New Diagnostic Fast, Effective, Finds TB in Elephants

Posted By News On August 25, 2012 - 12:24pm

A serological test is highly accurate at finding tuberculosis infection in elephants, and can determine such infection years before culture, according to a study in the August Clinical and Vaccine Immunology. The issue is critical not only for elephants, which are an endangered species, but for human public health, because elephants are among the very rare animal species which are commonly infected by human strains of tuberculosis, says first author Konstantin P. Lyashchenko of Chembio Diagnostic Systems, Inc., Medford, NY.

Elephants are unusual when it comes to diagnosing M. tuberculosis. Generally, bacterial infections can be diagnosed much earlier with culture than with serology. "Serology is considered to be a late diagnostic marker of tuberculosis in humans and various animal species," says Lyashchenko. "However, efforts to isolate M. tuberculosis from live, infected elephants have proven very disappointing, whereas antibody assays in such animals showed promise for early detection of TB," says Lyashchenko.

Lyashchenko and his collaborators first realized the promise of serology for diagnosing TB in elephants, in 2004, when they were asked to test serum samples from a zoo elephant that had been diagnosed with M. tuberculosis in 2000. They requested serum samples going back 10 years, and discovered the elephant had become antibody positive in 1996. Later, they found that all 26 elephants that had been culture confirmed as infected tested antibody positive for M. tuberculosis.

The novelty in the new paper is that instead of conducting serological tests in elephants that were known, via culture, to be infected with M. tuberculosis, the researchers began conducting serology in elephants that were not known to be infected, deflecting criticisms that interpretations of antibody test results might be biased by the knowledge that the elephants were infected, says Lyashchenko.

That is important, because the earlier research and development leading to approval of the Chembio antibody assays by the United States Department of Agriculture (USDA) in 2007, resulted in USDA's implementing a requirement by the USDA two years ago that all ~450 elephants in the US be tested annually using the Chembio serology. When elephants are diagnosed with tuberculosis, and confirmed by culture, their owners have a choice of treating, quarantining, or euthanising the elephants. Treatment takes around a year, and is very expensive, partly because such large animals require proportionately large quantities of drugs.

The incidence of TB appears to be higher in circus elephants than in zoos, says Lyashchenko. He notes that TB has more opportunity to spread among circus elephants than in zoo or sanctuary elephants, because they travel together more frequently and are both kept in closer proximity to, and exposed to greater numbers of people and other animals.

Positive cultures for M. tuberculosis are difficult to obtain because bacterial shedding occurs only intermittently, so that "trunk wash" samples, from which cultures are grown, must be taken repeatedly before success is achieved. Further slowing diagnosis, elephants can be infected with M. tuberculosis for a number of years before showing signs of disease.

"Very interesting and potentially right up my alley," says Suzan Murray, head veterinarian at the National Zoo. "We are beginning to do more field work with elephants and this could be very helpful."

(K.P. Lyashchenko, R. Greenwald, J. Esfandiari, S. Mikota, M. Miller, T. Moller, L. Vogelnest, K.P. Gairhe, HS. Robbe-Austerman, J. Gai, and W. Ray Waters, 2012. Field application of serodiagnostics to identify elephants with tuberculosis prior to case confirmation by culture. Clin Vaccine Immunol. 19:1269-1275.)

Chembio's tests are promising. However, it is unfortunate that the company is overextending the inference possible from its tests. For instance, the latest paper indicates that they demonstrated a high predictive value for their test, when in fact, they did not calculate this value and could not from what is essentially a selective case report. Similarly, due to the limitations of what is available, they stated test sensitivities and specificities at or near 100% in an earlier paper (Greenwald, 2009)when they were primarily looking at end-stage animals, rather than the earlier stage animals of interest. It would be more credible if Chembio were candid regarding the limitations of the data and their inferences.

The stated accuracy of humoral testing in this species has yet to be really determined. Both publications, this one, Lyashchenko, et al and another both base their sensitivities and specificities only on known infection status cases. No test of this technology has yet to be published that includes unknown status animals. I am disappointed to hear over and over again the wrong interpretation of these tests when they are used to then 'diagnose' the infection of unknown elephants. Mtb is diagnosed in elephants only when trunk culture is positive. Trunk washes do have limited sensitivity no one refutes that problem. These humoral response based tests are not without merit, but they need to be interpreted properly and real world uncertain status animals must be included in the statistics.
The most disturbing part of this technology is its misuse in USDA regulations. The authors may not have intended it but the USDA now quotes that these tests are 100% Specific and sensitive when used in field situations on animals whose infection status is unknown. There are a significant number of false positives out there and hopefully ongoing research projects looking at the entire population of elephants will show more real world results. In the meantime the government's involvement in this disease with their poor understanding of statistic and science is disturbing.

Wow, Anonymous and Zooexpert, was interesting to read your comments. I also work in the field and agree that these tests show promise but that their application in the actual 'field' with unknown status animals is necessary before the claims of highly accurate and diagnostic should be used. Presumptions about these tests have clouded the issue and allowed a lot of hype, confusion and the over blown sensationalism of the crazies at PETA and IDA to claim stat pak reactor elephants are infected and a threat to those that visit zoos and circuses. Even the state public health veterinarians don't think there is any significant threat to the public.
keep looking at the data.

Interesting. We have concerns about the accuracy of the TB serology for primates in the lab animal field, as well. Why is USDA approving these tests and why are they pushing them for TB control? Is there a financial or other link there? I am very concerned about USDA's ability to respond to outbreaks of disease if they are going to have such low standards.

Good science should be the sole basis for whatever the final decision from USDA. Why are people worried if the diagnostic sensitivity and specificity of this humoral antibody test in elephants is above 95%? No diagnostic test has 100% sen & spec, agreed. But remember we are dealing with elephants here, not cattle easily available in herds. The 100% sen figure is from their small sample size. We know almost nothing on the immunology of TB in elephants and still want compare these tests to the similar humoral antibody tests in humans, banned by WHO!! Lets not compare apples and oranges here. Please check the sen & spec of the FDA approved TB diagnostics in humans. Anything above 90%?! Still those tests are approved by statutory bodies. Why? Just because the risks of having false positives are far below the risks of having false negatives. As a field vet who struggled to diagnose tuberculosis in elephants with every damn diagnostic thing available to me, I must say the STAT-PAK indeed gives a very good indication on the infection status. For the <5% (the max probable false positives) lets not crucify the rest 95%. So again and always, let good science win.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Allowed HTML tags: <em> <strong> <cite><p><br><i><b><center><ul><li><div><html5:figure><html5:figcaption>
  • Lines and paragraphs break automatically.

More information about formatting options

CAPTCHA
Sorry, we know you're not a spambot, but they're out there