Abstract
Commercial enzyme-linked immunosorbent assays (ELISAs) for detection of serum antibodies to bovine viral diarrhea virus (BVDV), parainfluenza-3 virus (PI3V), respiratory syncytial virus (RSV), and infectious bovine rhinotracheitis virus (IBRV) were standardized to give a quantitative result when testing was performed at a single optimum dilution. For each test, serum samples were titrated and their end point titers calculated by an algebraic method directly from a plot of each titration series and also from a regression line fitted to this plot. The corrected optical density (COD) of each sample when tested at dilutions of 1/25, 1/50, and 1/100 was expressed as a percentage of the COD of a positive reference serum included on each plate, this value was the sample/positive (S/P) ratio. For each test, the linear relationship between the S/P ratio obtained at a dilution of 1/25, 1/50, and 1/100 and the end point titer calculated by each method was determined. In each case, the best linear relationship existed when samples were tested at a dilution of 1/100 (r = 0.973 for BVDV, 0.962 for PI3V, 0.961 for RSV, 0.947 for IBRV). From the equation of these lines, an increase in the S/P ratio between acute and convalescent serum samples of 31%, 23%, 21%, and 35% would correspond to a 4-fold rise in ELISA titer to BVDV, PI3V, RSV, and IBRV, respectively. ELISA titers calculated from S/P ratios at 1/100 were significantly related to virus neutralization titers to BVDV, RSV, and IBRV and to hemagglutination inhibition titers to PI3V (P ≪ 0.001 in all cases). Samples with low S/P ratios had the greatest intraassay and interassay variation. Intraassay reproducibility ranged from 3.5% to 22.3% (coefficient of variation), with a median value of 9.5%. Interassay reproducibility was lower, ranging from 6.0% to 50.6%, with a median of 17.4%.