Johne's Testing Strategies
The following 8 points summarize basic questions you should consider when deciding what type of herd testing approach is appropriate for a farm's situation.
What is the GOAL of your herd testing strategy?
- The answer to what testing should we do begins with the question of what do you want it to help you accomplish?
- Whether you want to take a peek, work toward control, assess the whole herd at once, get close to eliminating Johne’s as soon as is feasible, your goal will influence what test and how much testing you want or need to do, and how much you are willing to invest to get to your goal.
- Make it a point to consider possible ethical or legal implications for an operation if a diagnosis is made by fecal culture or PCR. These are definitive tests and confirms that there is knowledge of Johne's disease in the herd.
- It wise in the long run to investigate and control Johne's as early as possible. However, it is also wise to acknowledge this may pose short-term negative affects on merchandizing. It is best to have a plan to minimize this impact before proceeding.
What is the PREVALENCE in the herd ? How does it influence the predictive value of a test result?
An estimate of the prevalence of infection in the herd is essential information.
- It influences the approach you will take: goals, how aggressively to pursue control, what is needed in the control plan, time and resources required, etc.
- It influences how you interpret test results in the bovines just tested. Knowledge you already have about the probability a bovine is infected in a particular situation should be used to help you better estimate the likelihood that a positive or negative result in the tested bovine(s) in that situation accurately reflects true status or not (is correct or "false").
When tests are imperfect at categorizing infection status, such as Johne’s serology (false positives and negatives are possible) and fecal culture/PCR (false negatives are possible), you must use the other knowledge you have about the likelihood of infection to help fill in the inherent gaps in test information.
- Johne’s tests are not perfect in describing individuals’ infection status. One serology or culture test in known infected animals can in average terms be expected to miss over 60% of them; more false negative than true positive results are expected to occur (assume Se range of 25-40%). For Johne’s ELISA serology, testing known uninfected animals is expected to find 2% of them as false positive (Sp in range of 98%, 98% of uninfected cattle test truly negative).
- Se and Sp describe what the tests are expected to do in animals of known infection status. However, you must apply this information to situations made up of mixtures of infected and non-infected animals. Other existing information about the probability of infection in mixed infection groups is critical to develop a better estimate of the chance that a negative or positive result in an individual in the group accurately reflects their true infection status. What is the chance the animal’s result is a true or a false positive? A true or false negative?
Our best estimate of an individual’s Johne’s infection status is determined by combining information:
- What we already know about the probability of Johne’s infection in the group or individual (history, risk assessment),
- what we know about the accuracy of the test in known infected (Sensitivity)and non-infected (Specificity) animals, and
- the result itself. We do this intuitively, but this thinking process can also be described by some simple calculations.
- A prevalence estimate is formed by summarizing existing risk information about the herd (or individuals) to arrive at an estimate of the number of infected animals in the herd, which is the same as the probability that one individual in that herd is infected. This is referred to as a 'pre-test probability'.
Pre-test probabilities can be generated using the following:
- Johne’s disease history in the herd & individuals
- Risk of introduction and spread (management risk factors) in the herd
- Test data
- Clinical assessment of individuals
Estimates as crude as low, moderate, or high prevalence are very useful in interpreting subsequent test results. Clinical judgement, risk information, and test data are used to generate:
- The 'predictive value' of a result is an estimate of the chance that a positive or negative result in a particular animal is correct, or not. It is simply a function of the pre-test probability (the existing chance of infection in the herd or an individual) and the accuracy of the test used to detect infected and non-infected bovines (the Sensitivity and Specificity of the test).
- The predictive value of a result in a bovine in a client's herd can be calculated by plugging in the assumptions about the pre-test probability and the Se and Sp of the test used. It is not necessary to calculate these values per se but to properly interpret and make decisions on test results it is important to understand that the degree of infection expected in a situation influences the likelihood a test result you have in hand is correct.
Testing issues and interpreting results for Johne’s in individuals.
When the PREVALENCE is high, low, or unknown, it influences the predictive value of the result.
- Among known infected cattle, >60% will test falsely negative on fecal culture or serology.
- Among known non-infected cattle, ~ 2% will test falsely positive on serology.
- In herds, which may contain a greater or lesser proportion of either, the chance that a positive or negative test result in any particular individual is correct (predictive value) is a function of the accuracy of the test and the number of infected and non-infected animals in the herd (or the chance a specific individual has Johne’s disease based on other information).
