Search

In order to increase fertility in modern dairy farming, the interval between first insemination and conception should be as short as possible. Therefore, one approach is to diagnose non pregnant animals early to re-inseminate them as soon as possible. Commercial milk pregnancy assays are available to detect throphoblast derived bovine pregnancy-associated glycoproteins (bPAG) in milk. The aim of the present study was first to evaluate pre-analytical factors interfering with the correct detection of bPAG in milk. To achieve this aim, the stability of the bPAG was tested after repeated freezing and thawing cycles, as well as after a storage of seven days at room temperature and at 37°C. Secondly, the diagnostic performance of a commercially available PAG-ELISA was evaluated for pregnancy diagnosis between 28 to 60 days after artificial insemination by comparing the results with transrectal ultrasonography in a field study (n = 291 cows). After one freezing and thawing procedure the optical density (OD) increased and afterwards stayed stable. The storage of milk samples at room temperature had no effect on the OD, but after five days of storage at 37 °C the OD dropped sharply. It is therefore recommended to add an appropriate microbicidal preservative for shipment and storage of milk samples for bPAG analysis . The results of the field study showed that PAG-ELISA results and ultrasonography results agreed in 95.7% of the cases. Nine samples were tested as "false negative"and five as "false positive", however a re-check revealed that all "false positive"and "false negative" tested cows suffered from embryonic/fetal mortality. In conclusion, the pregnancy diagnostic by bPAG detection in milk samples is an accurate method to diagnose early pregnancy in lactating cows.

Late embryonic and early fetal mortality (EM/FM) might occur following a pregnancy diagnosis, decreasing the accuracy of pregnancy diagnostics in dairy cattle. Therefore, the detection of EM/FM may be useful in dairy reproductive management. Comparatively low concentrations of pregnancy-associated glycoproteins (PAGs) have been previously associated with EM/FM. The aim of the present study was to test the ability of a semiquantitative enzyme-linked immunosorbent assay for the detection of PAGs (PAG-ELISA) to either diagnose (EM/FM has already occurred) or to predict (EM/FM will occur) EM/FM. Three experiments were carried out as follows: (1) PAG-ELISA validation, (2) a retrospective verification of 141 PAG results in cases with suspected EM/FM and the identification of cut-off values for EM/FM, and (3) a field experiment, testing the proposed PAG-ELISA cut-off values defined in Experiment 2 under farm conditions by comparing transrectal ultrasound (TRU) results with PAG-ELISA results. The PAG-ELISA validation (Experiment 1) demonstrated that an optical density (OD) range of 0.4–2.8 reflected a linear relation to the PAG concentrations. The results from Experiment 2 indicated that more cows (85.7%) with PAG-OD values ranging from 0.40–0.79 showed EM/FM compared to cows (21.7%) with PAG-OD values ranging from 1.2 to 1.5. A cut-off value of 2.2 (sensitivity [SEN]=87.6%; specificity [SPE]=72.8%) was determined using ROC-curve analysis for prediction of EM/FM and a cut-off value of 1.2 for diagnosis of EM/FM (SEN=50%; SPE=94%). However, in the field experiment, EM/FM was best diagnosed using a cut-off value of 1.5 (SEN=26.7% and SPE=96.5%). Hence, the high SEN detected in Experiment 2 could not be confirmed in the field experiment indicating that a singular PAG value is not a convenient tool to diagnose EM/FM. Though, the combined use of PAG-ELISA and TRU showed remarkable results (SEN=84.2%; SPE=98.1%) by revealing a technique for diagnosis of previously occurred EM/FM and for prediction of subsequent occurring EM/FM. [ABSTRACT FROM AUTHOR]