WASHINGTON, June 8 (UPI) -- The U.S. Department of Agriculture claims it tested 500 cows with signs of a brain disorder for mad cow disease last year, but agency documents obtained by United Press International show the agency tested only half that number.
USDA officials said the difference is made up in animals tested at state veterinary diagnostic laboratories, but these animals were not tested using the "gold standard" test employed by the agency for confirming a case of the deadly disease. Instead, the state labs used a less sensitive test that experts say could miss mad cow cases.
In addition, the state lab figures were not included in a March 2004 USDA document estimating the number of animals most likely to be infected among U.S. herds, and apparently were not given to a congressional committee that had requested agency data on the number of cows with brain disorder signs that had been tested for the disease.
"This is just adding to the demise of USDA's credibility," said Felicia Nestor, senior policy adviser to the Government Accountability Project, a group in Washington, D.C., that works with federal whistleblowers.
"If the USDA is going to exclude from testing the animals most likely to have the disease, that would seem to have a very negative impact on the reliability of their conclusion," Nestor told UPI.
Nestor, who has monitored the USDA's mad cow surveillance program closely for several years, asked, "Are they deliberately avoiding testing animals that look like they have the disease?"
Concerns about the number of cows in U.S. herds with brain disorder symptoms have been heightened due to the recent case in Texas, in which USDA officials failed to test an animal with such symptoms, also known as central nervous system or CNS signs. This was a violation of USDA policy, which stipulates all CNS cows should be tested because they are considered the most likely to be mad cow infected. To date, the Washington cow that tested positive last December is the only confirmed case of mad cow disease -- also known as bovine spongiform encephalopathy -- among U.S. herds.
The Texas incident has alarmed the public and members of Congress because humans can contract a fatal brain disorder called variant Creutzfeldt-Jakob disease from consuming meat infected with the mad cow pathogen. If the USDA's surveillance program is allowing the riskiest cows to go untested, it raises concerns about the ability of the monitoring system to detect the disease reliably in U.S. herds, Rep. Henry Waxman, D-Calif., charged in a May 13 letter to Agriculture Secretary Ann Veneman.
Dr. Peter Lurie, of the consumer group Public Citizen, said CNS cows should be the one category that absolutely has to be tested to have a sound surveillance system.
"CNS animals are far and away the most important animals to test," said Lurie, who has done several analyses of the USDA's mad cow surveillance program.
"If there's any category that needs 100 percent testing, that's it, because they would be the most likely place to find mad cow in America," he told UPI. "Any CNS cow that slips into the food supply represents a major case of malpractice by USDA, and similarly, the failure to test the brain of that animal to see if it was indeed infected is really a failure to protect the public."
USDA officials said the agency has no estimate on how many CNS cows occur in U.S. herds. But spokesman Ed Loyd has told UPI, and at least one other media outlet, that 500 CNS cows were tested in fiscal year 2003. Yet agency testing records for the first 10 months of FY 2003, obtained by UPI under the Freedom of Information Act, show only 254 animals that fall under the CNS category -- or about half the number Loyd cited.
After failing to respond to repeated requests from UPI for clarification of the apparent discrepancy, Loyd finally offered the explanation that an additional 45 CNS cows were tested by the USDA during the final two months of FY 2003. The remainder, he said, was made up by CNS cases tested at various state veterinary diagnostic laboratories.
"We also include data reported to us from state veterinary diagnostic laboratories, and all of these are CNS cases that have been tested for BSE using a histological examination," Loyd said.
"We were not using any other labs during this period, other than (the USDA lab), to run the IHC tests for BSE, which is the gold standard," he said. "This (state laboratory) information contributes important data to our surveillance effort."
However, the state labs did not use the immunohistochemistry test, which the USDA has called the "gold standard" for diagnosing mad cow disease. Instead, the labs used a different test called histopathology, which the USDA itself does not use to confirm a case, opting instead for the more sensitive IHC test.
The histopathology test, unlike the IHC test, does not detect prions -- misfolded proteins that serve as a marker for infection and can be spotted early on in the course of the illness. Rather, it screens for the microscopic holes in the brain that are characteristic of advanced mad cow disease.
According to the USDA's Web site, histopathology proves reliable only if the brain sample is removed soon after the death of the animal. If there is too much of a delay, the Web site states, it can be "very difficult to confirm a diagnosis by histopathology" because the brain structures may have begun to disintegrate.
