Panel Discussion II: Usefulness of in vitro and in vivo Model Systems for Predicting the Adverse Public Health Outcome for Antimicrobial Drug Residues in Food

Article English OPEN
Beaulieu, Andrew J. ; Boisseau, Jacques ; Cerniglia, Carl E. ; Corpetgt, Denis E. ; Fernández, A. Haydée ; Wilkins, Tracy D. ; Woodward, Kevin N. (2011)
  • Publisher: Microbial Ecology in Health and Disease
  • Journal: Microbial Ecology in Health and Disease (issn: 1651-2235)
  • Related identifiers: doi: 10.3402/mehd.v12i1.8342

Dr. Beaulieu: We asked the panelists to spend about ten minutes each, addressing two questions. The first question is which in vitro and in vivo models or methods should be considered to evaluate the public health risk associated with antimicrobial residues in food, and can the model or method be used to evaluate more than one microbiological or functional endpoint in one test? The second question is how should the results obtained be used to establish an ADI? Is the standard toxicology formula appropriate to calculate an ADI for antimicrobial residues?
  • References (4)

    Dr. Kotarski: As I recall, studies performed by Dr. Corpet sometime ago resulted in an ADI for resistant enterobacteria of less than 0.75 mg.

    Dr. Corpet: What we saw is that 0.5 ppm in the water was enough to select for resistance. When we go back to the mg:person:day, this is purely speculative. This is extrapolation. How could we translate 0.5 ppm in mice? In my opinion, it does not translate to 7.5. It translates to 0.75 as it is in your publication with Dr. Cerniglia. When doing my calculations, the results I obtained are not exactly the same as yours; my results are ten times lower.

    Dr. Corpet: In my study, it was 0.5 ppm and in yours it was 1 ppm, really equivalent. Those are not no-effect levels. Those are the minimum selecting concentration. We did not try at the no-effect level.

    Dr. Kotarski: My understanding is that the inoculum used in the chemostat system was different to the inoculum used in the HFA mice. Therefore, since N ¾ 1 in both cases, we do not have a good indication of what the variation is for any of the variables for either system at this point.

  • Metrics
    No metrics available
Share - Bookmark