Small ruminant Johne’s (Mycobacterium avium subspecies paratuberculosis)

By Larell Bermudez-Koch, CS McConnel, Veterinary Medicine Extension, Melissa Holahan, Veterinary Medicine Extension, and K Poonsuk, Immunodiagnostic Section Head WADDL

As part of a larger project investigating small ruminant submissions through the Washington Animal Disease Diagnostic Laboratory (WADDL), we have focused on diagnostic trends related to Johne’s disease (Mycobacterium avium subspecies paratuberculosis; MAP). Small ruminant submissions to WADDL from May 2015 to June 2024 were evaluated for this project. Within this timeframe, 105,404 samples were submitted for Enzyme-Linked Immunosorbent Assay (ELISA), and 1,004 samples were submitted for fecal Polymerase Chain Reaction (PCR). Analysis of ELISA test results revealed an overall apparent animal prevalence of 1.0% (966/95,136) within domestic goat submissions, and 3.2% (330/10,268) within domestic sheep submissions. Prevalence based on PCR results was 10.8% (99/916) within domestic goats and 2.3% (2/88) within domestic sheep. Between May of 2015 and June of 2024, WADDL sequentially utilized two commercial MAP ELISA tests: IDEXX, and VMRD. IDEXX was utilized from May 2015 – October of 2023, and VMRD from October of 2023 – present day, because the newly developed VMRD ELISA is validated for small ruminants. A Chi-Squared test revealed differences in apparent prevalence (p < 0.0001) when comparing results from goat samples using either IDEXX (0.73%; 646/88,826) or VMRD ELISAs (5.1%; 320/6,310). Similar differences (p < 0.0001) were found when comparing results from sheep samples using either IDEXX (2.4%; 224/9,154) or VMRD ELISAs (9.5%; 106/1,114). A subpopulation of goat samples (n = 270) of unknown MAP infection status tested with both IDEXX and VMRD ELISAs showed little agreement (𝜅 = 0.15) between the two tests.

So what is the takeaway? Expect some goat herds and sheep flocks that have a history of negative IDEXX ELISA results to have positive results if tested using the VMRD ELISA. Based upon established test characteristics it seems likely this is a result of enhanced sensitivity resulting in the identification of some truly infected animals. Keep in mind, however, that even though the specificity of the VMRD ELISA is also high (>98%), prevalence of disease within a herd ultimately drives the positive predictive value (PPV) which is the probability that subjects with a positive screening test truly have the disease. A low prevalence of infection will drive down the PPV regardless of the test specificity.

That brings up the second major takeaway. Given that the ELISA tests are imperfect, diagnostic evaluation of a herd likely requires the integration of PCR (and/or MAP culture). ELISA has historically been considered more of a herd level assessment than an individual animal test. Screening a herd through the use of ELISA provides a snapshot of potential levels of infection. More refined management can be accomplished by utilizing PCR at defined periods of production (pre- or post-kidding/lambing for example) to establish the risk of infection to neonates. Given the results from our study, it appears likely that PCR has predominantly been used to test ELISA positive animals individually rather than ELISA positive herds. It may be worth considering expanding the use of PCR in herds with a positive ELISA result to investigate the breadth of shedding and contamination potentially aligned with infection at the herd level as indicated by ELISA positivity.