Removal of lipid from serum increases coherence between brucellosis rapid agglutination tests and ELISA in Alaskan grizzly bears (Ursus arctos horribilis) and Kodiak brown bears (Ursus arctos middendorffi)

In acute Brucella spp. infections, a positive rose bengal plate test (RBPT) result appears usually within 10-15 days, followed by a positive enzyme-linked immunosorbent assay (ELISA) result within a week, whereas in chronic... [ view full abstract ]

In acute Brucella spp. infections, a positive rose bengal plate test (RBPT) result appears usually within 10-15 days, followed by a positive enzyme-linked immunosorbent assay (ELISA) result within a week, whereas in chronic infections (>1 month) both RBPT and ELISA are positive for prolonged period of times, as demonstrated for mice and livestock species during experimental infections. In case of enzootic Brucella spp. infections (almost always the case for wildlife), there should be a good agreement between RBPT and ELISA results as it is likely that animals have experienced an infection for a month or more. The absence of agreement between tests results would suggest either that serological results are not compatible with brucellosis or that a technical problem occurred while performing tests. To highlight a technical pitfall when using brucellosis serological tests in wildlife species with high lipid content in sera by comparing RBPT and ELISA results before and after clean-up of the sera and assessing coherence between RBPT and ELISA results. We compared RBPT and ELISA results before and after a 30-minute chloroform/centrifugation clean-up of the sera, in Alaskan grizzly bears (n=64, Ursus arctos horribilis) and Kodiak brown bears (n=8, Ursus arctos middendorffi). The ELISA identified 42 positives (n=42/72, 58%) and the RBPT identified 53 (n=53/72, 73%). However, the coherence measured by pairwise coherence (Cohen´s Kappa, Κ) between tests was only fair (Κ=0,37). Following a 30-minute chloroform/centrifugation clean-up the ELISA identified 43 positives (n=43/72, 59%) and the RBPT identified 47 (n=47/72, 65%), and the coherence showed almost perfect agreement (Κ=0, 87). We consequently recommend cleaning up wildlife sera with high lipid content before performing RBPT and to perform RBPT and ELISA in parallel to assess coherence. Our results suggest that Alaskan grizzly and Kodiak brown bears have been exposed to Brucella spp.