General scenarios to be aware of:
- The predictive value of a positive serology test result is good in cattle from herds with known higher prevalence infection and very poor in individuals from herds with very low infection
- The predictive value of a negative test result is poor in cattle from herds with significant higher infection and very good in individuals from herds with very low infection.
- The predictive value of a positive or negative result is difficult to assess if there is no prior information about the probability of infection.
Test issues when the prevalence or pre-test probability of infection is estimated to be low:
- A random subsample of (30) adult animals is useful to help define a risk status for the herd. If 30 are test negative, estimate is that infection is 10% or lower; a positive estimates at least 10%.
- More "sensitive" testing (i.e. fecal culture/PCR) may be indicated to detect low level herd infection if infected individuals are likely in earlier stages.
- Repeat negative herd tests increase the assurance that a herd is not infected.
A "positive" serology result has a low predictive value because it has a high chance of being a false positive (incorrect).
- Because…there are few infected animals to test truly positive, relative to the 2% of non-infected animals expected to test false positive
- Estimated 15% chance that a positive ELISA result (S/P=0.25) is correct
- Estimated 99% chance that a negative ELISA result is correct
A "positive" fecal culture result has a high predictive value because it always has near to a 100% chance of being correct.
- Properly performed, the specificity of culture close to 100% meaning it is only positive if M. paratuberculosis is truly detected.
- A "negative" serology or fecal culture result has a high predictive value because there is a high chance it is a true negative result.
- Because…the estimate is that most animals in the herd are in fact not-infected.
Test issues when the prevalence or pre-test probability of infection is estimated to be higher:
- Higher prevalence makes control more difficult.
A large proportion of the highest risk animals have elevated ELISA values:
- estimated 88% chance that a positive ELISA result (S/P ratio = 0.25) is correct
- estimated 80% chance that a negative ELISA result is correct.
A very small proportion of lower risk animals have elevated ELISA values:
- estimated 2% of shedders that are Few on culture have positive ELISA’s
A "negative" serology result has a low predictive value because it has a high chance of being a false negative (incorrect).
- Because…there are many infected animals in the herd, but only 25-30% have significantly elevated ELISA’s relative to > 60% that do not have significantly elevated antibody interpretable as "positive."
- A "negative" fecal culture has a low predictive value for the same reason as serology – there is a high chance it is a false negative (incorrect).
- A "positive" fecal culture again has high predictive value because it is assumed to be 100% specific and correct.
- Caution is warranted in severely infected herds – the possibility that very low shedders may reflect exposure from high environmental contamination should be considered.
- A "positive" serology result has a high predictive value because there is a high chance it is indicating true infection.
- Because….it is estimated that there are many more infected animals in the herd that will test true positive relative to the 2% of non-infected animals expected to be false positive.
Test interpretation when the prevalence or pre-test probability of infection is unknown:
If you have no information about the source herd prevalence, or risk of infection in an individual, there is no pre-test probability context to assist you in interpreting the likelihood that:
- An elevated serologic result reflects true infection or is one of 2% false positive results in a non-infected animal.
- A negative serology or fecal culture result indicates a non-infected animal or is one of the >60% of infected animals that test negative.
Your option is to recognize the range of possible prevalences and predictive values. For example:
- If the probability of infection in cattle on the open market is estimated to be 20%, the predictive value of a positive ELISA is in the range of 80%. The predictive value of a negative culture or ELISA is in the range of 85%.
- If the estimate is 10%, the predictive value of a positive ELISA is 60%; of a negative culture or ELISA is 90%.
TIME: In what time frame do you want to implement your plan and reach your goals?
- Shorter time requires more aggressive management, testing, culling, at higher cost.
- Longer time complements the less aggressive plan and is needed with higher prevalence of infection.
- Without major destocking, control cannot go faster than the biologic progression of infection and the time necessary for infected animals to cycle through the herd and be replaced by non-infected animals.
Johne’s Tests Available Through Cornell Diagnostic Lab
|Johne's Direct Fecal PCR||
Detects M. paratuberculosis shedding
Sensitivity for detection of heavy and moderate shedders is 98%
Sensitivity for detection of light shedders is 100%
Specificity=100% (no false positives)
Sensitivity is poor in immature animals
Final results take 7-10 days
This is the default test for individual Bovine samples.