That is one reason the agency began using the IHC test -- it can confirm a diagnosis if the brain has begun disintegrating or been frozen for shipping.
The state labs used histopathology to screen 266 CNS cases in FY 2003, as well as 257 cases in FY 2002, according to Loyd. He did not explain why this information was not included in the testing records the agency provided to UPI and has not responded to requests for the identity of the state labs.
Linda Detwiler, a former USDA veterinarian who oversaw the agency's mad cow testing program, told UPI the histopathology test probably is adequate for screening CNS cows. If they have mad cow disease, she said, it would likely be an advanced stage that should be obvious.
Other mad cow disease experts, however, said having a back-up test such as IHC would be advisable, because histopathology tests sometimes can miss evidence of infection.
The Food and Agriculture Organization of the United Nations offers similar recommendations in its protocol for conducing a histopathology test. The protocol states that even if histopathology is negative, "further sampling should be undertaken" in cases "where clinical signs have strongly suggested BSE" -- a criteria that includes all of the cows tested at the state labs.
The USDA seems to agree on the need for a back-up test. Its expanded surveillance program, which began June 1, calls for using IHC -- or another test called Western blot -- to confirm any positives found on rapid tests. The March 15 document that describes the new program does not mention using histopathology to confirm cases of mad cow disease.
"Subtle changes can be missed on histopathology that would probably not be as easy to miss using IHC," said Elizabeth Mumford, a veterinarian and BSE expert at Safe Food Solutions in Bern, Switzerland, a company that provides advice on reducing mad cow risk to industry and governments.
"Therefore I believe it is valuable to run (histopathology)," Mumford told UPI.
She noted that in Europe, two tests -- neither one the histopathology test -- are used to ensure no cases are missed. A rapid test is used initially for screening, followed by IHC as a confirmatory test.
Markus Moser, a molecular biologist and chief executive officer of the Swiss firm Prionics, which manufactures tests for detecting mad cow disease, agrees about the possibility of a case being missed by histopathology.
"There were cases which were (histopathology) negative but still clearly positive with the other (testing) methods," Moser said. "BSE testing based on histology on sub-optimal tissue was probably one of the reasons why Germany was allegedly BSE-free until our test discovered that they were not" in 2000, Moser told UPI.
He agreed with Detwiler that histopathology should be suitable for most cases of CNS cows, but added it still can fail to detect the disease in some CNS cases -- particularly if the sample is not optimum.
"It is difficult, if not impossible, to distinguish the subtle changes in a diseased brain from artifacts like ruptures in the tissue due to tissue damage during the sampling, transport or preparation," he said.
Loyd asserted the additional CNS cases from the state labs actually yielded a total of 565 such cows the USDA had tested -- 65 more than his original figure of 500. Whether the USDA considers its total to be 500 or 565, however, either figure would exceed the agency's own estimates for the total number of such cows that it identifies annually.
According to data the USDA provided to the House Committee on Government Reform, and numbers the agency included in the March document about its expanded surveillance plan, only 201 to 249 CNS cows are identified at slaughterhouses. Approximately 129 additional cases occur on farms annually. At most, that yields a combined total of 378 CNS cows, or nearly 200 less than the 565 Loyd claims the agency tested.
The USDA surveillance plan document makes no mention of the number of CNS animals tested at state veterinary diagnostic labs. The figure also does not appear to be included in the agency's estimates of the number of high-risk animals that occur in the United States each year. The latter number was used to help the USDA calculate the number of animals it will screen for mad cow disease in its expanded surveillance plan.
USDA officials also did not include the state lab figures in response to a question from the House Committee on Government Reform, a source close to the issue told UPI. The committee, on which Waxman is the ranking Democrat, had requested in a March 8 letter to Veneman that she provide "the number of BSE tests that were conducted on cattle exhibiting central nervous system symptoms" for each of the last five years.
Loyd did not respond to a request from UPI asking why agency officials did not provide that information to the committee or include it in USDA's explanation of its expanded surveillance plan.
The committee has taken note of the CNS issue and plans to delve into it further in a hearing slated for sometime in the next few months.
"The committee will explore this and other issues surrounding USDA and BSE testing at a hearing later this summer," Drew Crockett, spokesman for the committee, told UPI.
Steve Mitchell is UPI's Medical Correspondent. E-mail email@example.com