Short communication: Selection of extended-spectrum β-lactamase-producing Escherichia coli in dairy calves associated with antibiotic dry cow therapy—A cohort study

Abstract

Antimicrobial residues in milk have been discussed as a possible selector for Enterobacteriaceae that produce extended-spectrum β-lactamases (ESBL) in dairy herds. Such residues are found in waste milk after antibiotic treatment of mastitis, but antibiotic dry cow therapy might also lead to antibiotic residues in colostrum and in milk during early lactation. While it is known that feeding of waste milk selects ESBL bacteria in calves, this was not investigated for colostrum yet, which is supposed to contain much lower antibiotic concentrations than waste milk. In this observational prospective case study on 2 farms, we hypothesized that blanket dry cow treatment with β-lactams would have more selective (here: increasing) effects on ESBL concentrations than selective (here: individually chosen) antibiotic dry cow therapy. Thus, we compared concentrations of ESBL-producing Enterobacteriaceae in feces of calves (n = 50) at 2 dairy farms with different management of antibiotic dry cow therapy. Considerably higher concentrations of ESBL-producing Escherichia coli were observed in blanket antibiotic dry cow therapy on d 3 of the calf's life (7.6 vs. 5.3 log cfu/g of calf feces). Both farms used narrow-spectrum penicillin combined with aminoglycosides for drying off, and the majority of ESBL isolates (93%) were co-resistant to aminoglycosides. No waste milk was fed to calves and no calf was treated with β-lactam antibiotics or aminoglycosides during the first 3 d of life, thus differences were most likely associated with different frequency of antibiotic dry cow therapy on farms (19 of 25 mother cows on farm A, 9 of 25 on farm B). Even though the presumable selection effect of antibiotics used for drying off decreased within the next 3 wk, this result further emphasizes the need for the reduction and prudent use of antibiotic dry cow therapy on farms.