Short-term repeatability of measurements of antimicrobial susceptibility of Escherichia coli isolated from feces of feedlot cattle

Short-term stability of measurements of antimicrobial susceptibility of Escherichia coli isolated from feces of feedlot cattle is important in developing monitoring and surveillance programs. Frequent evaluations (i.e., daily) are resource intensive and in some situations may be impractical for long-term sampling protocols. Consequently, a point-in-time measurement will need to be used to represent conditions in the peri-sampling period. In this study, 30 fecal samples were collected from each of 6 cattle pens on a commercial cattle feedlot on 2 occasions separated by 48 hours. Escherichia coli was isolated from single and pooled samples. The isolates were tested for antimicrobial susceptibility against a panel of 17 antimicrobials. Resistance to 5 antimicrobials (ampicillin, nalidixic acid, streptomycin, sulfamethoxazole, and tetracycline) was detected in single and pooled samples from both sampling periods (days 1 and 3). The prevalence of isolates resistant to these 5 antimicrobials was 2% or higher in all treatment combinations except for pools obtained from day 3 samples. Lower levels of resistance to 6 more antimicrobials were detected inconsistently across the single and pooled samples. Logistic models constructed for the antimicrobials to which the E. coli isolates were most commonly resistant demonstrated that there were no significant differences between periods (P > 0.10) and between single and pooled samples (P > 0.20). The distribution of the number of antimicrobials to which isolates were resistant was consistent for the single samples across periods, but there appeared to be a lower prevalence of any resistance in day 1 pooled samples. A larger number of resistant phenotypes were detected in the single samples than in the pooled samples, and resistant phenotypes with prevalence of less than 2% were detected inconsistently across periods and single and pooled samples. Resistance to individual antimicrobials was consistent by all measures when the prevalence was at least 2%. Inconsistent results were obtained for antimicrobials to which resistance rarely occurred. The apparent inconsistencies do not appear to be related to external factors but rather to sampling intensity. Short-term stability is a plausible assumption under sampling strategies that are designed to detect specific levels of prevalence. However, when resistance levels fall below these levels, there will likely be fluctuations in the presence or absence of rare resistant phenotypes and in their prevalence and central tendency measures.