Salmonella is a significant cause of foodborne illness in the U.S., resulting in an estimated 1.3 million human cases, 15,600 hospitalizations and 550 deaths each year. Escherichia coli O157 that produce shiga-like toxin (STEC) results in an estimated 62,458 human cases per year (0.45 percent of all foodborne illnesses), it is estimated to cause 1,843 hospitalizations and 52 deaths each year.

The beef industry has implemented many intervention strategies in harvest facilities to reduce the likelihood of carcass contamination with STEC, Salmonella and other potential foodborne pathogens.
In addition to these post-harvest strategies, there continues to be interest in pre-harvest strategies for reducing the pathogen loads in the gastrointestinal tracts of animals or on the hides of animals presented for harvest. To understand the potential for pre-harvest intervention, it is important to understand the distribution of these pathogens in the feedlot setting.

The Animal and Plant Health Inspection Service (APHIS) of USDA has recently released information about these two pathogens in U.S. feedlots.

"In 1994 we looked at feedlot management practices, cattle attributes and the presence of Salmonella and E. coli," says Dave Dargatz, DVM, PhD, USDA:APHIS Veterinary Services. "One reason that we wanted to evaluate these organisms again in the feedlot environment in the recent study was to take another look at the factors we saw that were associated with the organisms being present in the first study. We're still doing that analysis and hope to have something to report early next year."

Salmonella
In 1999, USDA's National Animal Health Monitoring System (NAHMS) conducted Feedlot '99 – a study of feedlots with 1,000-head-or-more capacity within the 12 leading cattle feeding states. These operations represented 84.9 percent of the U.S. feedlots in 1999 and contained 96.1 percent of the U.S. cattle inventory on feedlots with 1,000-head-or-more capacity on Jan. 1, 2000.
As part of this study, 73 feedlots were recruited to collect fecal samples from pen floors throughout a one-year period. In each feedlot, 25 fecal samples were collected from the floors of three pens. The pens were selected to represent cattle that had been on feed the shortest time, the longest time, and a randomly selected pen (75 total samples). Sampling occurred in each feedlot twice over the course of the year.

Overall, 6.3 percent of fecal samples were positive for Salmonella. There was little difference in the percentage of samples positive for Salmonella by type of pen.

The percentage of culture positive-positive samples collected within a pen ranged from 0 to 100. However, the median percentage of samples positive from all pens was 0. For pens that had at least one positive sample, the median percentage of samples positive was 14. It's important to note that because samples were taken from the pen floor, it cannot be determined how many individual animals were positive for Salmonella.

The largest percentage of samples was positive (11.4 percent) in the fourth sampling period, July through September.

Percent samples positive for Salmonella by season
October-December: 4 percent
January-March: 2.8 percent
April-June: 6.8 percent
July-September: 11.4 percent

"Others have seen this seasonal pattern in the past with higher levels in the warmer months," says Dargatz. "We don't know why this occurs. Some researchers have hypothesized that there could be more replication of the organism in the environment at these warmer times. Others have considered the possibility of more replication in feeds that are being presented to the animals and consequently there are more infections. Others have suggested that the types of feed fed during these seasons could have an impact." Dargatz adds that some have suggested there may be increased shedding during periods of heat stress. "The bottom line is that we really don't know why it occurs."
The percentage of samples positive for Salmonella differed by the geographic location of the feedlot. Samples from Southern operations were more likely to test positive (7.7 percent) than samples from Northern operations (4.8 percent).

Overall, 22.3 percent (94/422) of pens had one or more positive samples. The 94 positive pens were located in 50.7 percent (37/73) of feedlots where samples were collected.

In this study, the most common serotypes of Salmonella recovered from samples collected from feedlot pen floors were dissimilar from those most commonly associated with human illness or cattle diagnostic specimens, with the exception of S. newport (see table)

"I think we tend to lump Salmonella together in one large group," says Dargatz. "And probably under a certain set of circumstances any of the Salmonella serotypes could cause illness. However, what we see from the public health surveillance data is that some Salmonella serotypes are much more commonly associated with human illness than others. To some extent this may be a reflection of the sources, including various food sources, for the various serotypes. But, it is also probably related to the attributes of the specific Salmonella serotype such as the infectious dose and the virulence."

E. coli
Fecal samples were collected for identification of STEC in the same pens that were sampled for Salmonella

Overall, 11 percent (1,148/10,415) of fecal samples were culture positive for STEC. The largest percentage of samples positive for STEC came from pens where the cattle had been in the feedlot the shortest time. The largest percentage of samples was positive (19.9 percent) in those collected in September.

Samples were collected from 422 cattle pens, with 58.8 percent (248/422) having one or more positive samples. Most (52.4 percent) of the positive pens had 1, 2 or 3 positive samples. There was no geographic trend in the percentage of pens with culture positive samples. The percentage of samples culture positive from the Northern Middle and Southern regions was 11.5 percent, 8.4 percent and 13 percent, respectively. All feedlots have one or more positive samples during the course of the study.

Data from NAHMS' 1994 Cattle On Feed Evaluation (COFE) reported 1.6 percent of samples positive for STEC. There are two important differences between the Feedlot '99 study and COFE that may explain these disparities:

  • A different, more sensitive culture method was used in the Feedlot '99 study.
  • COFE was carried out from October through December, when the expected STEC prevalence is lower.

STEC appears to be widely distributed in cattle populations at feedlots.

Reducing pathogen loads in the feedlot is a goal of feedlot operators and veterinarians, but it's not an easy task. "I think a lot of folks are searching for useful interventions in the feedlot environment," says Dargatz. "But right now we don't have much to offer the feedlot industry as far as solid recommendations go."

This story appeared in the January 2002 edition of Bovine Veterinarian magazine.