Validation of an indirect immunoenzyme assay for the detection of antibodies against Trypanosoma evansi in horses in Argentina

An indirect enzyme-linked immunosorbent assay (ELISA) to detect antibodies against Trypanosoma evansi was evaluated using 90 different sera, obtained from naturally-infected horses. As negative controls, 218 sera from the T. evansi-free zone of Argentina, and 90 uninfected sera from the enzootic zone were used. The results of the ELISA were expressed in terms of percent positivity (PP) when compared with a positive primary reference serum, obtained from a horse experimentally-infected with T. evansi. The inter-assay coefficient of variation (CV), expressed as PP, was 44.7% for the negative control serum, 8.8% for the mildly positive reference serum, and 9.2% for the secondary positive control serum, while the intra-assay CV for each of the above sera was 6%, 2.8% and 5%, respectively. Positive and negative serological results were differentiated using a histogram of the distribution of the results obtained using sera from infected and uninfected animals from the enzootic zone (expressed in PP). A PP of 50 indicated a sensitivity of 95.5% for a confidence interval (CI) of 91.3% to 99.7%, and a specificity of 98% for a CI between 95% and 100%. Positive and negative predictive values were established for each rate of prevalence between 0.01% and 25%. The use of reference control sera in each assay enabled reproducible results to be obtained. The author recommends that this methodology be used whenever certification of the T. evansi status of horses is required, and particularly when animals are to be moved from an infected to a disease-free area.