TEXT

Top Abstract Text References

Fluoroquinolones are a widely used class of antimicrobials that inhibit bacterial replication by their action on DNA gyrase (1, 2). Many members of this class are currently approved for use in humans; chief among these is ciprofloxacin, one of the most-prescribed drugs in the United States. Fluoroquinolones are also widely, and increasingly, used in veterinary medicine, with enrofloxacin being the most common. Recently, two additional members of this class have come into common use in companion animals: marbofloxacin and orbifloxacin.

Clinical laboratories routinely perform in vitro dilution susceptibility testing to determine the antimicrobial susceptibility of bacteria isolated from clinical samples. Interpretation of these tests requires that performance standards for antimicrobial MICs be set using a number of quality control (QC) organisms. The National Committee for Clinical Laboratory Standards sets such standards and has done so for many of the fluoroquinolones used in human medicine (7). Currently, however, no standard has been set for any fluoroquinolone approved for use in companion animals (a standard for enrofloxacin has been reviewed by the National Committee for Clinical Laboratory Standards and is currently considered provisional) (5, 8). It is common, therefore, for veterinary laboratories to use the standards for ciprofloxacin as representatives of those of fluoroquinolones used in animals, although no data exist to assess the reliability of this practice.

We sought to determine whether MICs determined by using ciprofloxacin provide an acceptable alternative to those of three fluoroquinolones used in veterinary medicine: enrofloxacin, marbofloxacin, and orbifloxacin. We chose for testing four representative QC strains: Staphylococcus aureus ATCC 29213, Enterococcus faecalis ATCC 29212, Pseudomonas aeruginosa ATCC 27853, and Escherichia coli ATCC 25922 (6). Each strain was grown overnight on Columbia agar with 5% defibrinated sheep blood; then colonies were used to inoculate brain heart infusion broth to match a 0.5 McFarland density standard (4). Samples were diluted 1:100 in normal saline, and then 0.1 ml of the diluted sample was used to inoculate test tubes (for an inoculum of ~1.5 × 10⁵ bacteria/ml). Ciprofloxacin was purchased from the U.S. Pharmacopeia, and enrofloxacin, marbofloxacin, and orbifloxacin were generously provided by their manufacturers (Bayer, Pfizer, and Schering-Plough, respectively) as pure analytical reference standards. Each antimicrobial was diluted twofold in cation-adjusted Mueller-Hinton broth (8) in a 1-ml final volume to achieve concentration ranges of 0.0024 to 10 µg/ml for the first three and 0.024 to 100 µg/ml for orbifloxacin. Tubes were inoculated and incubated for 24 h at 35°C; bacterial growth was compared to that of a positive control, without antimicrobial, and that control was judged by turbidity, by a single pellet of 2 mm, or by several pellets of smaller diameter. All tests were performed in triplicate, and control tubes without inoculum and without antimicrobial agent were included in each experiment.

As shown in Table 1, we found that the MICs of ciprofloxacin were within the accepted ranges for all tested organisms (7) (and those for enrofloxacin were within the provisional ranges under consideration by the NCCLS) (5). However, the MICs of all three veterinary fluoroquinolones failed to fall within the accepted ciprofloxacin ranges for at least one organism. The marbofloxacin MIC did not mirror that of ciprofloxacin for P. aeruginosa, while enrofloxacin MICs for the two gram-negative organisms, P. aeruginosa and E. coli, differed significantly from those of ciprofloxacin. Orbifloxacin showed the most generalized deficit; its MICs for all four organisms were different from those of ciprofloxacin. Although a representative antimicrobial need not necessarily match MICs of other drugs in the class, the failure of ciprofloxacin to do so means that, in the absence of pharmacokinetic data for the other antimicrobials tested here, there is no rationale to assign it the role of representative.

P. aeruginosa is an important veterinary pathogen against which fluoroquinolones are commonly used (3). Since all three of the drugs tested here failed to reproduce the ciprofloxacin MIC for a QC strain of this organism, we sought to further investigate the clinical significance of these findings. We chose 10 clinical isolates (obtained from the North Carolina State University Veterinary Teaching Hospital and the Rollins Animal Disease Diagnostic Laboratory) and tested the MIC of each for the four antimicrobials. Antimicrobials were tested in the following ranges: ciprofloxacin, 0.157 to 5.0 µg/ml; enrofloxacin, 1.25 to 5.0 µg/ml; marbofloxacin, 0.625 to 5.0 µg/ml; orbifloxacin, 0.625 to 5.0 µg/ml.

Four of the ten strains had a ciprofloxacin MIC of less than 1 µg/ml and so were considered susceptible (7), while the remaining six were resistant (Table 2). We anticipated that for those strains for which MICs of ciprofloxacin were low, MICs of the other three fluoroquinolones would be correspondingly low, while higher ciprofloxacin MICs would correspond to higher MICs of the others. We found, however, that there was no apparent correlation between ciprofloxacin MIC and those of any of the other fluoroquinolones. Most striking among these was orbifloxacin, for which all MICs, regardless of those for ciprofloxacin, were the same; 5 µg/ml.

The results presented here show that ciprofloxacin does not provide an adequate representative for assessing the in vitro susceptibility of fluoroquinolones used in veterinary medicine. MICs of three of these antimicrobials failed, for some bacterial species, to mirror the ciprofloxacin MIC when we used QC strains. We also found no correlation between ciprofloxacin susceptibility and that of the other three fluoroquinolones when using clinical isolates of P. aeruginosa, suggesting that, at least for this organism, ciprofloxacin does not provide an acceptable alternative for susceptibility testing. It also appears that for these P. aeruginosa isolates, ciprofloxacin is more active than the veterinary fluoroquinolones in vitro. This work shows that fluoroquinolones used in veterinary medicine require their own standardized MIC ranges to ensure accurate interpretation of susceptibility tests.