Results are reported as:
Moderate and Many = advanced stage of infection and high risk for spreading MAP
Can be used in a herd control testing program and as a diagnostic tool for individual cows with clinical signs of Johne's disease.
|Johne’s Pooled Fecal Culture||
Pools 5 individual samples as the lab
Slightly lower herd level sensitivity than with individual culture screening
Better sensitivity than ELISA screening
Results may take 49-60 days
For positive pools – the five individual cows from that pool are tested using the direct fecal PCR at the additional costs stated above
Refer to Johne’s Pooled Fecal Culture Testing
Recommended in screening in lower prevalence herds (with less than 2% of clinical cases)
~7% of positive pools did not have a positive individual result in follow-up. This is due to the tendency of MAP to clump.
In higher prevalence herds, it may be more economical to utilize the ELISA test with fecal PCR follow-up.
|Johne’s Fecal Culture||
Detects M. paratuberculosis shedding
Specificity=100% (no false positives)
Sensitivity is poor in immature animals i.e. estimate 15% or less @ 12 months
Final results take 2-4 months
Total *CFU counts per .1 gm:
Moderate and Many = advanced stage of infection and high risk for spreading M. pt
TNTC= >300 and Too Numerous to Count
*CFU=colony forming unit (clumped M.ptb)
|Johne’s Commercial ELISA||
Detects M. paratuberculosis antibody
Sensitivity increases with stages of infection
If a "single" cut-off interpretation of the test was used to determine positive vs. negative, it would be in the range of 80-90.
2% of non-infected animals would be expected to have a positive ELISA, i.e. 2 of 100 cows in a non-infected herd will have a value >80.
There is a high chance that a "positive" elevated ELISA result, obtained on an individual animal from a low prevalence herd, is a "false" positive.
Recommended as a herd level test, not for individual diagnosis of Johne's disease
Sensitivity is very poor in immature animals
Results take approx. 1 week
Reported as Positive or Negative Individual animal results plus the profile of how values distribute in the herd or group are provided.
Herd prevalence influences interpretation:
Recommend follow fecal PCR on ELISA positive animals
|Johne’s Milk ELISA||
Detects M. paratuberculosis antibody – at later stages (III-IV) of infection
Sensitivity is very poor in immature animals (<18 months of age)
Results in 2-5 days
A herd level test
Not an individual cow level test
Can potentially miss heavy shedders due to low sensitivity
Similar results to the Commercial ELISA
Can be influenced by stage of lactation. May be more accurate early or late in lactation.
Johne's ELISA and fecal PCR are useful in combination
Because these two tests measure different things, blood antibody and fecal shedding of M. ptb bacteria, the Cornell Diagnostic Lab recommendation has been to use fecal PCR follow- up of animals with ELISA values elevated above background. This is useful for three reasons:
- To establish a more definitive status in the herd by using a more definitive test on individuals identified at greater risk of infection based on elevated antibody.
- To determine that there is good (comfortable) correlation between the ELISA and fecal PCR on high risk animals in the herd or in a "Targeted Testing" group.
- ELISA is inexpensive and can detect animals at a higher risk of advanced infection. Fecal PCR is used to further assess shedding of M. ptb in those individuals. Definitive evidence of shedding helps to prioritize which animals are the best candidates for culling or other Johne’s control decisions.
Cornell DL recommendations for herd level testing
Since the inception of the Johne's program we have recommended fecal culture of animals with a positive ELISA as a cost- effective strategy for herd, "Targeted Testing," or Herd Status testing for most farms. Two test results on animals detected to likely be at higher risk for infection provides more information for control decisions or to increase the chance that herds are low risk with no or a low prevalence of Johne's. We now have additional testing methods which allow for pooling of samples and the fecal PCR test which replaces the fecal culture test to reduce costs and time needed for test results to be reported.
The "20% Rule"
We slightly modified our original strategy to the "20% Rule" for herd testing:
- Selected animals with positive ELISA should be tested using fecal PCR
- Fecal PCR up to a maximum of 20% of animals tested by ELISA
- The number of animals with positive ELISA results will be less than 20% in most low or moderately infected herds,
This uses fecal PCR more conservatively in order to:
- Reduce costs compared to using individual PCR testing
- Determines correlation between ELISA test and reliability for finding Johne's positives for that farm.
- Still provide good test information that adequately supports most Johne's control goals and plans
Herd veterinarians may contact the DL directly if a farm has control goals and a farm plan that may require other strategies.
