The sensitivity and specificity of a modified ELISA for the diagnosis of Johne's disease from a field trial in cattle.

An enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies against Mycobacterium paratuberculosis in cattle was evaluated in three herds known to have Johne's disease. Prior to testing, the plasma was absorbed with dried Mycobacterium phlei in order to remove cross-reacting antibody specificities. The sensitivity and specificity of the ELISA were calculated after repeatedly testing 327 cattle in the infected herds. Of these, 53 animals had one or more positive faecal cultures or had post-mortem histopathological evidence of infection. The other 274 had three or more negative faecal culture results, and were regarded as non-infected for the purpose of evaluation of the ELISA. Using these criteria for the presence or absence of infection, the M. phlei-absorbed ELISA under field conditions had a 57% sensitivity and a 98.9% specificity. The sensitivity of the absorbed ELISA depended on the stage of disease of the animal under test. In general, it appeared that animals in the more advanced stages of disease were absorbed ELISA positive, whereas those in the early stages of infection were not detected. These results indicate that the M. phlei-absorbed ELISA has an important role as a test for the diagnosis and control of Johne's disease in cattle.