Prevalence of and factors associated with antimicrobial resistant escherichia coli from indigenous poultry and farmers in Kitui Rural Sub-County, Kenya

Abstract

Introduction: Antimicrobial resistance (AMR) is a global public health threat amplified by inappropriate antimicrobial-use in humans and poultry. However, data on AMR in households are limited. Escherichia coli has been proposed as one of the pathogens to be used for AMR surveillance. Objective: This study estimated prevalence of resistant E. coli and associated factors among farmers and poultry in Yatta/Kwavonza and Kanyangi Wards, in Kitui rural sub-County, Kitui County, Kenya. Methods: Study design: A cross-sectional study was conducted between July 2017 and January 2018. Study population was poultry farmers and indigenous poultry in two randomly selected wards. We targeted households with ≥3 indigenous poultry and farmers’ aged ≥18 years. Sampling strategy: Number of households sampled per ward were calculated proportionate to size and geocodes randomly generated using ArcGIS to identify household to be sampled. Data collection tools: Semi-structured questionnaires were used to collect data on demographics, poultry management and antibiotics. Inhibition zones were measured using a ruler and used a camera to capture indigenous poultry and zones of inhibition. Sample collection: In each household, stool sample from one household member was collected. Cloacal swabs from three poultry were also obtained and pooled together to form one sample for that particular household. E. coli was isolated and drug sensitivity testing done using disc diffusion assay. Multi-drug resistance (MDR) was defined as resistance to ≥3 antibiotics. Data entry: Data was entered into EPI databases. Data analysis done using Ms Excel 2007 and Epi Info. Proportions for individual 10 antibiotics was calculated and Odds ratio (OR) with 95% Confidence Intervals (CI) to identify factors associated with AMR in poultry and farmers. Data presentation was in prose and tables. Results: A total of 134 farmers were enrolled, 91 (68%) from Yatta/Kwavonza and 43 (32%) Kanyangi with a mean age of 44 and 48 years respectively. Overall, 134 farmers’ stools and 134 poultry cloacal swabs were collected. E. coli was isolated from 82 (62%) farmers among whom 59 (72%) were from Yatta/Kwavonza and 23 (28%) from Kanyangi. Fifty (84.7%) farmer E. coli isolates in Yatta/Kwavonza and 18 (78.3%) Kanyangi had resistance to at-least one antibiotic. Tetracycline was the antibiotic with the most resistant in both Yatta/Kwavonza 25 (42.4%) and Kanyangi 9 (39.1%). In poultry E. coli was isolated in 90 (67%) of the sample collected, 61 (68%) from Yatta/Kwavonza and 29 (32%) Kanyangi. Resistance to at-least one antibiotic was observed in 57 (93.4%) in Yatta/Kwavonza and 26 (89.7%) Kanyangi in poultry E. coli isolates. Amoxicillin 29 (47.5%), in Yatta/Kwavonza and streptomycin, 15 (51.7%) in Kanyangi were the most resistant antibiotics. Multidrug resistance was demonstrated in 24 (41%) and 10 (43.5%) farmers E. coli isolates and in 23 (38%) and 16 (55%) poultry E. coli isolates in Yatta/Kwavonza and Kanyangi respectively. There were no statistically significant factors associated with antimicrobial resistant E. coli in relation to poultry husbandry or antibiotic use in farmers and poultry. Conclusion: The study found a high AMR prevalence in poultry and farmers with significant levels of MDR. No factors were statistically responsible for E. coli resistance. Recommendation: Genotyping of all the E. coli isolates in this study and further research to investigate other causes for AMR at poultry-farmer interphase.