• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Animal & Veterinary

  • Print
  • Share
  • E-mail

CVM Samples Feed Ingredients for Bacteria Under NARMS

by Marcia L. Headrick, DVM, MPH, DACVPM

CAPT, U.S. Public Health Service

Center for Veterinary Medicine NARMS Coordinator

Collaborating Authors: Drs. Joseph Paige, David Wagner, and Robert Walker, FDA CVM
FDA Veterinarian Newsletter January/February 2004 Volume XIX, No 6

Feed ingredients could be a source of bacteria in the food-animal production environment. Some of those bacteria could be resistant to antimicrobials, creating a public health concern. CVM officials are using the established NARMS program to find out more about this possible link.

The potential for the selection of antimicrobial resistant bacteria in animal production settings and for those bacteria to negatively affect antimicrobial chemotherapy in human medicine is an important food safety and public health issue. While there may be multiple sources of antimicrobial resistant bacteria in the animal production environment, there are minimal data on the role of animal feeds and the ingredients used to manufacture these feeds in the development and dissemination of antimicrobial resistant bacteria in the animal production environment.

To address this issue, the Center for Veterinary Medicine (CVM) conducted pilot studies as a part of the National Antimicrobial Resistance Monitoring System-Enteric Bacteria (NARMS). The goal of these studies was to collect data on the prevalence of foodborne enteric bacteria present in animal feed ingredients and to determine the antimicrobial susceptibility profile of those isolates.

In addition, this study provided the opportunity to develop labora-tory methods for the testing of animal feed ingredients and to determine the feasibility of conducting a nationwide surveillance of animal feeds as part of NARMS.

CVM has divided the program into three phases.

Phase I: animal-derived proteins

The objectives of phase I of the NARMS Animal Feed Ingredient Studies were to:

  • Estimate the prevalence of Salmonella, E. coli, Campylobacter, and Enterococcus in animal-derived protein feed ingredients;
  • Identify Salmonella serotype diversity occurring in these commodities; and
  • Determine antimicrobial susceptibility profiles of the Salmonella, E. coli, Campylobacter, and Enterococcus organisms isolated from the animal feed ingredient samples.

Samples of animal-derived protein ingredients used in animal feeds were collected in 2002 from rendering plants across the United States by the Food and Drug Administration’s Office of Regulatory Affairs (ORA) Field Inspectors. Samples were shipped to the CVM’s Office of Research for analysis.

Composites were made from these samples and cultured for Salmonella, E. coli, and Campylobacter as described in the FDA Bacteriological Analytical Manual (BAM). The samples were cultured for Enterococcus using methods developed at CVM’s Office of Research. All samples were cultured in duplicate. The antimicrobial susceptibility profiles of isolates were determined using NARMS laboratory methods. The antimicrobial agents tested were the same as those used in the NARMS custom panel. National Committee for Clinical Laboratory Standards (NCCLS) interpretive criteria (susceptible and resistant breakpoints) were used when available.

The animal-derived protein feed ingredients cultured for Salmonella, E. coli, Campylobacter, and Enterococcus included:

  • meat and bone meal (72 samples);
  • blood meal (16 samples);
  • bone meal (2 samples);
  • feather meal (10 samples);
  • poultry meal, (17 samples); and
  • fish meal (5 samples).

Campylobacter was not isolated from any of the samples whereas E. coli was isolated from 40 percent. None of the E. coli was enterohemorrhagic E. coli.

Forty-two (34 percent) of animal protein derived feed samples were positive for Salmonella. There were 70 Salmonella isolates recovered and they were distributed into 27 identifiable serotypes. S. Tennessee was the most common serotype recovered, but it comprised less than 10 percent of the isolates (6 total).

In general, the Salmonella and E. coli isolates that the researchers recovered from these feed ingredient commodities were susceptible to all drugs tested, except tetracycline, to which 17 of the E. coli isolates were resistant.

Enterococus were isolated from 84 percent of the samples. These isolates were resistant to erythromycin (4.5 percent), penicillin (2 percent), tetracycline (18.5 percent), ciprofloxacin (8 percent), and streptomycin (2.5 percent). There was no resistance detected to quinupristin/dalfopristin (Synercid™) in any of the non-faecalis Enterococcus species. All isolates tested were susceptible to vancomycin and linezolid, with the exception of an isolate of E. gallinarum, which exhibited intermediate susceptibility to vancomycin.

These data were presented by FDA CVM scientists at the 2003 and 2004 American Society for Microbiology meetings.

Phase II: plant-derived protein

In 2003, 79 samples of plant-derived protein animal feed ingredients were collected as Phase II of the NARMS Animal Feed Ingredient Studies. These animal feed ingredients are primarily by-products of the oilseed industry. However, some cereal based products were received.

The following plant-derived protein animal feed ingredient samples were cultured for Salmonella, E. coli, and Enterococcus:

  • alfalfa meal/pellets (14 samples)
  • canola meal (2 samples)
  • corn (14 samples)
  • gluten (2)
  • cottonseed meal (8 samples)
  • hominy (1 sample)
  • linseed meal (3 samples)
  • soybean meal (30 samples), and
  • sunflower meal (5 samples)

The samples were collected by FDA ORA Field Inspectors from different sources across the United States. Inspectors from the FDA District Offices in Cincinnati, Dallas, Denver, Florida, Kansas City, Los Angeles, Minnesota, New Orleans, and Seattle were involved in collecting and submitting the samples.

Primary isolation and identification of the bacteria have been completed. Susceptibility testing of Phase II plant-derived protein isolates indicated that all Salmonella isolates (4 of 79 samples, or 5 percent) tested were susceptible to all 16 antimicrobials tested in the NARMS custom Sensititre panels. Forty-three percent of the plant based feed commodities were positive for E. coli. With the exception of tetracyline (9 percent) and cephalothin (13 percent), the 54 E. coli isolated were also susceptible to the 16 antimicrobials tested in the NARMS gram negative panel.

Ninety-one percent (72) of samples were positive for Enterococcus. A total of 162 enterococci were isolated and tested for susceptibility to 17 antimicrobials. All isolates were susceptible to chloramphenicol, penicillin, vancomycin, and linezolid. In addition, the non-fecalis species were susceptible to the streptogramin, quinpristin/dalfopristin (Synercid). Resistance was detected to erythromycin (9.3 percent), tetracycline (9.9 percent), and ciprofloxacin (6.2 percent). High-level kanamycin and streptomycin resistance was detected in 5.5 percent and 1.2 percent of enterococci isolates, respectively.

Phase III: complete feeds

Samples of complete animal feeds will be collected as Phase III of the NARMS Animal Feed Ingredient Studies in 2005. These samples will include approximately 50 samples from swine complete feeds. In addition, the program will include samples from the FDA Feed Contamination Program. Plans for continued sampling, particularly of animal-derived protein feed ingredients, are being considered.


Integrating animal feed data into the ongoing NARMS program will provide data needed to estimate the magnitude of animal feed contamination, understand the sources of contamination, evaluate changes in prevalence and susceptibility patterns of the enteric bacterial isolates over time, and help determine potential mitigation and control strategies to minimize the spread of resistant foodborne pathogens through animal feeds and thus through animal production environments.

Better understanding of the prevalence of antimicrobial resistant bacteria in animal feed ingredients will facilitate development of strategies to interrupt the spread of antimicrobial resistant bacteria to food animals via animal feed ingredients.

Once analysis of the data is complete, summary results of the NARMS Animal Feed Ingredient Studies will be posted on the FDA CVM NARMS web page. For more information on general antimicrobial drug resistance issues, visit CVM’s web site. NARMS reports are published annually.