Leukosis and paratuberculosis: frequently asked questions (Proceedings)

Every year I get numerous calls about Johne's disease and bovine leukosis. Occasionally veterinarians or producers want to discuss the two of them at the same time. As much as I try to keep things separated, inevitably I have trouble keeping the listener from getting them confused. I thought it might be interesting to discuss them together in one session in order to contrast the differences. I think most practitioners have a pretty good understanding of the basics surrounding the diseases, particularly the clinical syndromes. Therefore I've decided to try to answer some of the most frequently asked questions and highlight some of the most frequently misunderstood issues concerning these diseases

How are these diseases spread in a herd, vertically and horizontally?

The classical means of transmission of M. avium paratuberculosis (MAP) is fecal-oral. Susceptibly is age-related. There is no magic age after which a calf becomes resistant to the organism; however, the number of organisms required to produce infection rises as age increases. Therefore, it is increasingly difficult for an animal to become infected as it ages. Studies have shown that about 10% of infected cows shed MAP in their milk and about 22% of infected cows shed MAP in their colostrum. Approximately 8% of infected cows without clinical signs and 25% of infected cows with clinical signs of paratuberculosis have infected fetuses.

An understanding of the means of transmission of this organism is particularly useful when advising clients about control and elimination programs.

The spread of BLV is primarily by direct contact. In fact, a separation of 2 to 3 m appears to be adequate to prevent transmission between cattle. The virus exists primarily in lymphocytes, therefore exchange of lymphocytes is typically required. The exchange of blood is the most efficient method of transmission. White blood cells are present in the body secretions so this may be the most frequent natural means of horizontal transmission. There are several proven and theoretical iatrogenic means of transmission. Tattooing, dehorning and even rectal examination with the same sleeve could spread the disease. Dehorning with a Barnes-type dehorner has been shown to be associated with increased seroconversion among a group of calves compared to no dehorning. Interestingly, in that study, calves that were dehorned with a Barnes, and then had the surgical wound cauterized, had no transmission of BLV. Rectal palpation appears to be a means of transmission of the virus both experimentally and in the field. However, there is evidence that routine rectal palpation with a single sleeve is less likely to cause transmission than intensive repeated palpation by inexperienced personnel. It seems reasonable to change sleeves between cows when a substantial amount of blood is visible on the sleeve. When no blood is visible, the risk is probably low. The use of individual needles is usually recommended as a control measure, but there is not a lot of research to support the transmission of BLV by needles when visible blood is not present. Perhaps clients' perception should play some role in our decision to use individual needles. Transmission by insects has been studied extensively. Only tabanids are considered to be a likely vector. This is not been proven in a field situation.

It is clear that BLV is shed in milk and colostrum. However it is not clear whether calves are likely to become infected by consuming BLV infected milk. It is clear, however, that colostrum and colostral antibody against BLV is protective. Despite the fact that transmission by consumption of milk is not considered a primary route of transmission, several studies have been performed to demonstrate that pasteurization inactivates BLV.

Is either organism shed in the semen? On embryos?

There is strong evidence that BLV virus is not shed in normal, properly collected semen. Also, there is strong evidence that properly washed embryos from infected cows close negligible risk of producing infected calves when placed in BLV negative recipients. For this reason, The European Union Commission has declared the risk of transmission of BLV and semen are embryos as negligible.

It is known that MAP can be isolated from the semen of infected bulls. Whether this can cause infection of the fetus or the cow is not known. It is generally accepted that frozen processed semen poses negligible risk of transmission of Johne's disease. During natural service, it would seem that the risk of infection to calves exposed to the feces of infected will would be greater than the risk of transmission by semen of that same bull. It is generally accepted that properly washed embryos from infected cows pose negligible risk of producing infected fetuses. Even intentionally contaminated embryos can be decontaminated by appropriate washing.

How long does the organism live in the environment?

The MAP organism is extremely resistant in the environment. Although it cannot multiply outside of the host, it can persist in the environment for long periods of time. It has been shown to survive for at least one year in tap water, in mud and feces, and in pond water. On the other hand, BLV is very labile in the environment thus environmental contamination is unlikely to be a significant problem.

What's the best test?

There are two serologic tests for BLV, AGID and ELISA. Both have relatively high sensitivity and specificity. Both detect antibody. Because BLV is a retrovirus, infection is considered lifelong. Therefore, cattle with positive antibody titers are considered to be infected. A PCR is available which detects viral antigen. It could be argued that it is slightly more specific. From research data it appears that the PCR will detect animals earlier in the stage of infection than will the serologic test. However the difference is only a few weeks. All of these tests are much more robust than even the best test for paratuberculosis. Colostral antibody will cause a calf to be serologically positive for over 6 months.

Briefly, the most popular serologic test for Johne's disease is the ELISA. The sensitivity is roughly 0.25 and the specificity and 0.99. Fecal culture is the most sensitive test, but it requires at least 42 days. It has a sensitivity of only slightly greater than 0.5. A direct fecal PCR is also available, but in most laboratories is less sensitive than culture. It also tends to be more expensive. For more information on the best test, see Collins MT, et al. J Am Vet Med Assoc, 2006: "Consensus recommendations on testing for bovine paratuberculosis".

How should I interpret the results?

There are actually two questions embedded in the question above. One of them can be answered by examining the positive predictive value (PPV) and the negative predictive value (NPV) which are based on sensitivity, specificity and prevalence. The other question would better be worded "What actions should the owner and veterinarian take based on the results?" The answer to this question is going to hinge upon the goals of the owner. You should have a serious conversation about the owner's goals and how you will use the test results BEFORE you begin testing. While that sounds obvious, it is not always done.

Let's begin by defining PPV and NPV. The PPV is the probability that an animal with a positive test is infected. The NPV is the probability that an animal with a negative test is uninfected. While we often talk about sensitivity and specificity, the PPV and NPV are more important values for the practicing veterinarian. The prevalence has a tremendous impact on the PPV and NPV of a test. Because the sensitivity and specificity of the ELISA for BLV are both very high and because the in-herd prevalence of the infection is usually greater than that of paratuberculosis, there are fewer false positives and false negatives for BLV then for Johne's disease. For example, in a dairy herd with a prevalence of 10% for Johne's disease, the PPV is .82 while the NPV is .94. That means that 82% of cattle with a positive test will be infected while 18% will be false positives. Also, 94% of cattle with a negative test will be uninfected, and 6% will have a false negative test. If that same herd has a 50% prevalence for BLV infection, over 99% of the cattle with a positive test will be infected and virtually all of the cattle with a negative test will be uninfected. It is obvious that the test for BLV functions much more effectively than that for Johne's disease. The magnitude of the problem presented by the relatively high false positive rate for Johne's disease depends on the owner's goals and action taken in response to a positive test. For example, if the plan is to identify test positive cows and discard their colostrum, the fact that 18% of the identified cattle are uninfected will not have a great impact. If the plan is to cull test positive animals, an 18% false positive rate is probably unacceptable. On the other hand, the accuracy of the BLV ELISA is very acceptable for making culling decisions, as fewer than one in a hundred animals will be culled unnecessarily. (NOTE: the false positive and false negative rate in this example apply only to the conditions of this example, not to all situations involving these tests.)

Table 1 shows the effect of prevalence on PPV and NPV on the ELISA for Johne's disease with a sensitivity of .4 and specificity of .99 and BLV with a sensitivity of .994 and specificity of 1.0%.

Table 1: Predictive value of Johne’s ELISA