By: Dr. Scott Dee
I read a recent publication, written by Hong and others from the University of Minnesota (Serotypes and antimicrobial resistance in Salmonella enterica recovered from clinical samples from cattle and swine in Minnesota isolates 2006-2015, PLOS One). The research team evaluated trends in antimicrobial resistance (AMR) across Salmonella isolates sourced from the University of Minnesota Veterinary Diagnostic Laboratory. Samples were compared over a 9-year period of time and trends in the frequency of Salmonella strains recovered, along with any changes in AMR across species groups were analyzed. A total of 2537 strains tested originated from swine samples submitted to the laboratory, in contrast to 1028 from cattle. The authors reported that the types of Salmonella detected on farms are changing over time and that more isolates from pigs demonstrated AMR than isolates from cattle. The authors then concluded that the results of their study provide “insight into the dynamics of AMR in Salmonella in Minnesota livestock and can help monitor emerging trends in AMR”.
Overall, I liked the study. I felt it was well written and appreciated the sample sizes employed in the analysis. The fact that different livestock species were involved (some poultry data would have been nice) and that a lengthy (9-year) period of time was used were also strengths. Finally, the study also involved an excellent diagnostic facility and used nationally-standardized methods (MIC: Minimum Inhibitory Concentration) were used to test for AMR were good approaches.
However, the question that comes to mind is whether this type of sampling is truly representative of trends in AMR at the population, or herd level. We must remember that these samples are biased, as they were “conveniently” collected from the laboratory computer. To further bias the outcome, the isolates originated from tissues collected from subsets (n =1?) of dead animals submitted from clinical cases of disease. In other words, the results may not represent what is happening in the rest of the population (i.e. the healthy animals in the herd that were not sampled) or what is present in the environment (the feedlot or the wean-to-finish facility) in which the animals were raised.
Therefore, while I applaud the authors for their efforts and I am all in favor of collecting data from traditional diagnostic sources, I feel it’s important that we think “out of the box” and attempt to develop alternative means of tracking AMR through testing healthy animals at the population level and in their environment. This is underway at Pipestone Applied Research and we will let you know what we learn. Who knows? The results could be the same or perhaps the patterns and trends in AMR are different? Perhaps these traditional samples over-estimate the level of AMR in commercial livestock?
Maybe, maybe not, but we will never know until we try. Stay tunedJ