Milk electrical conductivity in Manchega ewes: Variation throughout milking and relation with mammary gland health status

The aim of this work was to study the effect of milking fraction and mammary gland health status on the electrical conductivity (EC) of milk from Manchega ewes, considering also the lactation number. To this end, we also studied the relationship of EC with milk macrocomposition, and the relation existing between EC and somatic cell count (SCC). Finally, the use of EC thresholds as a mastitis detection method (sensitivity, specificity, positive (PPV) and negative predictive value (NPV)) was assessed in each of the three fractions: first streams (F1), machine milk (F2) and stripping milk (F3). Milking fraction, mammary gland health status and lactation number had a significant effect on EC and SCC. In the case of EC, the milking fraction caused a more pronounced effect than health status of the glands or number of lactation (F = 19.95, 15.88 and 6.55, respectively; P < 0.5). In SCC, the gland health status caused the most pronounced effect followed by milking fraction and lactation number (F = 112.02, 6.89, and 5.28, respectively; P < 0.05). Changes in the milk composition, especially fat and lactose contents, explained the EC variation to a great extent. For the same EC threshold, specificity and sensitivity varied slightly depending on the milking fraction. NPV above 80% was obtained in the three milking fractions and at all EC thresholds tested, but PPV was only higher than 20% as of the threshold of 4.5, 4.4, and 4.2 mS/cm in F1, F2 and F3, respectively. From the results obtained, we concluded that the algorithm design for mastitis detection in sheep should include those factors affecting the composition and which therefore cause variations in EC, such as milking fraction, individual differences, lactation stage or lactation number.