Sensitivity and specificity of two enzyme-linked immunosorbent assays and a quantitative real-time polymerase chain reaction for bovine paratuberculosis testing of a large dairy herd

Serum and fecal samples from a large California herd were used to estimate the diagnostic sensitivities and specificities of two ELISA kits and a fecal quantitative real time PCR (qPCR) for paratuberculosis. Both ELISA kits were marketed by the same company but one (Mycobacterium paratuberculosis; MP) was advertised as having increased sensitivity over an older kit that is no longer available (HerdChek; HC). Because of the large amount of data accumulated using the older kit, there was concern that the transition to a new test kit would provide significant complications in following herd level seroprevalence rates for Johne's disease. Furthermore, both kits diagnostic accuracy in large herds was not known. A 3 test (2 dependent, 1 independent) no gold standard analysis of the ELISA and qPCR results from a large dairy herd in California was conducted. Results of this study showed an improvement in the new ELISA kit compared to the previous one in terms of sensitivity (34% compared to 31%, respectively) and specificity (96% compared to 94%). The fecal qPCR assay had a sensitivity of 68% and a specificity of 96%. Prevalence of paratuberculosis in the study herd was 10.5%. Estimates reported here may differ from previous studies due to the differences in source herds, reference tests used and results variability due to laboratory and over time.