Evaluation of the Brucella abortus species-specific polymerase chain reaction assay, an improved version of the Brucella AMOS polymerase chain reaction assay for cattle.

In a blind test, 344 samples representing 80 bacterial isolates were analyzed by the Brucella abortus species-specific polymerase chain reaction (BaSS PCR) assay for the identification and discrimination of B. abortus field strains (wild-type biovars 1, 2, and 4) from 1) B. abortus vaccine strains, 2) other Brucella species, and 3) non-Brucella bacteria. Identical samples were tested in 2 laboratories. Half the samples were fully viable, and half were bacteria that had been killed by methanol fixation. The results in 1 laboratory correctly identified 100% of the samples, resulting in a predictive value of 100% for all categories and 100% sensitivity and specificity under the prescribed conditions. The second laboratory misidentified 31 samples, resulting in a range of 66.7-100% sensitivity, 93.2-99.7% specificity, and 77.3-98.2% predictive values depending on the category. There was no significant difference in viable versus fixed bacteria for either laboratory. Subsequent review of the protocol indicated that contamination was the likely cause of 26 of the 31 erroneous identifications. The results show that the BaSS PCR assay has the potential to be a very reliable screening tool for B. abortus identification. However, the data also provide a cautionary reminder of the importance of preventing contamination in diagnostic PCR.