Practice relevant rapid tests for milk progesterone in cattle compared with a laboratory-connected routine method

Four commercially available semiquantitative milk progesterone tests (Ovucheck-Praxistest: Cambridge Veterinary Science/Smith Kline), Progestassay-Milchprogesteron (Pitman-Moore/Janssen), Reprostrip-Progesteron-Schnelltest (Noctech/Albrecht), Enzygnost-Milchprogesteron (IQ, 'Bio' UK/Hoechst Veterin?r) were examined for their accuracy by using them for the determination of progesterone levels of 64 milk samples, i.e. 1556 single assays. Several test series were performed, using codified samples and changing sequences. Three or four test persons respectively, performed the tests independently and classified the samples semiquantitatively. These test results were then compared to the results acquired by measuring the progesterone levels of the same samples by means of an approved quantitative, labor-bound progesterone test (Hormonost: Biolab). These control tests were performed at a specialized routine labor, by different personnel and at a different location. Lastly, in 48 out of the 64 sampled animals the reproductive status could be evaluated clinically and was taken into account as well. Samples yielding high progesterone levels, i.e. greater than or equal to 9 ng/ml were classified correctly in 84.4 to 96.5% of the cases, whereas samples with low levels (less than or equal to 2.5 ng/ml) were classified correctly in 68.8 to 90.0% of the cases only. Samples ranging between this spectrum (greater than 2.5 less than 9 ng/ml) were classified correctly only in 42.1 to 52.6% of the cases. However, this range appears to be of the most interest for the veterinary practitioner since cows in proestrus or early interestrus tend to have mild progesterone levels within these values. On the other hand, clinical findings are often insufficient for a proper diagnosis just in these animals.(ABSTRACT TRUNCATED AT 250 WORDS)