Evaluation of a commercially available radioimmunoassay and enzyme immunoassay for the analysis of progesterone and estradiol and the comparison of two extraction efficiency methods

The measurement of progesterone (P4) and estradiol (E2) is essential for monitoring reproductive cycles and can aid in diagnosing the cause of poor reproductive performance in dairy cattle. Readily available, reproducible, accurate, non-radioactive assays are needed for the assessment of P4 and E2 in bovine serum. The gold standard for hormone assessment, radioimmunoassay (RIA), was compared with enzyme-linked immunoassay (EIA). Serum collected from various points in the estrous cycle was extracted with radiolabeled P4 (ie, ³H-P4; HE) and without ³H-P4 (CE) before being used in the assay. For the assessment of P4, there is a great degree of correlation between the RIA and EIA (adjusted R-square = 0.95; Pearson correlation coefficient (PCC) = 0.98, P < 0.001). A difference between the RIA and EIA methods was not detected for E2 concentrations (P = 0.16), but the correlation between techniques was poor (adjusted R-squared = 0.73; PCC = 0.87, P = 0.002). There was no difference in the serum extraction efficiency as measured with ³H-P4 as opposed to without (P = 0.94). The two methods for the measurement of serum extraction efficiency were highly correlated (adjusted R-square = 0.83; PCC = 0.92, P < 0.001). The concordance correlation coefficient (CCC) showed an excellent agreement between RIA and EIA for P4 determination (0.89) and between HE and CE methods (0.90). Although the 95% limits of agreement of the Bland-Altman plots encompassed 89% (8/9) and 92% (12/13) of the differences between methods for P4 quantification and extraction respectively, the CCC indicated an excellent agreement among them. The CCC between RIA and EIA for E2 quantification was 0.68 which corresponds with a fair agreement; however, the 95% limits of agreement of the Bland-Altman plot encompassed 100% (9/9) of differences between methods. The EIA and CE methods are comparable alternatives to the RIA and HE methods, respectively and can be used to quantify P4 and E2 for bovine serum.