Prevalence of paratuberculosis in cattle and control measures within the herd influence the performance of ELISA tests.

ABSTRACT

Commercial ELISA kits are widely used in the diagnosis of paratuberculosis of dairy cattle. It is critically important to understand the influences on these test results and their relation to faecal culture (FC) results in order to interpret the findings and to make decisions concerning serial testing and control measures. A total of 1021 cattle (423 FC positive, 598 FC negative) from 14 Mycobacterium avium subspecies paratuberculosis (MAP) positive herds were tested with four ELISA systems and FC simultaneously to calculate the kappa coefficients for the agreement of the different ELISA systems as well as find influencing factors. For the agreement of FC and ELISA, the kappa coefficients were low and ranged from 0.19 to 0.24, whereas, results of the different ELISA were consistently high (0.74-0.90). Agreement with FC was enhanced with the duration of control (P≤0.001) and the lactation number (P≤0.01), and reduced with within-herd prevalence (P≤0.001). There were substantial differences in the detection rate of low (15-24 per cent) and high (85-100 per cent) MAP shedders. In conclusion, the factors shown to influence test sensitivity, should be taken into account for validation and interpretation of ELISA tests. The benefit of serial ELISA testing is low.