Characterization of pregnancy-associated glycoproteins and progesterone as a predictor of twins and conceptus loss in high-risk-pregnancy Holstein cows

The objective of this study was to characterize plasma concentrations of pregnancy-associated glycoprotein (PAG) and progesterone (P4) as predictors of twins and pregnancy loss in Holstein cows with high-risk pregnancies. High-risk pregnancy was characterized using transrectal ultrasonography 37 d after artificial insemination (AI) based on the following criteria: small embryo size <15 mm, n = 10), slow heartbeat (<60 beats per minute, n = 11), and extra-amniotic membrane (additional amniotic membrane, n = 3). A cohort of twins (n = 41) diagnosed at d 37 post-AI was also enrolled. Each high-risk and twin pregnancy cow was paired with a cow of the same parity carrying a normal singleton at d 37 post-AI (control, n = 65). Blood samples were collected to measure PAG and P4 at d 37, 44, and 51 post-AI. Statistical analysis was performed using ANOVA, logistic regression, and receiver operator characteristic (ROC) curve with JMP software (SAS Institute Inc., Cary, NC). Pregnancy loss at d 51 post-AI was greater in high-risk pregnancies than in controls and twins (control = 1.5%; high-risk = 87.5%; twins = 12.2%). Concentration of PAG at d 37 post-AI did not differ among groups (control = 5.3 +/- 0.7; high-risk = 4.8 +/- 1.2; twins = 4.0 +/- 0.9 ng/mL). Concentration of P4 at d 37 post-AI was greater in twins than in high-risk pregnancies and control, and lesser in high-risk pregnancies than in controls (control = 7.0 +/- 0.3; high-risk = 5.9 +/- 0.4; twins = 8.4 +/- 0.3 ng/mL). Regression and ROC analysis for PAG at d 37 post-AI did not find a threshold to predict pregnancy loss or twins. Regression and ROC analysis for P4 at d 37 post-AI found that a threshold of 6.5 ng/mL predicted pregnancy loss with an area under the curve (AUC) of 0.64, and a threshold of 7.2 ng/mL predicted twins with AUC of 0.71. In summary, pregnancy loss and twins were predicted with only moderate accuracy by P4 concentration at d 37 post-AI, and the variability in PAG concentrations at d 37 post-AI was not sufficient to generate a threshold to predict pregnancy loss and twins in Holstein lactating cows.

Automated summary

This study aimed to predict pregnancy loss and twins in Holstein cows with high-risk pregnancies by measuring plasma concentrations of PAG and progesterone. Results indicated that progesterone concentration at day 37 post-AI had moderate accuracy in predicting pregnancy loss and twins, while the variability in PAG concentrations was not sufficient for prediction in Holstein lactating cows.