Toxic bugs taint large numbers of cattle.

Each year, a particularly virulent food-poisoning agent sickens some 73,000 people in the United States, killing about 60. Though scientists have known that beef is the primary route of exposure to this bacterium--an Escherichia coli strain known as O157:H7--most tests had indicated that no more than 2 percent of sampled cattle were infected.

A federal study now suggests that those earlier surveys grossly underestimated the share of animals carrying the fecally transmitted germs. Using more sensitive assays than had traditionally been employed, investigators found that 28 percent of cattle entering large slaughterhouses were contaminated.

Last July and August, scientists from the Agriculture Department's Meat Animal Research Center in Clay Center, Neb., visited four major Midwestern meat processing plants. At each site, the researchers followed several lots of animals--each from a single ranch or feedlot--from stunning through butchering. Overall, they swabbed hides of 357 animals and meat from 330. They also sampled feces still in the intestinal tract of some 330 cattle.

Their assay turned up E. coil O157:H7 bacteria on hides or in feces of at least some animals in 21 of 29 lots sampled. More significantly, 43 percent of the skinned carcasses were tainted, the scientists report in the March 28 PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES.

E. coli doesn't penetrate muscle, so the meat should be sterile immediately after skinning, explains veterinarian William W. Laegreid, a member of the research team. Any O157:H7 bacteria on the carcasses must have come from contamination during slaughter, he observes.

"There had been an assumption that much of the carcass contamination was coming from the hide," he notes. "We were therefore a little surprised when we didn't find a tight correlation [between the two]."

The good news, Laegreid points out, is that by the end of processing only 2 percent of sampled meat bore contamination. This is less than a tenth of the contamination seen in carcasses--evidence that meat processors have very effective decontamination procedures, he says.

To further lower the share of infected meat exiting a processing plant, Laegreid argues that farmers may need to substantially reduce contamination in their livestock. That will prove a challenge on two counts, he notes. First, the bacterium is already endemic. In one study of animals reared on isolated ranges, he recalls, "we couldn't find an uncontaminated herd, basically."

More importantly, Laegreid points out that livestock managers "have absolutely no proven, effective treatment to reduce O157:H7 infection in cattle." Currently, he and others are exploring a host of possible options, including the development of vaccines, chlorination of livestock drinking water, and cattle diets laced with bacteria designed to kill O157:H7 or keep it from colonizing the animals' guts.

Microbiologist Alison D. O'Brien of the Uniformed Services University of the Health Sciences in Bethesda, Md., is among those investigating ways to block the infection of livestock. However, she and a colleague argue in their commentary accompanying the new report that the motivation for targeting O157:H7 on the farm should go beyond concern about tainted meat.

Germs shed in manure can enter water and plants, including crops, O'Brien notes. Indeed, she observes, "a fair number of small [O157:H7] outbreaks have been associated with alfalfa sprouts, lettuce, and unpasteurized apple cider." The bacteria may even be spread by flies, Laegreid adds.

The large-scale contamination reported by Laegreid's group is "a big deal," she says. "It not only means we have to be even more careful about food safety, but we also need to realize that cattle are contaminating the environment more widely than we had been aware of."