Tel.+49 30 838 62618 Fax.+49 30 838 62620
Objectives: Especially heifers suffer from a high prevalence of dystocia. Dystocia is associated with major pain in the dam and an increased risk of stillbirth in calves. One requirement to prevent and manage dystocia in heifers is to recognize the onset of stage 2 of calving. Therefore the objective of our study was to predict the onset of stage 2. We set out to test the predictive value of six signs of imminent parturition (SIP) and the alteration of the broad pelvic ligaments and filling of the teats.
Materials and Methods: Prior to the experiments a systematic literature search was conducted in which six signs of imminent parturition were identified to be plausible parameters to indicate the onset of stage 2 (i.e., tail raising, scurrying, turning the head towards the abdomen, clear vaginal discharge, bloody vaginal discharge and lying lateral with abdominal contractions). The experiments took place on a commercial dairy farm in Germany milking 2,250 cows. In the first experiment the inter-observer reliability of SIP was tested. A total of 22 Holstein-Friesian heifers around calving were included. Two investigators observed each heifer for 15 seconds once per hour walking behind the cows in the pen. In the second experiment SIP were observed every hour 24 hours from 267 d before calving to parturition. 37 Holstein-Friesian heifers were included. Additionally to SIP, relaxation of the broad pelvic ligaments (PL) and filling of the rear teats (TF) were examined twice a day and evaluated on a 4-point scale. We calculated a receiver operating characteristic (ROC) analysis for a combination of SIP. One score point was given for the occurrence of each SIP meaning it ranged from 0 to 4. Besides a receiver operating characteristic analysis was conducted for the combination of score points for PL and TF.
Results: Cohen’s kappa results for the inter-observer reliability were moderate until almost perfect, ranging between 0.507 and 0.912. The cut-off value for occurrence of SIP was >1 with 83.8 sensitivity and 99.2 specificity. A proportion of 99.2 % of heifers having a score under the cut-off value was correctly diagnosed as not calving. Prediction of calving was possible in 36.5 % of calving events. The results of the area under the curve (AUC) for the combination of PL double weighted and TF was about 0.8, implying that in 80% of the examinations a calving heifer will have a higher PL and TF score than a non-calving one. As the results of SIP and PL TF of the negative predictive value were considerably higher than the positive predictive value, we contemplated predicting no calving. The accuracy of excluding calving for the next 12 hours was 96.8% for 29 heifers examined from day 269 days carried calf until parturition corresponding with a cut off value >5 and a 68.97 sensitivity and 87.91 specificity. A binary logistic regression analysis was conducted to determine the effect of group (i.e., late pregnancy vs. 24 h before onset of stage 2) on the occurrence of SIP. Tail raising, clear vaginal discharge and bloody vaginal discharge were more likely to occur 24h before calving then in late regnancy (P ≤ 0.001). While there was a difference in detecting imminent parturition between hourly observation and observation every 3h, 4h, 5h, and 6h, respectively (P ≤ 0.05), there was no difference between hourly observations and observation every two hours (P = 0.139).
Conclusions: Prediction of calving using SIP or the evaluation of PL and TF is challenging. Nevertheless, it is useful to exclude the possibility of calving. Knowing which cow will not calve during next 12 h facilitates monitoring of cows around parturition. The appearance of SIP was very individual regarding the different SIP and the time of occurrence. We could not find a difference between hourly observation and observation every second hour. Hence, we suggest to observe pregnant heifers every 2 hours for tail raising, clear and bloody vaginal discharge and lying lateral with abdominal contractions.