18 Oct 2009, 6:52 PM
How reliable and workable is the existing intradermal tuberculin (skin) test (currently the only approved test under EU regulations)? This test has been used for over 50 years. It is claimed to be the most reliable test for live cattle and many countries claim it has led to the control and even eradication of bTB in some countries (although cases still crop up with surprising regularity). However, in the UK more and more skin tested cattle are being identified as reactors and inconclusive reactors and despite regular testing for many decades hot spots remain throughout the country, where it is alleged bovine TB is endemic. It is not a totally reliable test. Field studies and scientific data regarding its reliability are minimal. As a test, it is very time consuming (involving two visits for each test), and requires all cattle to be handled twice each time, so it can have significant health and safety risks for those involved. If reactors or inconclusive reactors are found and testing is every 60 days with subsequent movement restrictions, it can cause hardship for farmers and animal welfare issues. It has proved to be unreliable for species such as alpacas (www.farminguk.com/news/TB-in-Alpacas---My-Story-So-Far_17047.html). This is disturbing as the Welsh Assembly has recently submitted proposals to make TB testing for other species compulsory.
The infamous gamma interferon blood test has even more question marks against it. This blood test evaluates the same immune response that the skin test does but is conducted in the laboratory using blood samples. Back in 2006 DEFRA began a pilot study involving this new blood test. Apparently they secretly selected 24 farms in three to four year testing areas, using bloods from other testing regimes, and tested for bovine TB. It is claimed that in the areas selected they would not have expected to find any evidence of bTB in the cattle. However, they did – around 7% of samples proved positive. As the blood test was not a recognised diagnostic test at that time, they could not confirm their findings with slaughter. The pilot study highlighted problems associated with issues such as the differing results for length of time in transit of the blood samples, possible co-contamination with recent skin tests, and several other known contaminants, including skin granulomas, the presence of antibodies or vaccinates to Johnnes disease (M. avium paratuberculosis) and other diseases. So, despite the fact that no post mortems were carried out to confirm their results, DEFRA presented the results to the Standing Veterinary Committee of the EU for approval as a secondary diagnostic test for bTB and approval was obtained.
As the blood test is an ancillary test, one would assume that it would need to be backed up with a positive skin test. However, this is not the case. If an animal fails the blood test it must be slaughtered and this was confirmed when a test case appeal was won by DEFRA.
The blood test was sold to an unsuspecting industry as ‘flexible in interpretation’. This has not been the case and no allowances are made for other known contaminants. It was also described as picking up ‘very early cases’. This too is somewhat of an exaggeration. The difference between the latency of the intra-dermal skin test and gamma IFN is about two weeks, with the skin test averaging 42 days and gamma 28 in experimentally infected animals. The reliability of the blood test is highly suspect. It was originally designed to be just a quick surveillance test, to be confirmed by the skin test. DEFRA apparently even admitted on its website that the blood test is cruder and less ‘specific’ than the skin test, in that it picks up the presence of diseases other than bovine TB, most not harmful to cattle at all. When the blood test is used over 80% of cattle slaughtered have no evidence of bTB.
Concern has been widely reported concerning the sensitivity and specificity of both the tuberculin skin test and the blood test. It is accepted that none of the tests currently available for the diagnosis of bovine TB allow a perfectly accurate determination of the M bovis infection status of cattle. The skin test does not mean that all identified reactors and inconclusive reactors actually have TB and can pass it on, yet all cattle identified by the skin test as being reactors or second time inconclusives are compulsorily slaughtered and owners are compensated from public funds. A reaction to the skin test indicates that the animal has mounted an immune response capable of recognising M. bovis (ref. 1) and does not conclude actual infection. It means that animals may have been exposed to a bacterium that can cause TB at some point in their life (or have been vaccinated!). Some animals display natural increased resistance to mycobacteria infections. The development of disease depends on the ability of M bovis to multiply. In the majority of cases natural and acquired immunity are often successful in limiting proliferation of tubercle bacilli so the disease will not progress. Even where lesions develop these can heal naturally without treatment, or may remain dormant and flare up in the future if the immune system is compromised. It is not actually known how many of the cattle currently slaughtered under the programme would go on to develop any clinical symptoms of TB, or if such cattle are really that infectious. How many of the cattle slaughtered would have developed a resistance to the disease as a result of being exposed to bacteria and could such resistance have been passed on to future offspring, thereby encouraging natural immunity in the herd? Why is there not more research into this? If the risk of TB infection is much less than currently claimed by officials, then could there be far more flexibility regarding testing and slaughter, with consequentially significant cost savings and other benefits to the farming community and other cattle owners?
