Use this URL to cite or link to this record in EThOS:
Title: Patterns of antimicrobial resistant E. coli and genetic interplay between livestock, humans and their shared environment in a high-density livestock-human population in western Kenya
Author: Kemp, Steven
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Access from Institution:
Antimicrobial resistance (AMR) is a global One Health issue. There has been a significant increase in the rates of AMR infections in both humans and veterinary medicine. The dynamics of AMR in developing countries are poorly understood, especially in community settings, due to a paucity of data on AMR prevalence and the drivers of resistance. Previous studies in subSaharan Africa indicated a high prevalence of multi-drug resistance (MDR) amongst Enterobacteriaceae such as E. coli in both humans and livestock. In western Kenya, there is significant overlap between humans, livestock, and their shared environments, due to the high density of smallholder farmers. This increases the risk of bacterial transmission via direct or close contact between humans and animals, or indirectly via the shared environment. The aims of this study were to determine the knowledge, attitudes and practices (KAP) amongst antimicrobial users and providers; to estimate the prevalence of carriage and identify risk factors for AMR E. coli including to the highest priority critically important antimicrobials (HPCIAs) amongst humans, livestock and the environment in an area of high-density smallholder farms in western Kenya and; finally, to explore AMR E. coli faecal carriage amongst livestock slaughterhouse workers. Three studies were conducted; a cross-sectional KAP study of 147 antimicrobial users and prescribers understanding of AMR across Busia county; a cross-sectional study (farm) study collecting and characterising faecal E. coli from farmers, livestock and their shared environments, and finally examination and characterisation of faecal AMR E. coli amongst slaughterhouse workers. During the KAP and farm studies a structured questionnaire was used as a framework to interview all participants. During the farm study, E. coli isolates were isolated from farmers and livestock faeces, and from living environments and water sources on 70 farms across Busia. The slaughterhouse study was a retrospective study which collected faecal E. coli from workers in 142 slaughterhouses across western Kenya. All E. coli isolates were subjected to antimicrobial susceptibility testing using the disk-diffusion method and a subset were characterised by Whole Genome Sequencing (WGS). Non-sequenced isolates with an ESBL phenotype were subjected to PCR to determine ESBL resistance genes. Mixed effect logistic regression models were utilised to assess risk factors for carriage of AMR E. coli using questionnaire-derived data from the farm study. The KAP study indicated that antimicrobials are accessed via agrovet shops, with a large proportion (~40%) being sold without a prescription. Concerning knowledge of antimicrobials; less than half of agrovets had sufficient training regarding livestock health and disease, and a registered pharmacist was often not present to dispense veterinary antimicrobials. Detailed information regarding dosage, withdrawal periods and the risks of AMR were not routinely provided by agrovets or animal healthcare assistants (AHAs) to farmers at point-of-sale. The most commonly sold/purchased antimicrobials were broad-spectrum oxytetracyclines and penicillinstreptomycin. Due to a lack of diagnostic facilities, broad-spectrum antimicrobials were often used empirically. There was good record keeping of antimicrobials sold by agrovets, but few records kept by farmers or AHAs who treated animals. Withdrawal periods were acknowledged by 80% of farmers, but only 28% had a good understanding and strictly adhered to them. A high proportion of agrovets (69.2%), AHAs (39.7%) and farmers (29.0%) had never heard of AMR or 'resistance' before. Faecal carriage of AMR E. coli on smallholder farms was high amongst sampled humans, livestock, and environment, with 95.3% of samples being resistant to at least one class of antimicrobial. Resistance to tetracycline (89.2%), trimethoprim (71.0%) and sulfonamides (69.4%) most prevalent. There was a high prevalence of MDR isolates (53.9%), with sheep and goats having the largest proportion compared to other animals. There was a moderate prevalence of ESBL-E. coli (14.8%) with two ESBL hotspots were identified in Nambale and Butula sub-counties. Molecular characterisation of E. coli isolates indicated a high diversity of E. coli with several similar strains found across animal, human and livestock populations between farms, confirmed by a large number of multi-locus sequence types (MLST) common to all groups. The most prevalent STs were identified in all three groups. Four distinct clonal groups with common ancestors were identified, all associated with carriage of blaCTX-M ESBL resistance genes. There was suggestion of clonal spread, as each member of the groups had similarities between serogroups, plasmids carried and resistance and virulence determinants, however directionality could not be determined. Two major risk factors were associated with increased risk of AMR - antimicrobial use (AMU) and animal vaccination. There was a high proportion of faecal AMR E. coli amongst slaughterhouse workers (95.1%). There was a lower proportion of MDR (45.5%) and ESBL-E. coli (9.6%) compared to that found in farmers, livestock, and the farm environment. The largest proportion of ESBL-E. coli was found amongst mixed ruminant slaughterhouse workers. The most common resistance phenotypes were to tetracycline (86.2%), trimethoprim (59.3%) and sulfonamides (57.4%). A similarly wide diversity of E. coli was found amongst slaughterhouse workers, confirmed by numerous STs. Five isolates were identified as forming a clonal group of ST131 primarily amongst pig-only slaughterhouse workers. These were associated with serogroup O25:H4, the human pandemic clone and ESBL producer. Two other large clonal groups were associated with carriage of ESBL resistance genes, ST617 (blaCTX-M) and ST361 (blaSHV), also carrying ESBL resistance genes. These findings indicate that there are gaps in the knowledge of antimicrobial users and prescribers and prescribing of broad-spectrum drugs. There is a high prevalence of circulating AMR E. coli amongst humans, livestock, and the environment as well as a moderate prevalence of ESBL-E.coli associated with blaCTX-M-15 in all groups. This is important as there was little documented use of HPCIAs to provide selection pressure. There was overlap evident between humans, livestock, and the environment, as well as in slaughterhouse workers in rural settings. It is important to consider both direct and indirect transmission pathways of bacteria between humans, livestock, and the environment, using an integrated One Health approach to protect food safety and reduce the transmission of AMR bacteria. There is urgent need for surveillance systems to collect AMU and AMR data which will allow for targeted interventions and further identify reservoirs of resistance genes amongst different populations.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral