The complement fixation test is used worldwide in the confirmatory diagnosis of bovine brucellosis. For this technique the antigen is the same as the one used in the tube agglutination test. However, literature is poor in information about the stability of the whole cell Brucella antigen for use in the complement fixation test to establish a time of validity of the antigen. Hence the aim of this investigation was to evaluate the stability of this antigen under refrigeration for use in the complement fixation test. Fourteen batches of antigen prepared with Brucella abortus strain 1119/3, produced from 9 months to 23 years and 11 months before, were analysed. One hundred and sixty-seven cattle sera with varying titres of antibodies to Brucella were tested through the warm complement fixation microtechnique with five 50% haemolytic units of complement. Sera with at least 25% of complement fixation in dilution 1:4 were considered positive. The results with 13 of the antigen batches were compared with the results obtained with the batch produced 9 months before by the McNemar chi2 test and kappa statistic. The oldest antigen batch gave a higher proportion of sera titres which were exactly the same observed with the 9-month-batch (90.4%), and the antigen produced 4 years and 3 months before the test gave de lowest proportion of sera with the same titre of the 9-month-antigen (73.7%). The comparison of the results after being classified as positive and negative showed that the highest proportion of agreed results was observed with the antigen produced 21 years and 4 months before (98.8%, kappa 0.98). The antigen with the lowest proportion of agreed results was the one produced 3 years and 2 months before (91.6%, kappa 0.84). The results of the study show that most sera gave very similar results with all antigen batches evaluated, and that there was no relationship between the period of antigen production and the difference in test results.
Bovine brucellosis; serological diagnosis; complement fixation test