenotype and in some cases resistance to imipenem. Ongoing WGS analyses will help to shed light on the clonal structure of the isolates and possible transmission pathways. This study underlines the importance of screening and infection control at equine clinics.