Detection of delayed cyclicity in dairy cows based on progesterone content in monthly milk samples

We examined whether infrequent milk sampling for progesterone analysis could be used as a management or diagnostic tool by the dairy farmer to predict delayed ovarian cyclicity in dairy cows. The data included 1040 lactations from 324 Swedish Red cows and 183 Swedish Holstein cows and were randomly divided into two datasets. A logistic regression model was fit to the first dataset and the model was then validated on the other dataset. The model was also validated using a British dataset comprising 1212 lactations from 1080 British Holstein-Friesian cows. The dependent variable was whether delayed ovarian cyclicity occurred or not, delayed ovarian cyclicity defined as progesterone levels below threshold value for the first 56 or 45 days postpartum in the Swedish or British dataset, respectively. The basic model included the effects of breed, parity, season and housing type. To the basic model various progesterone-based measurements were added. These were the interval from calving to commencement of luteal activity and the percentage of samples with luteal activity within 60 days after calving, using all samples in the databases or one sample per month. The accuracy of the conditional probability of delayed ovarian cyclicity calculated with the different models was obtained using receiver operating characteristic (ROC) curves and calculating the area under curve. Sensitivity and specificity were calculated for cut-off probabilities using the ROC analyses. The accuracy was highest (0.94–0.99) when including the progesterone measurements based on milk sampling for progesterone analysis 2–3 times per week. The accuracy was between 0.85 and 0.88 when the progesterone measurements with monthly milk sampling were added to the model and 0.76 or 0.67 with the basic model. This study clearly shows that infrequent milk sampling for progesterone analysis, such as once a month as in the regular milk recording system, could be used to predict delayed ovarian cyclicity in dairy cows. This increases the opportunity for earlier treatment of anovulatory dairy cows and may therefore decrease involuntarily extended calving intervals in the herd.