Abstract
Clinicopathologic criteria were used to group 68 cats according to high, moderate, or low suspicion of having feline leukemia virus (FeL V)-related disease. Peripheral blood samples were tested for FeL V antigen by enzyme-linked immunosorbent assay (ELISA) and for FeL V DNA by polymerase chain reaction (PCR). There was no significant difference between ELISA and PCR results in the 68 cats. In the high-suspicion group, 46% (11/24) of cytopenic cats were test positive (ELISA and PCR) and 87% (13/15) with hemopoietic neoplasms were test-positive. Also within the high suspicion group, test-positive cats were 2.5 times more likely to die within the 1 year follow-up period than were test-negative (ELISA and PCR) cats. Among cats in the moderate-suspicion group, 15% (2/13) were test-positive, and none (0/16) of the cats in the low suspicion group was test positive. The relative risk of a positive test (ELISA and PCR) in the high suspicion group was 3.7 times that for the moderate-suspicion group and 22.8 times that for the low suspicion group. There was no significant difference in the relative risk of a positive test result between the moderate and low suspicion groups. The results indicate that FeL V detection by PCR can be adapted for diagnostic purposes using peripheral blood samples, however, results do not differ significantly from FeL V ELISA results. Also, a proportion of cats with a high suspicion of having FeL V-related cytopenia and hemopoietic tumors are negative for both circulating FeL V antigen and DNA. These cats may not have FeL V-related disease, or FeL V may exist in a disease-producing but nonreplicating form ultimately detectable by PCR in tissues other than peripheral blood.