Pregnancy diagnosis based on pregnancy-associated glycoproteins detection in dairy cow milk: Factors influencing measurement

Milk pregnancy-associated glycoproteins (PAGs) provide a noninvasive approach to early pregnancy diagnosis in dairy cattle. We retrospectively analyzed records from 2460 cows (Holstein and Brown Swiss) retrieved from a national livestock database (January 2022–April 2024) to identify cow-level and production factors associated with milk PAGs concentrations. Cows were categorized as pregnant, recheck, or not pregnant according to milk ELISA results, applying the threshold values provided by the manufacturer; outcomes were verified from insemination and subsequent calving data. Mixed models evaluated effects of breed, gestational age, parity, and days in milk (DIM), and Pearson correlations quantified associations with milk composition. Milk PAGs testing allowed a clear and unambiguous discrimination among the three diagnostic categories. Among cows classified as pregnant, 12% did not calve subsequently, whereas none of the recheck cows calved, underscoring the need for confirmatory follow-up. Absolute PAGs concentrations increased with advancing gestation (p < 0,01) and were higher in multiparous than primiparous cows (p < 0,01); DIM showed little effect. Weak positive correlations were observed between PAGs and milk protein (p < 0,01), casein (p < 0,01), and urea (p < 0,01), whereas associations with other milk traits were negligible. Collectively, these results support milk PAGs measurement as a practical and sensitive aid to pregnancy detection while emphasizing integration with routine clinical monitoring (e.g., rectal palpation and ultrasonography). Understanding how breed, gestational stage, and parity influence milk PAGs may improve test interpretation and contribute to more efficient reproductive management and reduced losses in dairy herds.