TY - JOUR
T1 - Fetal age assessment for Holstein cattle
AU - Krog, Camilla Hessel
AU - Agerholm, Jørgen Steen
AU - Nielsen, Søren Saxmose
PY - 2018/11/19
Y1 - 2018/11/19
N2 - Although transport and slaughter of cattle during the last 10% of the gestation period is prohibited in the European Union, such cattle are sometimes sent for slaughter. The late term pregnancy is usually not recognized by the authorities until the uterus is inspected after slaughter and a near term fetus is observed. Accurate post mortem determination of age of bovine fetuses is therefore of major importance as evidence for the subsequent prosecution of the owner. Fetometric measurements such as crown-rump length (CRL) have been used, but these existing estimators have often been established based on insufficiently described study populations or phenotypes that may have changed in the past decades. Morphological characteristics are also used, but few data are available on the correlation between fetal age and the development of these characteristics. The objectives of this study were to investigate the correlation between fetal age and morphological features of bovine Holstein fetuses and to evaluate the use of these features alone and in combination with fetometric measurements to predict fetal age. We collected fetuses from 274 pregnant Holstein cows with recorded insemination dates slaughtered at a Danish abattoir. Gender, teeth development, occurrence of pigmentation, coat, tactile hair and other morphological features were recorded along with CRL, head width, head length and body weight (BW). The gestational length was calculated based on recorded insemination and slaughter dates, and coefficients of variation (R2) were determined for all recorded variables. Notably, the highest R2 was recorded for head length (0.985) followed by CRL (0.979) and head width (0.974). The categorical (morphological) variables were less informative. When used in multivariable models, they did offer statistically significance, but for practical purposes, limited additional information. A multivariable model including the fetometric variables head length and width in combination with CRL resulted in R2 = 0.99 with predictions that were roughly within +/- 11-12 days in 95% of cases. We conclude that the model based on the fetometric variables only provided the most precise predictions, while combination with morphological features such as eruption of teeth, pigmentation and coat mostly increased the width of the prediction intervals.
AB - Although transport and slaughter of cattle during the last 10% of the gestation period is prohibited in the European Union, such cattle are sometimes sent for slaughter. The late term pregnancy is usually not recognized by the authorities until the uterus is inspected after slaughter and a near term fetus is observed. Accurate post mortem determination of age of bovine fetuses is therefore of major importance as evidence for the subsequent prosecution of the owner. Fetometric measurements such as crown-rump length (CRL) have been used, but these existing estimators have often been established based on insufficiently described study populations or phenotypes that may have changed in the past decades. Morphological characteristics are also used, but few data are available on the correlation between fetal age and the development of these characteristics. The objectives of this study were to investigate the correlation between fetal age and morphological features of bovine Holstein fetuses and to evaluate the use of these features alone and in combination with fetometric measurements to predict fetal age. We collected fetuses from 274 pregnant Holstein cows with recorded insemination dates slaughtered at a Danish abattoir. Gender, teeth development, occurrence of pigmentation, coat, tactile hair and other morphological features were recorded along with CRL, head width, head length and body weight (BW). The gestational length was calculated based on recorded insemination and slaughter dates, and coefficients of variation (R2) were determined for all recorded variables. Notably, the highest R2 was recorded for head length (0.985) followed by CRL (0.979) and head width (0.974). The categorical (morphological) variables were less informative. When used in multivariable models, they did offer statistically significance, but for practical purposes, limited additional information. A multivariable model including the fetometric variables head length and width in combination with CRL resulted in R2 = 0.99 with predictions that were roughly within +/- 11-12 days in 95% of cases. We conclude that the model based on the fetometric variables only provided the most precise predictions, while combination with morphological features such as eruption of teeth, pigmentation and coat mostly increased the width of the prediction intervals.
U2 - 10.1371/journal.pone.0207682
DO - 10.1371/journal.pone.0207682
M3 - Journal article
C2 - 30452469
AN - SCOPUS:85056732057
SN - 1932-6203
VL - 13
JO - PLoS Computational Biology
JF - PLoS Computational Biology
IS - 11
M1 - e0207682
ER -