Isotype-specific antibodies in horses and dogs with immune-mediated hemolytic anemia.

Classes of antibody bound to erythrocytes were determined using direct immunofluorescence (DIF) flow cytometry in 3 horses and 12 dogs with immune-mediated hemolytic anemia (IMHA). Background levels of antibody binding were determined in samples from 12 horses and 12 dogs that were free of clinical disease. The range of nonspecific binding of a fluorescein isothiocyanate (FITC)-conjugated goat anti-equine immunoglobulin G (IgG) was 19.9-36.7%, but was eliminated by the use of the F(ab')2 fragment of FITC-conjugated goat anti-equine IgG. Background binding by other class-specific antibodies to equine and canine erythrocytes was negligible. The DIF results were compared to the direct antiglobulin (Coombs') test in 5 horses and 20 dogs with anemia. The former assay was more sensitive in dogs with IMHA than was the Coombs' test (100% versus 58%). In contrast, the Coombs' test had better specificity than the DIF assay (100% versus 87.5%, respectively). Using clinical parameters or response to therapy as the comparison, the positive and negative predictive values for the DIF test were 92% and 100% compared to the values of the Coombs' test of 100% and 62%. The DIF assay detected low levels of cells bound with antibody (<30%) in 5 dogs that were Coombs' test-negative. For both species, performance of the DIF test was independent of the prozone effect. Five dogs with IMHA had IgG and IgM on erythrocytes, 5 had IgG, and 2 had IgM. Three horses had surface-bound IgG, including a horse with suspected penicillin-induced IMHA, a foal with neonatal isoerythrolysis, and a foal with clostridial septicemia. The DIF method was valuable in monitoring the response to therapy in the foal with neonatal isoerythrolysis.