Abstract
Dear Editors,
I am writing on behalf of the Veterinary Antimicrobial Susceptibility Testing (VAST) subcommittee of the Clinical and Laboratory Standards Institute (CLSI) with regard to the paper by Lund et al, 1 ‘Antimicrobial susceptibility in bacterial isolates from Norwegian cats with lower urinary tract disease’, published in the June 2015 issue of the Journal of Feline Medicine and Surgery (JFMS).
The authors are commended for providing new information on the bacterial isolates from cats with lower urinary tract disease. New data on bacteria cultured from cats can be helpful for clinicians and scientists who wish to improve treatment of lower urinary tract disease.
However, we have some questions and some points with regard to the susceptibility testing presented in the paper. The authors indicated in the Materials and methods section that ‘Susceptibility to antimicrobial agents was determined in accordance with the recommendations of the European Committee on Antimicrobial Susceptibility Testing (EUCAST)’. They indicated that susceptibility of isolates from cats was determined for ‘penicillin, ampicillin, amoxicillin with and without clavulanic acid, cephalexin, trimethoprim, trimethoprim/sulfonamide, tetracycline, fucidic [sic] acid, enrofloxacin, spiramycin, lincomycin and nitrofurantoin’. However, the EUCAST recommendations do not have any clinical breakpoints for isolates from cats. EUCAST only provides breakpoints for isolates from humans. Differences in pathogens, antimicrobial doses, and especially differences in pharmacokinetics and pharmacokinetic–pharmacodynamic relationships between cats and humans make the use of human breakpoints to predict susceptibility for cat isolates potentially unreliable. In the Discussion of the paper, the authors compared their results with those of other published studies. Unless each study uses the same standards for testing, such comparisons can lead to incorrect conclusions.
An examination of the EUCAST breakpoints indicates that clinical breakpoints are not listed for many of the drugs listed in the authors’ results. For example, some drugs are not used in humans (eg, enrofloxacin) and no human breakpoint exists. The authors tested susceptibility of Enterococcus and Pasteurella species to 13 antimicrobials, yet there are only a few antimicrobials for which EUCAST has a human breakpoint listed for these bacteria. The CLSI standard for testing Enterococcus species (M100) 2 indicates that cephalosporins, clindamycin and trimethoprim-sulfamethoxazole may appear active in vitro, but they are not effective clinically, and isolates should not be reported as susceptible. Most of the drugs listed in the susceptibility results for Enterococcus species isolates in the paper are not recommended for testing by the CLSI.
It is possible that the authors of this paper may have used the wild-type cutoff (also known as the epidemiological cutoff or ECOFF) to indicate susceptibility of these bacteria to the antimicrobials listed in their paper. The ECOFF and the breakpoint (also called the clinical breakpoint) are not the same. 3 They are determined using different criteria and the values are often quite dissimilar for many antimicrobials. An isolate with an MIC value above the ECOFF is not necessarily defined as ‘resistant’ using the breakpoint and interpretive criteria.
The VAST is a subcommittee of the CLSI. This is the only global organization that provides public standards for antimicrobial susceptibility testing for bacteria isolated from animals.4,5 The CLSI procedure for establishing susceptibility testing breakpoints is a consensus-driven process. Breakpoints for animals are determined after analyzing pharmacokinetic and pharmacodynamic data, MIC distributions, and considering clinical data. An examination of Table 2 from the CLSI–VAST document shows many differences between the human breakpoint and the veterinary-specific breakpoint for the same antimicrobial agent. The human breakpoint is often not accurate for animals and we encourage laboratories to use the criteria determined by the CLSI whenever possible.
We appreciate the opportunity to clarify some of the points addressed in the paper by Lund et al. 1 We invite authors of manuscripts submitted to JFMS to consult with our CLSI–VAST subcommittee when there are questions regarding susceptibility testing of bacterial isolates from animals. This will help ensure that the most accurate information possible can be presented, which will ultimately improve antimicrobial therapy for animals.