How to reduce the risk of Johne’s when adding animals
- Johne's status of the source herd provides the most reliable information for estimating the infection status of an individual, particularly with regard to true low risk status.
- Today there is a high probability that you will buy the infection in: Have a management plan in place in the home herd that will prevent its spread.
Before purchasing animals:
- Know the risk within the home herd for Johne’s and other diseases of importance.
- Consider the risk from the possible source.
- Decide on what is an acceptable level of risk for the several other possible diseases one might bring in: BVD; Strep ag, Staph, Mycoplasma mastitis; Salmonella; Neospora; Digital dermatitis; Cryptosporidia, etc.
- If Johne's is a priority, consider Johne’s serology of the source herd - at least 30 mature animals as is recommended in the National Voluntary Johne’s Herd Status Program.
Johne's tests in individual animals have relatively low ability to detect earlier stages of infection, even in mature cattle. This is due to the relatively low sensitivity (25%-40% range) of serology and fecal tests. Thus, there is a 60-75% chance that an infected animal will test negative on a single test.
The utility of an individual Johne’s test in immature animals is even more limited. Less than 5%-10% of infected animals may be detected because most are in stages of infection that are too early to detect.
- Serology in immature cattle adds some minimal additional information if animals are from the same herd source and or ELISA values are very high. They are expected to be negative. If they are not, infection or cross reactivity may be present.
- Fecal culture or PCR in immature animals is not an efficient or convenient pre- purchase test.
- It has use in extra precaution situations, as the first part of a repeated testing regimen.
Pre- and post-testing of acquired cattle does provide some information about an individual’s risk of infection:
- Serology provides some information in mature cattle pre-purchase, although knowledge of the source as low risk is more valuable. Low risk sources are not readily available – the added value of cattle documented as low risk for Johne's, or other diseases, is not yet high enough to encourage producers to establish low risk status.
- Repeated negative fecal culture or serology over time indicates a low risk that an animal is propagating infection in the herd. Actual number and frequency of testing will be a function of the status of the herd and Johne's goals of the owner.
- If the status of the source herd is not low risk, repeated testing over time with negative results is the only way to increase assurance that an individual animal is not infected, or to eventually detect that an individual animal is infected.
Johne's Additional Strategies
In order of decreasing risk of introducing Johne's infection:
- No risk assessment
- Pre-buy ELISA testing, with negative "background" results
- Post-buy ELISA + Fecal Culture 2-3X @ 3-10 mos intervals, with continued negative results
- Low risk herd history of source, statement by owner/vet
- Low risk evidence based on 30 mature animals tested, with negative results on ELISA and or fecal culture
- Low risk herd history, plus 30 mature animals tested negative, plus critical management practices in place on source farm
- Low risk herd history, herd tested low risk on testing - even if positive has been discovered, plus critical management practices in place on source farm
- Low risk source history, and herd status test negative (very high assurance if multiple negative tests)
Test strategy checklist – Define these points before you test
What is the goal of testing?
- Begin to assess infection or herd status
- Estimate herd prevalence
- Control the spread of infection
- Eliminate Johne’s infection
- Establish negative or low risk herd status
- What test will be used?
Who will you test?
- Target risk groups
- Management groups
- Majority or all of the herd
When will animals be tested – what is the best timing strategy?
- Coordinate testing timing so current results are available to make control decisions
- Figure in the extra time if fecal culture is used.
- Consider the advantage and or disadvantage of testing the herd all at once: Herd snap shot at one time – establish prevalence and patterns
- Identify and prioritize animals if want to cull ASAP Results become old after 6 months
- Consider the advantage and or disadvantage of testing by groups: Over time the whole herd is tested i.e. 1 year
- Results can be timed to be more current at decision points. Regular attention is paid to results and control
Cost and effort is distributed over time; more logistics but easier Can estimate prevalence over time.
- Monthly, quarterly, semi-annually, annually, bi-annually, etc
- Pregnancy check(s)
- Mid to late gestation/lactation
- Before dry off
- Before calving (before 4 weeks from due date)
- Before putting on pasture
- Before breeding
- Other strategic/convenient management time
What culling / management decisions will be made based on test results?
- High-risk positive animals - pregnant and not pregnant?
- Low-risk positive animals - pregnant and not pregnant?
- Test negative, lower risk animals?
- Animals with mixed test results?
- What other health and performance criteria will be used with the Johne's test results and to make what decisions?