Associations of antimicrobial and multidrug resistance in Escherichia coli from the fecal flora and housing environment of calves on dairy farms with atmospheric variables.

The objective of this study was to describe the non-linear relationships between antimicrobial resistance to at least one tested antimicrobial (AMR) or multidrug resistance (MDR; resistance to at least three antimicrobial drug classes) of Escherichia coli isolates from calf feces and from the calf environment on dairy farms, and farm-level atmospheric variables, as well as selected calf- and herd-level factors. Data sets of 601 fecal isolates and 777 environmental isolates, collected from 57 dairy farms on 204 and 212 sampling dates, respectively, were combined with atmospheric data closest to each farm (monthly average of values prior to each sampling date). By employing random forest models, we found that the maximum age of the sampled calves in the pooled fecal sample showed the highest importance for the outcome MDR in the calf fecal data set. For the outcome AMR in the calf data set and for both outcomes in the environment data set, all the atmospheric variables had the highest importance. In E. coli from the calf environment, AMR was negatively associated with the average, maximum, and minimum temperature in the month prior to sampling. In both the calf and environment data sets, both AMR and MDR were positively associated with increasing deviation from historical temperature. These findings warrant confirmation in larger data sets.IMPORTANCEThe dynamics of expansion and transmission of resistant bacterial populations and of antimicrobial resistance (AMR) genes within as well as between human, animal, and environmental reservoirs are complex. An association between average minimum ambient temperature and AMR rates in selected human pathogenic bacteria has recently been reported. The present study highlights the relative importance of atmospheric variables for the outcomes AMR and multidrug resistance (MDR), as well as associations between atmospheric variables and the AMR outcomes in two Escherichia coli data sets from calves' fecal and environmental samples from Swiss dairy farms. In both the calf and environment data sets, both AMR and MDR were positively associated with increasing deviation from historical temperature. In the face of rapid anthropogenic global warming, these findings warrant further research on the association between AMR in livestock species and atmospheric variables.