The milk urea content in dairy goat farms of Lombardy

The measurements of milk urea concentration can be used as a diagnostic tool of protein feeding in the dairy cow, providing also accurate prediction of urinary N excretion. For the dairy goat only a few studies have considered the milk urea content in order to diagnose the protein nutritional status, and the sole indication we have found of a "normal" level is the one reported by of Brun-Bellut et al. (1991) who indicate a range of 28-32 mg/dl. The "normal" baseline for bulk cow milk urea concentration ranges from 17-26 to 21-30 mg/dl. In this work we studied the trend of bulk milk urea level (MUL) during three consecutive years (2005-2007) in the dairy goat farms (n=118, on average) of Lombardy involved in the project "Quality of goat Milk”. This project, funded by Lombardy Region, was carried out by the Technical Assistance Service for goat and sheep Farms (SATA) of Lombardy. Bulk milk samples (n=2792) were analysed monthly (8 samples per farm and year, on average). Milk urea concentration was determined by differential pH-metry (EFA 2000 - Hamilton, Bonaduz, Switzerland). Milk fat, crude protein, casein (since July 2007) and lactose content were determined by an automated Fourier Transform infrared analysis (FT IR 6000, FOSS A/S, Denmark), while the counts of somatic cells were per- formed by means of flow cytometry with Fossomatic 5000 FC (FOSS A/S, Denmark). Data were statistically analysed by ANOVA, with year, month and farm as main effects, and by simple or multiple regression, using GLM and REG procedures of SAS (2000). MUL was influenced (P<0.001) by the year: the Least Square Means of MUL were 39.2, 39.9 and 41.2 mg/dl, for 2005, 2006 and 2007, respectively. The variability observed was very high (Range=68.8; SD=11.2) but it resulted independent from the level of production of the herds. For the last two years it was possible to divide the farms into two different systems of breeding: natural breeding season (NBS) (85% of the farms) vs controlled breeding season (CBS). With NBS the MUL increased after kidding season (Jan-Feb-Mar) from 34 to 40 mg/dl, reaching a peak from April to June (44) and then it decreased until November (37). The regression between MUL and the other milk chemical constituents (fat, crude protein, lactose) and SCC (expressed as linear score - LS), had always determination coefficients lower to 0.01. On the other hand, multiple regression analysis (Stepwise method) between casein (% CP) and crude protein, LS and urea, resulted statistically sig- nificant with a model R-square of 0.40 and partial R-square of the independent variables of 0.344, 0.033 and 0.026, respectively. The equation obtained is as follows: Casein (%CP)=2.51xCP(%)–0.28xLS–0.024xUREA(mg/dl)+68.5; (n=330, R2=0.40; P<0.0001). About 75% of the bulk milk urea content in dairy goat farms of Lombardy is higher than the standard level suggested by Brun-Bellut et al. (1991). The average CP content of the diets predicted from the average MUL registered, would be 18-19% on DM, applying the prevision equation proposed by Bonanno et al. (2008). This protein content is well above requirements and the implications in terms of N excretion and even animal welfare, are evident. Further research is needed in order to correctly diagnose the protein nutritional status of the dairy goat.