Bayesian latent class estimation of sensitivity and specificity parameters of the PMS-PCR test for the diagnosis of cattle sub-clinically infected with Mycobacterium avium subsp. paratuberculosis.

The objective of this study was to estimate the performance of the peptide magnetic separation PCR test (PMS-PCR) for the diagnosis of Mycobacterium avium subsp. paratuberculosis (MAP) in sub-clinically infected dairy cattle. Twenty-one herds were randomly selected from a source population of 131 commercial dairy herds with a known history of MAP infection, located in the De Los Rios and De Los Lagos regions, in southern Chile. In the selected herds, all milking cows with ≥2 parities and without any clinical signs were sampled, collecting feces and blood-serum samples. The PMS-PCR test was used to analyze the fecal samples, while serum samples were analyzed using a commercial ELISA kit. A Bayesian latent class model was used to estimate the sensitivity (Se) and specificity (Sp) of the diagnostic tests. A total of 1381 animals were sampled in the 21 selected dairy herds, with an average sample size of 65 animals per herd (range 10-721). The PMS-PCR test had a greater Se than the ELISA test, with a median of 85.5 % (posterior probability interval (PPI) 95 %: 79.3-91.0%), while the ELISA test presented a median of 21.7 % (95 % PPI: 18.3-25.4%). On the other hand, the ELISA test had a better Sp than the PMS-PCR test, with a median of 97.7 % (95 % PPI: 96.6-98.5%), whereas PMS-PCR presented a median of 90.8 % (95 % PPI: 88.3-93.9%). Model results showed that PMS-PCR has a better Se than all available tests for MAP diagnosis in subclinical animals. However, this test should be used with care in herds with high infection rates, where a high MAP environmental load is expected, potentially increasing the frequency of false positive cases due to the pass-through phenomenon.