Developmental aberrations following transfer of in vitro-produced bovine embryos can result in early gestational losses and offspring abnormalities. An ongoing study tests the hypothesis that such aberrations occur with equal frequency among commonly employed culture systems. In year 1, embryos were produced using oocytes from abattoir-derived ovaries (breed unspecified) and a proven Angus bull selected for low birth weight. IVC treatments were 2×2 factorial for medium (KSOMaa or SOFaa) and oxygen concentration (5% or 20%). Angus recipients (n=61; 32 cows, 29 heifers) were randomly allotted to treatments for Day 7 transfers. Pregnancy was diagnosed with ultrasound several times during gestation (Table 1). At parturition calf weight, shoulder height, chest circumference, crown-rump length, and humeral and femoral length data were collected. Statistical analyses (Statistical Analysis System, Cary, NC) were logistic regression with a binomial distribution for pregnancy rate, and the general linear models procedure for calf measurements; included were fixed effects of medium, oxygen, and their interaction, with additional fixed effects of dam parity and calf sex where appropriate. No significant effects of medium or oxygen were found for pregnancy rate or calf measurements other than birth weight. Mean birth weight was higher in the KSOM, 20% oxygen treatment (Table 1), and medium-oxygen interaction for calf weight was also significant (P<0.01). In year 2 embryos were produced using the same Angus bull and Angus oocytes. Angus recipients (n=38; 32 cows, 6 heifers) were randomly allotted to treatments. Fetal crown-rump lengths were measured by ultrasound weekly from Days 33 to 54 and were analyzed as repeated measures using the mixed procedure. Pregnancy outcome and LS means for crown-rump lengths are included in Table 1. Though insufficient recipient numbers preclude determination of statistical significance, of interest is the relatively small fetal size in early gestation and large birth weights in the KSOM, 20% oxygen treatment. This treatment also contained a Day 33 pregnancy, subsequently lost by Day 40, in which the fetus was too small to obtain an accurate measurement. Fetal growth will continue to be monitored throughout gestation. Data will be collected at parturition as in year 1, and pooled analyses will be done.
Table 1
Improved estimation of fixed effects panel data partially linear models with heteroscedastic errors