Comparison of conventional immunoassays and the oestrogen radioreceptor assay for screening for the presence of oestrogenic anabolic compounds in urine samples

Screening for the presence of anabolic growth promoters in urine samples from cattle grown for meat production can be performed by (semi)quantitative methods such as immuno-, receptor- or cell-based assays or by quantitative methods with mass spectrometric detection which can also include confirmation of compounds. In this study conventional immunoassays used at two different institutes [Veterinary Sciences Division (VSD) in Northern Ireland and TNO Nutrition and Food Research Institute (TNO) in The Netherlands] were compared with the oestrogen radioreceptor assay (ORRA), with GC-MS as the reference method. Urine samples were generated by treating calves (n = 2 per group) intramuscularly with ethynyloestradiol (EE2), diethylstilbestrol (DES) or alpha-zearalanol (zeranol, ZER). Urine samples were collected up to 21 d after administration of the oestrogenic compounds. Samples were screened by enzyme immunoassay or radioimmunoassay and by the ORRA and also by GC-MS. Values found by VSD were lower by a factor of 1-20 than those measured by TNO. These differences could be explained by differences in sample clean-up (immunoaffinity chromatography versus solid-phase extraction) and by differences in cross-reactivities between the antisera used. The ORRA and GC-MS showed similar results for EE2 and DES, but produced lower results (by a factor of ca. 3) for ZER owing to the relatively low affinity of ZER for the oestrogen receptor. The most important finding was that the withdrawal period for calves treated with EE2, DES or ZER was similar for each of the screening methods used. Therefore, it is concluded that the choice of screening method does not affect the probability of finding a positive sample.