WASHINGTON, July 16 (UPI) -- A new study suggests the mad cow disease test used by the U.S. Department of Agriculture in its new surveillance program may be unreliable and could be missing cases of the deadly disorder under the majority of conditions in which it is used.
The USDA's monitoring program, which began June 1, relies on a rapid test manufactured by Bio-Rad Laboratories of Hercules, Calif.
The Bio-Rad test, however, may be unreliable on brain samples that come from cows that die on the farm or other locations, according to a study by Japanese researchers that appears in the May issue of the Journal of Veterinary Medical Science. These types of animals account for approximately 70 percent of the animals the agency has tested so far, USDA officials told reporters earlier this week in response to a scathing report from the agency's Inspector General that cited numerous problems in the USDA's mad cow surveillance program.
Another type of test called a western blot performed reliably in the experiment, but this is a slower test that is not feasible for screening thousands of animals per week as the USDA now is doing. The agency has licensed a rapid version of this test, manufactured by the Swiss firm Prionics, but so far it has not been approved for use in the 12 labs conducting initial mad cow screenings.
"In this study, we showed that several problems undermine the utility of the (Bio-Rad test) with deteriorated samples, whereas (western blot) remains very dependable," Takashi Yokoyama and other researchers from the Prion Disease Research Center in Tokyo -- one of only three mad cow disease reference laboratories recognized by the international authority, the World Organization for Animal Health -- wrote in the journal.
"Therefore, (western blot) might be the only reliable procedure to detect (the mad cow pathogen) in severely damaged samples from fallen stock," the researchers concluded.
The western blot test used in the study was different from the Prionics test.
In the study, Yokoyama and colleagues simulated conditions of decaying carcasses, such as would be found in cows that are left in the field for one to four days after death.
The Bio-Rad test detected prions -- the agent believed to cause mad cow -- in all of the samples, but the signals began to drop off significantly after only one to two days of simulated decay. In addition, the Bio-Rad test failed to detect prions in the samples with higher dilution levels.
Testing experts said the study findings could pose problems for the reliability of USDA's mad cow surveillance plan because under its testing strategy cases that test negative might not truly be negative.
"This is quite heavy stuff," Markus Moser, a molecular biologist and chief executive officer of Prionics, told United Press International.
This means the Bio-Rad test could "result in mistakenly false results" on animals that lie around for a few days after death before a sample of their brain is collected, Moser said.
Bio-Rad Vice President Brad Crutchfield defended the company's test, saying it had been extensively tested by experts in the United Kingdom on 250 degraded samples from infected animals and had not missed a single case. This study is not yet published but was conducted by the U.K.'s Veterinary Laboratories Agency -- one of the three world-recognized mad cow testing labs -- which consequently chose Bio-Rad's test as its rapid test of choice. The test is also used in Germany, Belgium, Japan and Canada.
Crutchfield conceded, however, the western blot type of test often is the best choice when confirming liquefied samples.
"Indeed, for many governments around the world, the Bio-Rad western blot is often used," he said.
The USDA's Beth Johnson, an adviser to Agriculture Secretary Ann Veneman, told UPI, "Certainly, (the study is) a valuable piece of research ... that adds to our wealth of knowledge."
Johnson added the impact on the USDA's testing would be minimal, however, because "most of the samples we get from on the farm or dead animals are not at that point" where the tissues are decaying and in a liquefied state.
Johnson was unable to cite specific percentages for how many samples have been in a deteriorated state by the time they were tested.
Current and former USDA veterinarians said it may not be known how much time passes before a sample is collected from animals that die on the farm and it could be six days or longer -- time enough for significant decay to have set in.
"I don't think we have any idea (how much time passes before a cow's brain is sampled after death) because we don't have any access to farms," a USDA veterinarian who requested anonymity told UPI.
Lester Friedlander, a former USDA veterinarian who left the agency in 1995, said as many as nine days can pass in some cases. It could be two to three days after the animal dies before the farmer calls a renderer, who could in turn take two to three days to come out to the farm. It could be another two to three days before the USDA is notified and collects a sample, Friedlander said.
"The key question is what percentage of brains the USDA has sampled since June 1 are dead or have been deteriorating on the farm for a few days," Michael Hansen, a biologist and senior research associate with the watchdog group Consumers Union in Yonkers, N.Y., told UPI. "If that's any kind of significant fraction, they should be using other tests or retesting those animals."
"It does again raise yet more questions and makes you wonder about the whole program," Hansen said. "If they're testing all these dead and decaying animals, the deadstock, you'd think they'd be aware of all of the limitations and actually have tried to determine which test is most accurate."
The Japanese researchers said the problem with the Bio-Rad rapid test encountered in their study appears to be that as the brain tissue begins deteriorating it also may cause changes in the prion that interfere with the way the test operates.
Any test that relies on the same technology used in the Bio-Rad "might have problems with dead stock," Stuart Wilson, a molecular pathologist and scientific director at Microsens Biotechnologies, told UPI. Microsens, based in London, manufactures technology used in a mad cow test made by Idexx Laboratories, which has been licensed by the USDA but not approved for use yet.
Moser agreed and said his company makes a rapid test -- separate from its western blot test -- that uses similar technology to Bio-Rad's and it has encountered problems with detecting mad cow infection in decaying tissue samples.
Wilson said the Idexx test is similar to Bio-Rad's but uses a little different technology that might help it avoid problems with detecting infection in decaying brain samples.
Wilson cautioned the Japanese study is just one experiment that would need to be confirmed before the results are fully accepted.
Moser said it is not uncommon in Switzerland for the tissue samples to be somewhat deteriorated by the time they are tested.
"Most of the samples we get ... are mostly liquid," he said. It also depends on the time of the year.
"In summer, it's worse than winter," due to the hotter weather, which causes the carcasses to begin decaying more rapidly, he said.
If the Japanese results hold true, this could pose significant challenges for the early stages of USDA's monitoring program, which was initiated just as summer and warmer temperatures were beginning in the United States.
--
Steve Mitchell is UPI's Medical Correspondent. E-mail [email protected]
