What is the problem?

Antibacterial drug resistance (ABR) is estimated to cause 8000-10,000 deaths per year in the UK. The economic cost of longer stays in hospital, more time off work, and a greater requirement of social care resulting from slower cure is extremely large. However, ABR also has implications for domestic animals. For food producing animals, the inability to treat infections because of ABR affects food security. For food producing and companion animals there is a perceived threat that ABR can transfer to humans and ultimately make the threat of ABR in humans much worse. We want to understand how ABR is selected in food producing and companion animals – what are the risk factors – and whether there is any evidence of transmission between animals and humans in a defined geographical region (50 x 50 km).

What research is being carried out?

This project has involved surveys of ABR on 54 dairy farmers in the South West of England. Samples were collected from young animals as they get older, and from the adult population monthly for 2 years in 2017-18. Management practice and antimicrobial usage data have also been collected. All samples were processed to determine and identify the levels and types of ABR found in Escherichia coli, a key human and animal pathogen, which is carried in the intestines of all mammals.

We have subsequently related the presence of ABR with the management practice risks and the levels of antimicrobial usage on farms. Following usage restrictions for critically important antibiotics (3^rd generation cephalosporins and fluoroquinolones) on the farms in mid 2018, we then returned to 20 of the dairy farms in 2021 and collected samples again. The aim is to determine if levels of antibiotic resistance have fallen as usage has been reduced.

We also use whole genome sequencing of resistant E.coli to investigate farm to farm transmission, and other One Health sharing events.

We recruited 250 puppies at 12-16 weeks of age and 600 adult dogs from within the study region. We have collected faecal samples from these dogs and used owner surveys to identify behavioural and “management” risk factors for carriage of ABR E. coli of different types in these dogs.

We conducted a survey of ABR E. coli within Bristol Zoological Gardens, and on smallholder farms in the South West of England.

We are monitoring the levels of ABR in E. coli from human urinary tract and bloodstream infections across the South West over >5 years and relating trends in ABR with the rates of prescription of antibiotics in primary care. We are characterizing the types of ABR E. coli found in these clinical samples and using whole genome sequencing to relate these bacteria with the bacteria identified in dogs and on farms, with a view to testing whether there is evidence of transmission from one group to the other.

Outcome and next steps

We have identified that the way samples are collected on farms affects the levels of ABR E. coli observed. The place and age of animals sampled, the way samples are collected, and the temperature at the time of sample collection all make a difference (Schubert et al, 2021 Applied and Environmental Microbiology, 87:e01468-20; Turner et al, 2022 Journal of Applied Microbiology, 132:2633-2641). This will inform appropriate farm ABR survey practices for research and “benchmarking”. We have identified that antimicrobial usage on dairy farms has only a weak influence on the prevalence of ABR E. coli, but there are specific antimicrobial use “drivers”, particularly the use of 4^th generation dry cow therapy tubes, amoxicillin-clavulanate and fluoroquinolones, which select different types of ABR E. coli (Schubert et al, 2021 Applied and Environmental Microbiology, 87:e01468-20; Alzayn et al, 2020 Journal of Antimicrobial Chemotherapy 75:2471-2479; Mounsey et al, 2021 Journal of Antimicrobial Chemotherapy 76:3144-3150). Investigation of whether antibiotic usage reduction results in ABR reduction on these dairy farms is ongoing.

We have shown that feeding raw meat to dogs is a very strong risk factor for them excreting ABR E. coli, and that walking dogs close to cattle can result in sharing ABR bacteria between cattle and dogs (Mounsey et al, 2022 One Health, 14:100370; Sealey et al, 2022 Journal of Antimicrobial Chemotherapy, 77:2399-2405; Sealey et al, 2023 One Health 17:100640). We are currently surveying uncooked meat that might be fed to dogs (and handled by humans) to determine whether this explains the specific types of circulating E. coli found in dogs.

We completed a survey of mammals at Bristol zoo, showing extensive circulation of ABR E. coli (Sealey et al, 2023 Journal of Antimicrobial Chemotherapy 78:1667-71).

Within the same study area, have sequenced >1000 urinary and >1500 bloodstream E. coli, with longitudinal sampling from 2018-2023 so far, and more sampling ongoing.  We have developed bioinformatic pipelines to identify ARB within these genome sequences (Reding et al, 2024 Briefings in Bioinformatics 25:bbae057). We have shown that antimicrobial usage in humans has a large impact on the prevalence of ABR E. coli urinary tract infections in humans (Hammond et al 2020 PLoS One 15:e0232903) and have identified emerging mechanisms of resistance to the first choice urinary antibiotic nitrofurantoin (Dulyayangkul et al, 2024 Antimicrobial Agents and Chemotherapy 68:e0024224, and the first choice bloodstream infection antibiotic piperacillin/tazobactam (Dulyayangkul et al, 2024 PLOS Pathogens 20:e1012235).

We have found very little evidence that ABR E. coli found in animals in our study region are an immediate cause of ABR infections in humans (Alzayn et al, 2020 Journal of Antimicrobial Chemotherapy 75:2471-2479; Mounsey et al, 2021 Journal of Antimicrobial Chemotherapy 76:3144-3150; Findlay et al, 2021 Applied and Environmental Microbiology 87:e01842-20; Sealey et al, 2022 Journal of Antimicrobial Chemotherapy, 77:2399-2405; Sealey et al, 2023 One Health 17:100640).

We are currently following up on the impacts of antimicrobial usage changes in farming and in human medicine, and testing our predictions about how these changes will effect the ecology of ABR E. coli in our region. We are focussing on transmissible resistance to first line agents used to treat urinary tract infections and how knowledge of emergent co-resistance can be used to inform antibiotic prescribing practices. We are interested in long-term colonisation and its impacts on infection, and how One Health transmission can be measured in this context. We are starting to sample water within the region to identify possible reservoirs of ABR bacteria.