Animal & Veterinary
CVM Releases NARMS Retail Meat Survey Results
FDA Veterinarian Newsletter September/October 2004 Volume XIX, No V
The Center for Veterinary Medicine (CVM) on September 30 released its first annual National Antimicrobial Resistance Monitoring System (NARMS) retail meat survey report, which provides data on the prevalence of antimicrobial resistant foodborne pathogens and commensal bacterial among retail meat and poultry samples.
The data for the report were generated in a 2002 survey. To gather the data, personnel from labora-tories in six participating States collected approximately 40 retail meat samples from retail sites each month during the year. The retail meat samples they collected from each site consisted of 10 samples each of chicken breasts, ground turkey, ground beef and pork chops.
The NARMS retail meat surveillance represents a collaborative effort of the Food and Drug Administration, the Centers for Disease Control and Prevention (CDC), and the Foodborne Diseases Active Surveillance Network (FoodNet).
NARMS retail meat surveillance was initiated in 2002 after a feasibility study was conducted in Iowa. Participating 2002 FoodNet laboratories include those from Connecticut, Georgia, Maryland, Minnesota, Oregon and Tennessee. By January 2004, the number of FoodNet labora-tories had increased to 10, with the addition of New York, California, Colorado and New Mexico.
For the NARMS survey, retail meat samples are collected from local grocery stores by participating FoodNet laboratory personnel. A similar retail meat sampling scheme is followed by all NARMS FoodNet participants. All FoodNet laboratories culture for Campylobacter and Salmonella using standard methods described by FDA. Four sites (Georgia, Maryland, Oregon and Tennessee) also culture for the presence of enterococci and E. coli using FDA-described methods.
FoodNet laboratory personnel ship the bacterial isolates to CVM’s Office of Research in Laurel, MD. Upon receipt of the isolates, the Office of Research confirmed the identity of the bacteria and developed a comprehensive antibiogram (which is the antimicrobial susceptibility profile of an organism) for the Salmonella, E. coli and enterococcal isolates using the NARMS antimicrobial panels. Agar dilution is used to determine antimicrobial susceptibility patterns of Campylobacter species.
Antimicrobial susceptibility results are interpreted, where appropriate, according to internationally recognized standards established by the National Committee for Clinical Laboratory Standards (NCCLS). NCCLS is an international, voluntary standards-developing organization for healthcare.
All Salmonella and Campylobacter isolates are also subjected to Pulsed-field gel electrophoresis (PFGE) to determine genetic relatedness. Resultant PFGE patterns are submitted to the CDC-led PulseNet program, which is a national network for DNA fingerprinting of foodborne pathogens.
Results from the survey demonstrate that retail meats, in particular chicken breast, are contaminated with Campylobacter, including antimicrobial resistant variants. Salmonella may also be found on retail meats, in particular ground turkey. However, further studies are needed to determine the relationships between antimicrobial use in animal husbandry and the development of antimicrobial resistance in these organisms. In addition, more study is needed to explore mitigation strategies to reduce the presence of these foodborne pathogens on retail foods of animal origin.
The researchers said their observations also suggest that Enterococcus spp. and E. coli commonly contaminate retail meat products and that differences observed in antimicrobial susceptibility phenotypes may reflect the extent of use of antimicrobials in specific food animal production environments.
Enterococci of foodborne origin have not been conclusively identified as direct causes of clinical infections. Also, with the possible exception of E. coli O157:H7 and other shiga-toxin producing strains, the current data are insufficient to accurately assess the hazard and the potential public health risk associated with the presence of generic E. coli in foods, regardless of their antimicrobial resistance traits. Further study is also warranted to determine the significance and virulence potential of these organisms that contaminate retail food of animal origin.
All of the data are available in a full report on CVM’s website at NARMS Retail Meat Annual Report, 2002.
In 2002, 2,513 retail meats were analyzed for the presence of Campylobacter and Salmonella. This included 616 chicken breasts, 613 pork chops, 642 ground beef and 642 ground turkey samples. Campylobacter was isolated more frequently from chicken breast (47%) than from the other three meat types tested (ground turkey, 0.6%; pork chop, 0.8%; ground beef, 0.0%). C. jejuni was the predominant Campylobacter species identified, followed by C. coli.
Because there are presently no NCCLS-approved interpretive criteria (susceptible, intermediate, or resistant breakpoints) for Campylobacter, “resistance” refers to those isolates exhibiting ciprofloxacin minimum inhibitory concentrations (MICs) of > 4 µg/ml and erythromycin MICs of > 8 µg/ml. Fifteen percent of C. jejuni recovered from chicken breast exhibited minimum inhibitory concentrations (MIC) > 4 µg/ml to ciprofloxacin, as compared with 10% of C. coli. Nineteen percent of C. coli recovered from chicken breast exhibited MICs > 8 µg/ml to erythromycin, as compared with no C. jejuni (0%).
Salmonella was recovered from ground turkey (12%) more often than the other three meat types tested (chicken breast, 10.0%; pork chop, 1.6%; ground beef, 1.4%). S. Heidelburg was the predominant serotype recovered (n=35/153) and was more often associated with ground turkey samples (60%).
Overall, antimicrobial resistant phenotypes differed by Salmonella serotype and retail food of animal origin. For example, five multi-drug resistant S. Newport were recovered from ground beef, ground turkey and pork chops. The majority of S. Newport isolates exhibited resistance to at least nine antimicrobials, including cephalosporins, phenicols, and potentiated sulfonamides. Thirteen percent of Salmonella isolates exhibited resistance to gentamicin, and 11% of Salmonella isolates demonstrated resistance to ceftiofur. Nalidixic acid resistant Salmonella were isolated only from ground turkey and were predominantly S. Saintpaul (n=4/6). All isolates were susceptible to ciprofloxacin and ceftriaxone; however, a decrease in susceptibility to ceftriaxone was noted among ceftiofur resistant isolates. Indistinguishable Salmonella genetic DNA fingerprints (PFGE patterns) were also recovered from different retail meats collected at different sampling times, and from different States. This information suggests a possible common origin of particular Salmonella serotypes.
Enterococcus and E. coli
With regards to Enterococcus and E. coli prevalence, 1,574 meats were analyzed (only four of the NARMS/FoodNet sites participate in E. coli/Enterococcus surveillance). This included 390 chicken breasts, 390 pork chops, 399 ground beef and 395 ground turkey samples.
Sixty-eight percent of these retail meat samples were contaminated with E. coli. The majority of the 1,070 E. coli isolates recovered were susceptible to the antimicrobials tested. However, 52% were resistant to tetracycline, 36% to streptomycin, 27% to sulfamethoxazole, 19% to ampicillin and 14% to gentamicin.
Ninety-seven percent of the 1,574 retail meat samples were contaminated with enterococci. Among the 1,520 enterococci speciated, Enterococcus faecalis was the predominant species recovered (59%), followed by E. faecium (33%) and E. hirae (7%). Resistance to linezolid or vancomycin was not detected in any isolate, but high-level gentamicin resistance was observed in 9% of enterococci isolates and 52% (excluding E. faecalis isolates, which are intrinsically resistant) demonstrated resistance to quinupristin-dalfopristin.