Based on MALDI TOF MS identification of colonies displaying distinct

Based on MALDI-TOF MS identification of colonies displaying distinct colony appearance on Slanetz–Bartley agar, E. faecium (50%) was detected as the most prevalent enterococcal species, followed by E. faecalis (27%), E. hirae (16%) and E. avium (9%). AMP resistance was detected in E. faecium and E. faecalis isolated from 12% and 3% of the dogs, respectively. The faecal concentrations of AMPR enterococci varied from 8×101 to 2.1×103cfu/g for E. faecium (<1–100% of total E. faecium) and from 103 to 6×103cfu/g for E. faecalis (<1–1.7% of total E. faecalis). Each laboratory replicate alone had a sensitivity of 100% for identification of E. faecalis, AMPRE. faecalis and E. avium/raffinosus carriers. The same sensitivity was observed using one of the replicates for identification of E. coli, CTXRE. coli and E. hirae, whereas the other replicate would have missed between one and two carriers for each of these bacterial targets (Se=98%, 88% and 94%, respectively). Use of a single replicate for detection of the remaining bacterial targets (AMPRE. coli and total and AMPRE. faecium) would have missed between one and four carriers (Se=92–96%). Data analysis showed that total enterococci were less frequent amongst dogs with gastrointestinal problems (p=0.05, OR=0.07, 95% CI 0.003-0.7). A borderline significant association was found between carriage of AMPR enterococci and AMPRE. coli (p=0.07, OR=3.2, 95% CI=0.9–1.3).
Discussion
The resistance phenotypes investigated in this GSK503 study are of high clinical relevance in small animal veterinary practice in view of the importance of β-lactams for treatment of common infections in dogs such as urinary tract infections, which are often associated to E. coli and to a lesser extent to enterococci. Various studies have shown that recent treatment with β-lactams is a risk factor for carriage of resistant bacteria in dogs (Gibson et al., 2011; Lawrence et al., 2013). The results of this study indicate that dogs without a recent history of antimicrobial treatment shed resistant bacteria such as AMPRE. coli (40%), AMPRE. faecium (12%), CTXRE. coli (8%) and AMPRE. faecalis (3%) at different frequencies and concentrations. The prevalence of AMPRE. coli carriers (40%) in Danish dogs was two to four times higher than those reported by previous studies in other countries (Murphy et al., 2009; Wedley et al., 2011), whereas the prevalence of CTXRE. coli carriers (8%) fell within the range of expected carriage frequencies (1–18.5%) (Haenni et al., 2014; Murphy et al., 2009). However, comparison between different studies is difficult due to biases associated to geographical, temporal and methodological factors. In addition, dogs included from a University veterinary facility may bias the results toward a higher proportion of carriers. However, our study population represented both primary and secondary/tertiary cases with 57% of dogs being primary cases and 1/3 of the dogs being healthy individuals.
Quantitative microbiological risk assessment is advised to assess the role of animals as a source of antimicrobial resistance in humans (Snary, 2008). The median faecal load of AMPRE. coli (3.2×104cfu/g) in dog carriers was slightly higher to that of CTXRE. coli (8.6×103cfu/g). However, CTXRE. coli are resistant bacteria of higher impact on public health due to the importance of third generation cephalosporins in the therapy of severe E. coli infections in humans. Faecal shedding of these bacteria was extremely variable amongst dogs (8×101 to 2×105cfu/g) and even higher variability (1×102 to 6×1010cfu/g) was reported by a recent study in The Netherlands (Baede et al., 2015). The average faecal concentration of CTXRE. coli in Danish dogs was close to those found in weaners and finishers (105 and 103, respectively) but lower than the average in piglets (107) in Danish pig farms positive for ESBL-producing E. coli (Hansen et al., 2013).