Prevalence of antimicrobial resistance genes and its association with restricted antimicrobial use in food-producing animals: a systematic review and meta-analysis

Abstract Background There is ongoing debate regarding potential associations between restrictions of antimicrobial use and prevalence of antimicrobial resistance (AMR) in bacteria. Objectives To summarize the effects of interventions reducing antimicrobial use in food-producing animals on the prevalence of AMR genes (ARGs) in bacteria from animals and humans. Methods We published a full systematic review of restrictions of antimicrobials in food-producing animals and their associations with AMR in bacteria. Herein, we focus on studies reporting on the association between restricted antimicrobial use and prevalence of ARGs. We used multilevel mixed-effects models and a semi-quantitative approach based on forest plots to summarize findings from studies. Results A positive effect of intervention [reduction in prevalence or number of ARGs in group(s) with restricted antimicrobial use] was reported from 29 studies for at least one ARG. We detected significant associations between a ban on avoparcin and diminished presence of the vanA gene in samples from animals and humans, whereas for the mecA gene, studies agreed on a positive effect of intervention in samples only from animals. Comparisons involving mcr-1, blaCTX-M, aadA2, vat(E), sul2, dfrA5, dfrA13, tet(E) and tet(P) indicated a reduced prevalence of genes in intervention groups. Conversely, no effects were detected for β-lactamases other than blaCTX-M and the remaining tet genes. Conclusions The available body of scientific evidence supported that restricted use of antimicrobials in food animals was associated with an either lower or equal presence of ARGs in bacteria, with effects dependent on ARG, host species and restricted drug.