Between the beginning of September and end of October 2017, a representative survey based on market share data collected 339 samples of raw chicken (whole or portions) and 342 samples of raw pork mince from retailers in England, Wales, Scotland and Northern Ireland. Samples included both domestically produced and imported meats. These were tested for Escherichia coli (including ESBL-producing E. coli), Klebsiella spp. and enterococci, as well as Campylobacter spp. in the case of chicken samples, and Salmonella spp. in pork. One isolate of each bacterial type from each sample was randomly selected for additional testing to determine the minimum inhibitory concentration (MIC) for a range of antimicrobials.
Salmonella spp. were only detected in 5/342 (1.5%) of pork mince samples, of which four were identified as Salmonella enterica serovar Typhimurium and one as Salmonella enterica serovar Derby. All four S. Typhimurium isolates were resistant to ampicillin and tetracycline (as well as having reduced susceptibility to sulfamethoxazole), with one also showing resistance to chloramphenicol, whilst the S. Derby isolate was susceptible to all antimicrobials except sulfamethoxazole. None of the salmonellae had phenotypes consistent with the production of ESBL or AmpC enzymes.
E. coli were detected more frequently in chicken samples (165/339; 49%) than in pork mince (35/342; 10%). Of isolates examined for MIC determination, a higher percentage of those from chicken had resistance to ciprofloxacin (34/131; 26%), nalidixic acid (33/131; 25%) and gentamicin (9/131; 7%) compared to those from pork (12/94 [13%]; 3/94 [3%] and 0/94 [0%] respectively). In contrast, resistance to chloramphenicol and tetracycline occurred more often in isolates from pork (68/94 [72%] and 22/94 [23%], respectively) than from chicken (48/131 [37%] and 9/131 [7%]). E. coli organisms demonstrating the ESBL (but not AmpC) phenotype were detected in 44/681 (6.5%) of meat samples tested, including 16/342 (4.7%) of pork and 28/339 (8.3%) of chicken samples. Furthermore, the AmpC phenotype alone was detected in 39/339 (11.5%) of chicken samples but not in any pork samples, while the ESBL+AmpC phenotype was detected in chicken samples only (6/339; 1.8%). The difference in prevalence of the ESBL phenotype (including those that were both ESBL and AmpC) between chicken (34/339; 10.0%) and pork (16/342; 4.7%) samples was statistically significant.
In contrast to E. coli and enterococci, Klebsiella species were detected more frequently in pork mince (127/342; 37%) than chicken (22/339; 6.5%). Of 85 Klebsiella isolates examined for MIC determination, rates of resistance were lower than observed in E. coli isolates for all antimicrobials tested, with the exception of ampicillin, to which Klebsiella species are intrinsically resistant.
This survey provides a baseline of the prevalence, types and levels of AMR bacteria found in UK retail chicken and pork mince which the FSA can use to monitor its progress in reducing AMR in these foods and inform UK AMR strategy.
The data generated from this study will provide a baseline dataset for longitudinal comparisons within country as well as comparisons with data from other countries, and will provide a useful benchmark against which to monitor the impact of future interventions.
Antimicrobial resistance was detected in a proportion of all the types of bacteria examined, with resistance to the most clinically important drugs generally appearing to be more prevalent in chicken isolates than pork. However, the risk of acquiring AMR related bacterial infections from these foods is very low provided that they are cooked and handled hygienically. Due to the strategy of sampling in relation to market share, there were insufficient samples from non-UK countries to allow statistical analysis of differences between UK and non-UK produced meat. However, this may be a focus for a future study.
September 2018/ FSA/ United Kingdom.