Estimation of the specificity of an antibody ELISA for paratuberculosis generated from a sector of the UK cattle population using results from a paratuberculosis control programme.

In the United Kingdom (UK) a voluntary programme to control paratuberculosis in cattle based on herd management and serological screening has been operating since 1998. The programme assigns a risk level to each participating herd according to the within herd seroprevalence and the confirmation of the presence of infection with Mycobacterium avium subspecies paratuberculosis (MAP) by faecal culture or polymerase chain reaction (PCR). From the outset a general concern over the specificity of the paratuberculosis antibody enzyme-linked immunosorbent assay (ELISA) resulted in the use of a faecal screen for the causal organism to negate or confirm infection in individual seropositive animals. Progress in improving the diagnostic tests has been gradual throughout the life of the programme and the under-pinning approach to using tests to determine the risk of paratuberculosis for a herd required to be re-examined. This study used a large data set of more than 143,000 test results over five years from the lowest paratuberculosis risk level category of herds to estimate the specificity of a commercially available paratuberculosis antibody ELISA for cattle. In each year of the study the estimated specificity reached or exceeded 0.998. We also examined the apparent impact that annual or more frequent application of the single intradermal comparative cervical tuberculin (SICCT) test for tuberculosis (TB), using purified protein derivatives of Mycobacterium bovis and Mycobacterium avium subspecies avium, had on specificity of the antibody ELISA for paratuberculosis. We found a statistically significant difference in three of the five years with herds that were officially tuberculosis free and not subject to frequent SICCT testing. This difference was small and considered to be of little practical importance for the paratuberculosis assurance programme. We concluded that, in the UK the mandatory TB surveillance programme of cattle herds is not a limiting factor in the use of serological testing to support herd-level assurance schemes for paratuberculosis. Furthermore, in paratuberculosis, where shedding of MAP is intermittent and the sensitivity of the commercially available PCR tests for detection MAP is highly variable, faecal screening of seropositive animals is an unreliable method for negating infection in seropositive cattle.