Animal & Veterinary
DEVELOPMENT OF A STANDARDIZATION SUSCEPTIBILITY TESTING METHOD FOR CAMPYLOBACTER
by P. F. McDermott, Ph.D., S. M. Bodeis and R. D. Walker, D.V.M.
FDA Veterinarian Newsletter 2002 Volume XVI, No VI
Campylobacter is a leading cause of human bacterial gastroenteritis worldwide. It is estimated that there are more than 2,000,000 cases estimated each year in the U.S. C. jejuni and C. coli are the most commonly isolated campylobacter species in cases of human disease. Other important species associated with disease in humans include C. lari, C. jejuni subspecies doylei and C. fetus. Campylobacter is considered mainly a foodborne pathogen. Contaminated milk, water, chicken, pork, beef, lamb, pets, and seafood are all known to contribute to human infections. In the U.S., the majority of sporadic cases of Campylobacter infection have been linked to mishandled or undercooked poultry meats. Surveillance data show that approximately 70-80% of retail raw chicken meats are contaminated with Campylobacter.
Intestinal campylobacteriosis is usually a mild to moderate self-limiting diarrheal disease, accompanied by fever and abdominal cramping. As for most cases of acute diarrhea, treatment is usually supportive, consisting of rehydration and symtomatic therapy. Antimicrobial therapy is employed in relapsing or severe intestinal infections or when extra-intestinal infections occur, such as bacterimia, endocarditis, or meningitis. The latter conditions arise mostly in elderly and immunocompromised patients. When antibiotics are recommended, macrolides (e.g., erythromycin) and fluoroquinolones (e.g., ciprofloxacin) are the drugs of choice. Doxycycline and gentamicin are sometimes used as alternative drugs for treatment. For infections caused by C. fetus, meropenem is one of the treatments of choice. Decreased susceptibility of campylobacter, particularly to the fluoroquinolones, has made empiric therapy less reliable. Effective patient management is further complicated by the lack of a standardized in vitro susceptibility testing method and interpretive criteria, which makes choosing an appropriate anti-infective agent difficult. A standardized in vitro antimicrobial susceptibility testing method, including the appropriate quality control organisms, is essential for generating accurate and reproducible susceptibility testing results.
In the United States, the National Committee for Clinical Laboratory Standards (NCCLS) is the recognized organization for developing standards for in vitro susceptibility testing of bacterial pathogens. To develop a reliable testing procedure for Campylobacter, a multiple-laboratory trial was conducted in accordance with NCCLS guidelines. Preliminary studies had established a suitable growth medium (Mueller-Hinton agar with 5% sheep blood), atmospheric conditions (85% N, 15% CO2, 5% O2), incubation temperature (36oC, 48 hr) and identified C. jejuni ATCC 33560 as the quality control (QC) strain. In subsequent deliberations, the NCCLS requested a study to compare testing at both 36°C for 48 hours and 42°C for 24 hours. In this study, they requested that five antimicrobial agents commonly used to treat human campylobacteriosis (ciprofloxacin, doxycycline, gentamicin, erythromycin, and meropenem) be tested. The multi-laboratory trial also included testing a collection of human isolates of Campylobacter, in order to validate the method with clinical strains.
The participating laboratories were: Clinical Microbiology Institute, Wilsonville, OR; Focus Technologies, Herndon, VA; Duke University Medical Center, Durham, NC; Michigan State University, College of Medicine, East Lansing, MI; The Food and Drug Administration, Center for Veterinary Medicine, Office of Research, Laurel, MD; Danish Veterinary Institute, Copenhagen, Denmark; Division of Immunity and Infection, The Medical School, University of Birmingham, Birmingham, UK; Department of Microbiology and Public Health, University of Alberta Hospital, Edmonton, Alberta, Canada; and Abbott Laboratories, Abbott Park, IL. These laboratories were chosen for their expertise in susceptibility testing of Campylobacter.
In each laboratory, 10 replicates of C. jejuni ATCC 33560 and 21 human isolates of Campylobacter were tested daily for two days using agar dilution. The 21 clinical isolates consisted of five C. jejuni, five C. coli, five C. doylei, three C. fetus and three C. lari. Each isolate was tested against the five antimicrobial agents at both 36oC for 48 hr and 42oC for 24 hr. Testing involved 10 independent suspensions of the QC organisms, C. jejuni ATCC 33560, and each of the 21 human clinical isolates per day for two days. This resulted in a total of 11,340 data points for the human clinical isolates (21 isolates x 5 drugs x 3 medium lots x 2 days x 2 temperatures x 9 laboratories).
The minimal inhibitory concentration (MIC) results for the QC organism against the five antimicrobial agents were highly reproducible within and between laboratories and with both incubation conditions. For all drugs, the QC limits encompassed more than 95% of the observed values under both incubation conditions. The MIC results for the 21 human clinical isolates showed that they spanned an MIC range around the established QC ranges. Comparison of the two testing procedures showed that C. jejuni and C. coli could be reliably tested using either incubation temperature. In contrast, there were many instances where C. lari, C. jejuni subspecies doylei, and C. fetus isolates failed to grow at 42ºC. Given these variations in growth among different isolates of these species, it is recommended that these three Campylobacter species be tested only at 36ºC.
Several in vitro methods have been used to measure the susceptibility of Campylobacter to various antimicrobial agents. Disk diffusion is attractive due to its convenience and low cost. Some researchers report consistent results obtained by disk diffusion within a single laboratory. In preliminary multi-laboratory experiments not reported here, we were unable to advance disk diffusion as a standardized method, due to very poor intra- and inter-laboratory reproducibility. This problem was ascribed to the peculiar growth characteristic of Campylobacter. This resulted in widely different interpretations of zone sizes for the same strain/antimicrobial combinations, depending on the angle and intensity of the light source. Thus, disk diffusion should not be used for susceptibility testing organisms in this genus until growth conditions are identified that eliminate ambiguities in end point determinations.
The epsilometer testing method (Etest, Solna, Sweden) is widely used to measure the antimicrobial susceptibility of Campylobacter. This method involves a carrier strip coated with an antimicrobial gradient that is placed on a seeded agar plate. This technique is convenient, and has the advantage of providing an MIC value. In separate studies, we have compared the Etest with the NCCLS agar dilution method at 36oC and found that, in general, the Etest endpoints fall one or more dilutions below those observed using agar dilution. Deviations from the agar dilution results were greater for certain antimicrobial agents. The availability of the agar dilution test as a standardized method will provide a reference method that can be used to advance other susceptibility testing methods that may be amenable to routine laboratory use. For example, we are currently using the method reported here to develop a standardized testing method based on broth microdilution.
The QC ranges, testing conditions and testing method, as well as the QC organism, has been accepted by the NCCLS for the susceptibility testing of Campylobacter, and will be published in the the M31-A2 and M7-A6 documents in 2002. In addition to improving the management of patients being treated for infections caused by Campylobacter, a standardized method allows for comparison of data between laboratories. This improves monitoring of susceptibility trends over time, and for precise data in both clinical studies and diagnostic laboratories.
Dr. McDermott, Ms. Bodeis, and Dr. Walker are scientists in CVM's Division of Animal and Food Microbiology, located at the Office of Research, Laurel, Maryland.