LETTER

This letter is in response to the article by Riddle et al. (4) that appeared in this journal in April 2000. As Chairholder of the National Committee on Clinical Laboratory Standards Veterinary Antimicrobial Susceptibility Testing (NCCLS V-AST) subcommittee, I wanted to provide additional information associated with the development of quality control (QC) ranges, in this case, for fluoroquinolones. The NCCLS is an independent volunteer organization composed of experts from government, industry, and the professions whose mission is to develop and communicate performance standards for laboratory activities. The V-AST subcommittee's mission is to establish performance standards for antimicrobial susceptibility tests for bacteria isolated from animals. The main documents pertinent to this mission are the M37A, which describes the data necessary for establishment of QC and interpretive criteria (1), and the M31A, which is the performance standard for veterinary antimicrobial susceptibility testing (2). Thus, the scope of the NCCLS V-AST includes the approval of QC ranges and interpretive criteria for fluoroquinolones approved for use in animals. The V-AST subcommittee has stated that a "class representative" for a given antibiotic class must be chosen with caution owing to the unique physicochemical, pharmacokinetic, efficacy, and microbiological activity differences among related analogs. To this end, there has never been a veterinary class representative fluoroquinolone identified and all drug sponsors have been encouraged to provide their own QC data for their respective molecules.

Consequently, the V-AST subcommittee has reviewed and approved quality control data for five fluoroquinolones approved for use in veterinary medicine (Table 1). Unfortunately, due to the length of time necessary to revise the M31A document, QC ranges for only two of these five fluoroquinolones are published in the current version. Since the publication of M31A in June 1999, new QC data for three other veterinary fluoroquinolones have been reviewed and approved and are presented in Table 1. So, for the five fluoroquinolones approved for use in veterinary medicine, the NCCLS V-AST subcommittee has approved interpretive criteria for three that are marketed for use in companion animals, i.e., enrofloxacin, orbifloxacin, and difloxacin (marbofloxacin QC data have not been presented to the V-AST subcommittee, and premafloxacin, an investigational candidate, has not been approved by the U.S. Food and Drug Administration), and two approved for use in food animals, i.e., sarafloxacin for use in poultry and enrofloxacin for use in poultry and beef cattle. (It should be noted that sarafloxacin has been withdrawn from the market by its manufacturer, effective October 2000.) The NCCLS Antimicrobial Susceptibility Testing subcommittee has approved QC ranges for ciprofloxacin (3). Ciprofloxacin is intended for human medical use, and while it may sometimes be used in companion animal medicine under extralabel provisions, its use in food animals is strictly prohibited by the U.S. Food and Drug Administration's Center for Veterinary Medicine.

It should be noted that the NCCLS M37A document (1) not only provides guidance for the development of QC data but also outlines the conduct of susceptibility testing that will generate data used to establish interpretive criteria. The process of providing MIC QC data involves numerous replications, resulting in several hundred data points, in a multilaboratory study, with data summarization and review by the V-AST subcommittee. The methods described by the authors in their publication, and the results generated, do not conform to this process nor have they been reviewed by the V-AST subcommittee. The NCCLS V-AST subcommittee encourages those laboratories conducting susceptibility testing of veterinary antimicrobial agents to obtain copies of the M31A and M37A, if they have not already done so, from the NCCLS (www.nccls.org) and to conduct their testing and interpretation of data in accordance with these methods.