The specificity and sensitivity of a diagnostic test should be determined in naturally sensitised M bovis-infected cattle and cattle in M bovis-free herds in the geographical region in which the test will be used. Has this been done for the different regions of the UK? This is important because non-specific sensitisation caused by other organisms that share antigenic determinants may vary for different regions and countries. For example, it is possible that cattle not infected with M bovis will respond to M bovis (ie will be revealed as reactors or inconclusive reactors) because of exposure to other mycobacteria. For example, are some cattle in the UK responding because of Johne's disease? This disease is closely related to the organism that causes tuberculosis. Johne's disease is rife in some areas of the UK among cattle.
The skin test used in the UK is the single intradermal comparative cervical test (SICCT). It measures hypersensitivity to tuberculin (both for M bovis and M avian) injected into the neck of cattle. The results are read after 72 hours. It compares the responses of M bovis and M avium. It only assumes that infection with M bovis promotes a larger response to M bovis tuberciulin than to M avium tuberculin and that infections with other types of mycobactrieum promote the reverse relationship. However, this assumption is only a statistical one and to some extent flawed in that the hypersensitivity response to avium tuberculin in the field equalled or exceeded the response to bovine tuberculin in 5 of 58 animals infected with bTB (Ref. 9.) This is why the the SICCT is used as a herd screening test rather than a test on individual animals!
Claims relating to the reliability of the existing skin test vary from 70 to 99%. According to DEFRA statistics (ref 2) well under half of the cattle slaughtered as TB reactors following testing are found to have evidence of TB lesions at post mortem. The figure for inconclusive reactors is even less, at around 10%. Tissue samples are taken from some of these carcasses. These are sent to TB diagnostic laboratories, where they will try to isolate M. bovis (bacteria causing TB). M bovis is confirmed in around 60% of those where lesions (and some lesions are due to other mycobacteria) had been found and just 5% of carcasses where no lesions had been found. If such large percentages have no actual evidence of bovine TB, then how can DEFRA continue to insist these animals have TB and how can slaughter and compensation payments be justified?
Is it ethical to claim that all cattle found to have TB lesions are likely to have been infectious, particularly as there is still uncertainty regarding how infection is spread? DEFRA (ref. 3) does state that reactions to the skin test can be caused by exposure to other mycobacteria. The skin test is known to give false results, and DEFRA states that the test ‘can be expected to detect approximately 80% of all the infected cattle in a herd at any one test'. DEFRA admits that the test is not always accurate. However, if no lesions are found and culture results are negative we are told (ref. 4) that ‘visible lesions of TB may not be present in the early stages of the disease’ and ‘the negative culture results do not imply that the slaughtered animals were not infected, or that they had not been exposed to the TB bacterium’. So how can DEFRA justify such statements where there is no evidence of TB other than a skin test reaction, particularly when it is clearly recognized that the test is not always accurate and these further diagnostic tests did not confirm the presence of bovine TB? Why is DEFRA’s emphasis to get as many reactors and inconclusive reactors identified as possible, with the subsequent high compensation costs for the taxpayer? The organisms can lie dormant within a host’s body for its lifetime without causing progressive disease or infection (ref. 5) so does the presence of any TB lesions in carcasses really prove active infection and a risk of spreading bovine TB? How does DEFRA justify that there is only ‘a 1 in 1,000 chance that a non-infected animal will be wrongly classified as a reactor’ and how reliable is this assumption? When asked to back up these claims using proper scientific data DEFRA and Animal Health fail to respond. This is somewhat disturbing as these are such fundamental questions.
The skin test fails to detect animals in the advanced stages of the disease (which is when they are at highest risk of spreading bTB). Advanced bTB can result in anergy - a state in which animals do not respond to the skin test because of a weak or suppressed immune response. Such old, anergic cattle may therefore shed M bovis secretions and excretions for long periods of time. Recently infected cattle do not react to the test either.
