Factors influencing the detection of antibacterial-resistant Escherichia coli in faecal samples from individual cattle

AIMS

To investigate whether on-farm antibacterial usage (ABU), environmental antibacterial-resistant (ABR) Escherichia coli prevalence, sampling and sample handling methodologies are associated with ABR E. coli positivity in individual faecal samples from dairy heifers.

METHODS AND RESULTS

Three hundred and sixty-four heifers from 37 farms were sampled via rectal or faecal pat sampling. Samples were stored at -80°C for variable periods before microbiological analysis. Data analysis was done through a multilevel, multivariable logistic regression approach. Individual rectal samples had increased odds of positivity for amoxicillin-, cefalexin- and tetracycline-resistant E. coli. Sample storage for 6-12 months was associated with decreased odds of finding amoxicillin- and tetracycline-resistant E. coli. On-farm ABU had little influence, and environmental ABR E. coli prevalence had no significant influence on the odds of sample-level positivity for ABR E. coli.

CONCLUSIONS

Sampling methodology and sample handling have a greater association than on-farm factors with the detection of ABR E. coli in individual faecal samples from dairy heifers.

SIGNIFICANCE AND IMPACT OF THE STUDY

Sampling and storage methodologies should be considered carefully at the point of designing ABR surveillance studies in livestock and their environments and, where possible, these methodologies should be standardized between and within future studies.

Limited phylogenetic overlap between fluoroquinolone-resistant Escherichia coli isolated on dairy farms and those causing bacteriuria in humans living in the same geographical region

BACKGROUND

Our primary aim was to test whether cattle-associated fluoroquinolone-resistant (FQ-R) Escherichia coli found on dairy farms are closely phylogenetically related to those causing bacteriuria in humans living in the same 50 × 50 km geographical region suggestive of farm-human sharing. Another aim was to identify risk factors for the presence of FQ-R E. coli on dairy farms.

METHODS

FQ-R E. coli were isolated during 2017-18 from 42 dairy farms and from community urine samples. Forty-two cattle and 489 human urinary isolates were subjected to WGS, allowing phylogenetic comparisons. Risk factors were identified using a Bayesian regularization approach.

RESULTS

Of 489 FQ-R human isolates, 255 were also third-generation-cephalosporin-resistant, with strong genetic linkage between aac(6')Ib-cr and blaCTX-M-15. We identified possible farm-human sharing for pairs of ST744 and ST162 isolates, but minimal core genome SNP distances were larger between farm-human pairs of ST744 and ST162 isolates (71 and 63 SNPs, respectively) than between pairs of isolates from different farms (7 and 3 SNPs, respectively). Total farm fluoroquinolone use showed a positive association with the odds of isolating FQ-R E. coli, while total dry cow therapy use showed a negative association.

CONCLUSIONS

This work suggests that FQ-R E. coli found on dairy farms have a limited impact on community bacteriuria within the local human population. Reducing fluoroquinolone use may reduce the on-farm prevalence of FQ-R E. coli and this reduction may be greater when dry cow therapy is targeted to the ecology of resistant E. coli on the farm.