Risk Factors for Antimicrobial Resistance in European Livestock Farming

Abstract

With the mass production and sale of various classes of antimicrobials, a growing number of antimicrobial classes have been used for the clinical treatment of humans, companion animals and livestock. As a result, antimicrobial resistance (AMR) in humans and animals is a challenging public health problem. As part of the EU-project “Ecology from Farm to Fork Of microbial drug Resistance and Transmission (EFFORT)”, this thesis aims to better understand AMR abundances in humans, animals and the environment, and to explore potential risk factors for AMR in livestock using genomic methodologies. Our studies analysed more than 9500 samples collected in nine European countries. AMR abundances of various sources, animals species and humans were quantified by using a variety of quantification approaches (next-generation sequencing (NGS), real-time PCR (qPCR) and minimum inhibitory concentration (MIC)). Farm-related characteristics, such as antibiotic use (AMU) and farm biosecurity were collected through standard questionnaires. The results demonstrate that qPCR analysis can be considered a valuable and cost-effective tool to assess AMR levels in large collections of animal, human, and environmental samples. AMR levels vary highly across animal species and sample sources, and largely agree with the results of metagenomic analysis. In pigs and broilers, we observed a decreasing trend in AMR levels along the production chain (‘from farm to fork’). A positive association was found between AMR and AMU. The impact of biosecurity measures on AMR is complex, which requires to be further analysed in a specific way. Although AMU and biosecurity measures were identified as main determinants, they could only explain a limited part of the variation in AMR. This suggests that there are still unstudied factors that need to be considered in future research. In addition, occupational livestock AMR exposure shows a relation to AMR abundance in faeces of livestock workers. Finally, we were able to give recommendations on an animal-specific sampling strategy to determine farm-level AMR in terms of individual sample numbers per farm. The findings of this thesis shed light on AMR variations and potential links of AMR between animals, humans and livestock environments. It is recommended to limit AMU at farms and implement appropriate biosecurity measures to control AMR in livestock. In the future, more in-depth longitudinal studies of AMR, AMU, and specific biosecurity measures will be needed, in which attention should be paid to the appropriate sample size.