Browsing by Subject "Milk constituents"

Now showing 1 - 1 of 1

Prediction of ruminal acidosis in dairy cows from milk constituents
(2022) Seyfang, Gero Marc; Rodehutscord, Markus
Subacute rumen acidosis (SARA) is a common, but hardly assumable disease in modern dairy cows’ herds. SARA incidences are prevalent in two circumstances. The first, when the cows have to adapt fast to a ration high in carbohydrates after parturition. Since the feed composition has to be changed fast, to meet the cows’ requirements energy- and nutrients wise, the rumen microbiota climate has to adapt fast, which can cause unbeneficial rumen circumstances. The second, when the lactating cows have, beside high milk yield also a high feed intake in mid-lactation, when feed high in energy but low in structural carbohydrates is fed. This can lead to high density of VFAs in the rumen, if the outflow and absorption through the ruminal wall, as well as the buffer capacity in the rumen is not sufficient for the high production of those acids. Then the ruminal milieu becomes more acid, which can negatively affect the cow’s health. The cows suffering SARA, if at all, show mild symptoms like reduced water and feed intake, depression, diarrhea, reduced rumen motility, laminitis or reduced milk yield and milk fat depression. Since those symptoms can also show up with a delay in time and can be caused by several other factors, monitoring SARA in herds can be difficult. An unambiguous definition of SARA circumstances in the rumen cannot be found in literature, although it is under research for decades. Since SARA can influence the milk yield and can lead to a milkfat depression and a change in composition of milkfat, we focused on milk parameters and milkfat composition in particular with the aim of correlating those with pH conditions in the rumen. Three trials were made with feeding rations that were predictably capable of inducing SARA conditions in mid-lactation. During the trials, besides performance and ruminal parameters, as well as continuous pH measurement, milk samples were taken. The cows used were all rumen cannulated. Therefore, datalogger with integrated pH meter (Large Ruminant Logger M5-T7, Dascor Inc., Oceanside, USA) were placed in the ventral sac of the rumen to measure reliably and continuously. In Trial 1, three feeding rations with constant 20% grass silage were used. One ration consisted of additionally 20% corn silage and 60% concentrate (treatment CS60), the other two rations included 20 respectively 60% pressed sugar beet pulp silage and 60 respectively 20% concentrate (treatments SBPS60 respectively SBPS20). With those rations, low pH values were induced in the rumen, leading to SARA incidences of 89% in the measured days in the CS60, 100% in the SBPS60, and 61% in the SBPS20 treatment. In Trial 2, for all three rations a fix concentration of 52% concentrate was used. The remaining 48% consisted of corn silage (treatment CS), grass silage (treatment GS) or hay (treatment Hay). In the CS treatment, SARA incidence was 23%, while the GS and Hay treatments did not show SARA incidence. While the first two trials were designed as a 3x3 Latin square, in Trial 3 the cows remained in their respective treatment. One group stayed in the barn with a TMR, including 30% concentrate (treatment CG), while the other group was full time grazing and got additional 1.75 kg concentrate per day (treatment PG). SARA incidences were 7% in the CG and 8% in the PG. Additionally, in an intertrial approach, regression models for SARA detection were developed. Therefore, easily accessible performance data from the barn and milk parameters from the official milk control and milk fatty acids were used to estimate the rumen parameters pH mean and the time spent below pH 5.8. One first model was designed to include 63 variables. Besides 11 parameters gained in the barn or from the official milk control, also 52 parameters that were gaschromatographically detected fatty acids and sums of these fatty acids. A second model was designed to be useable if no gaschromatographical milkfat analysis was available. Therefore, only those FAs were included that can be estimated in a good quality with MIR spectroscopy. With those regression models the SARA days from the 185 measurement days were calculated to test the accuracy of the models. From the original 47 SARA days the first model was able to detect 43 days and the second model detected 39 SARA days. Although the accuracy of SARA prediction based on these models might be too inaccurate for a decision if a single day was SARA prevalent or not, an information on herd basis seems assessable. Still the small number of cows and measured days, as well as the fact that two breeds of cows and only cows in the later lactation phase were integrated in the model establishment has to be considered and further developed before it becomes a useful tool in field use for SARA detection.