Abstract
We investigated the farm-level impact of the use of several different antimicrobial agents on the population of antimicrobial-resistant commensal bacteria of animal origin to appropriately assess the release risk of resistance. This study was carried out based on the results of a survey on the history of antimicrobial drug use in 297 pig farms and antimicrobial susceptibility testing for the 545 Escherichia coli isolates (one or two isolates/pig/farm). A comparative analysis with the nonexposed herd revealed that ampicillin (ABPC) resistance in E. coli increased in the herds that were exposed to penicillin (relative risk [RR], 1.75) and penicillin–streptomycin (RR, 2.28); dihydrostreptomycin (DSM) resistance, in the penicillin–streptomycin-exposed herd (RR, 1.75); and trimethoprim (TMP) resistance in the methoprim–sulfonamide-exposed herd (RR, 2.10). On the other hand, ABPC and DSM resistances increased in the tetracycline-exposed herd (RR, 1.66 and 1.58, respectively); TMP resistance, in the penicillin-exposed herd (RR, 1.77); and oxytetracycline and kanamycin resistances, in the penicillin–streptomycin-exposed herd (RR, 1.28 and 2.22, respectively). These results demonstrated that the development of cross-resistance and coresistance, imposed by the therapeutic use of the antimicrobials studied, contributed the farm-level prevalence of antimicrobial-resistant E. coli and that the influence of coselection was characteristic to individual antimicrobial agents used.