Comparison of fluorescence polarization assay, indirect ELISA and competitive ELISA methods for diagnosis of Brucella melitensis-infection in small ruminants

Three serological tests, the fluorescence-polarization assay (FPA), the indirect ELISA (iELISA) and the competitive ELISA (cELISA), were compared in detecting Brucella melitensis antibodies in 454 serum samples from sheep (71 from animals which had aborted, 228 from animals without a reported abortion for the last 2 years, 155 samples from animals vaccinated with the REV-1 vaccine) and 251 serum samples from goats (animals with sporadic abortions and of unknown vaccination-status). The sensitivity of the three tests was 85, 93.5 and 88.5% and their specificity 96, 94.5 and 98.5%, for the FPA, iELISA and cELISA, respectively. The agreement between the three methods was 63.5% when testing samples from abortion cases, 92% when testing samples from apparently healthy sheep and 84% when testing samples from vaccinated sheep. Agreement between the three methods increased, when the possibility of true infection decreased; it was higher when the samples were likely to be from brucella-free animals. It appears that iELISA would be valuable for use at the early stage of a control program, whilst cELISA at its final stage. The FPA did not appear to be as accurate as the other two tests, but it was the easier, cheaper and quicker to perform. © 2003 Elsevier B.V. All rights reserved