Comparison of indirect immunofluorescent-antibody assay, enzyme-linked immunosorbent assay, and Western immunoblot for the diagnosis of Lyme disease in dogs.

Enzyme-linked immunosorbent assay (ELISA), indirect immunofluorescent-antibody assay (IFA), and Western immunoblot were used to test serum samples from 128 dogs for the presence of antibody to Borrelia burgdorferi. Sera included 72 samples from dogs suspected of having Lyme disease, 32 samples from dogs residing in areas in which Lyme disease was not considered endemic, and 24 samples from dogs with clinical and serologic evidence of immune-mediated disease (n = 10), Rocky Mountain spotted fever (n = 5), or leptospirosis (n = 9). Results of Western immunoblotting were used as the standard against which performances of ELISA and IFA were measured. ELISA was significantly more sensitive than IFA (84.8 versus 66.7%), although both tests were equally specific (93.5%). Eight samples that were positive by Western immunoblot were simultaneously negative by ELISA and IFA. Of these eight, four were from dogs suspected of having immune-mediated disease, two were from dogs suspected of having leptospirosis, and two were from dogs suspected of having Lyme disease. These results may indicate that sera from dogs with immune-mediated disease, and to a lesser extent sera from those with leptospirosis, cross-react with B. burgdorferi antigens. Alternatively, Western immunoblot results may not truly reflect Lyme disease status, particularly in the case of dogs with immune-mediated diseases. At present, however, the use of Western immunoblotting as a diagnostic standard for dogs offers the best alternative to a clinical definition of disease.