Why are there so many reactors and inconclusive reactors that show no symptoms when alive and on postmortem and following tissue culture have no evidence of any bovine TB? This applies to a significant number of cattle slaughtered, particularly those which were inconclusive, making it particularly distressing for cattle owners and expensive for the taxpayer in terms of compensation payments made for these animals. Is the test too sensitive and is it revealing reactors for reasons other than TB? Apparently nutritional problems, post calving, stress, other infections and the use of some drugs can all give false results. Some carcasses, on postmortem, have evidence of liver fluke and this is notified to cattle owners. Could the presence of liver fluke cause a positive or inconclusive reaction to the skin test? If so are we killing and compensating for something that is treatable and not a problem for most cattle which may not even show any visual symptoms and even develop some immunity to the fluke? It is interesting to note that a report on 'Bovine tuberculosis and Fasciola hepatica infection' (Ref 8) on fluke infection and TB skin testing refers to the post mortems done by Liebana in an article entitled The Pathology of Bovine Tuberculosis. They were unusually thorough in Brougham's report. In the abstract it states that the number of positive lymph nodes in skin test positive and confirmed infected animals was 1.7. His figures would seem to indicate a small likelihood of false positive skin tests in fluke infected cattle that are not regularly treated. This is because fluke infection stimulates or turns up the Th2 response, the helper T cell response, in the bovine immune effort to control fluke. This is the same response as is detected in a TB skin test, the T helper cells invading the skin where the TB antigen is deposited, causing an indurated lump.
Why are alternative testing methods not approved? Why is the PCR test on faeces not approved for cattle in selected cases? It is currently used for testing badgers. This test can detect TB in live animals, and would be particularly useful for those cattle that are not used to being handled, are in a small, closed, low risk herd and are very difficult to pen. Back in 2005 an ‘electric nose’ was proposed (ref. 6). This apparently uses a serum sample. In experiments it allegedly identified infected animals as early as 3 weeks after infection with M. bovis. Defra have been testing this system at a cost of £450,000 and did admit that the e-nose together with SIFT-MS (selective ion flow tube mass spectrometry) could offer improvements in terms of diagnostic accuracy in both cattle and badgers and could have a significant impact on the control of bovine TB. Are there other tests that may be better at detecting actual infection and which would spare animals that may have been exposed to mycobacterium at some points in their lives but do not actually pose a risk? Currently, are we merely killing off many cattle that have actually become immune to the disease naturally - a similar immunity to that which a vaccine would give?
Are there other ways of keeping TB under control? Has the government investigated less costly and more reasonable methods of controlling TB? Some USA states (ref. 7), recognising, that testing is problematic, inefficient and costly, switched from area testing to tracing of lesions animals found during the inspection of animal carcasses at slaughterhouses. These animals are then traced back to the relevant herds and only these herds are tested. Could testing be restricted to dairy herds only and pre-movement testing, particularly as it is accepted that the biggest risk of passing on disease is via cattle movements?
In October 2012 there was more evidence (Defra funded research - www.cam.ac.uk/research/news/scientists-build-a-clearer-picture-of-the-spread-of-bovine-tuberculosis/) confirming just how unreliable the skin test is - the skin test is the backbone of the existing test/cull system currently used to 'eradicate' bTB. It reveals that testing misses many animals 'harbouring the disease' and shows that large herds are particularly vulnerable to rapid transmission.
1. Veterinary TB
Chembio Diagnostic Systems, Veterinary Health Products, Veterinary TB
2. letter from Animal Health, Access to Information, ref AH0458 dated 20 May 2009
3. DEFRA booklet - Dealing with bovine TB in your herd (last updated May 2008).
4. Standard Animal Health letter ‘Laboratory Analysis of Reactors/Irs/DCs to the Tuberculin Test
– Results of Bacteriological Culture
5. Bovine Tuberculosis research paper
see page 7
House of Commons Library, Research Paper on 98/63
1/6/2008 on Bovine Tuberculosis
6. Use of Electronic Nose Test
Use of an Electronic Nose To Diagnose Mycobacterium bovis Infection in Badgers and Cattle, report 2004.
7. History of Bovine TB
Emerging Disease Issues. Diseases that may effect humans or animals. Bovine tuberculosis.
8. 'Bovine tuberculosis and Fasciola hepatica infection' by JM Brougham, P Durr, R Clifton-Hadley, A Colloff, T Goodchild,
R Sayers, S Williamson and S H Downs
9. NI Morrison, FJ Bourne, DR Cox, CA Donnelly, CJP McInerney, R Woodroffe, 'Pathogenesis and diagnosis of infections
with Mycobacteria bovis in cattle'.