|Lobos, Nelson - UNIV OF WISCONSIN|
|Wattiaux, M - UNIV OF WISCONSIN|
Submitted to: Journal of Dairy Science
Publication Type: Abstract Only
Publication Acceptance Date: July 9, 2007
Publication Date: July 9, 2007
Citation: Lobos, N.E., Wattiaux, M.A., Broderick, G.A. 2007. Comparison of three analytical methods to assess urea nitrogen in colostrum [abstract]. Journal of Dairy Science. 90 (supplement 1):114. Technical Abstract: Milk urea nitrogen (MUN) obtained from cows fed mid to late lactation diets has been used as an indicator of diet composition adequacy and can be used to predict urine urea nitrogen. However, recent research has suggested that in early lactation, MUN was positively correlated with feed efficiency (FCM/DMI) and thus the metabolic status of the cow. Knowledge of urea-N (UN) in colostrum may provide useful information for early lactating cow management. This experiment was conducted to compare three analytical methods to determine the concentration of UN in colostrum. Twenty-seven Holstein cows, housed in a tie stall barn at the U.S. Dairy Forage Research Center (Prairie du Sac, WI) were fed a pre-calving diet with 13.2% crude protein (DM basis), starting two weeks before calving. Colostrum (i.e., the first milking) samples were collected in plastic vials containing the commonly used preservative bronopol® (2-bromo-2-nitropropane-1,3-diol) and kept frozen until analysis. After thawing, samples were divided in three subsamples. The first sample was analyzed in a commercial laboratory by near infrared spectroscopy (NIR) using a Foss 6000® instrument calibrated for milk. The second sample was centrifuged to remove fat and analyzed manually using a colorimetric urease-based assay with sodium nitroprusside as a catalyst (Berthelot reaction (BR)). Absorbance of blue indophenol was read at 550 nm. After precipitation of the protein, the third sample was analyzed by an automated colorimetric assay using the Diacetyl Monoxyme method (DAM) adapted to a flow-injection analyzer (Lachat QuikChem 8000®). Data were analyzed using Proc REG of SAS. Linear models assumed no intercept. Average and (standard deviation) of colostral UN were 20.9 (14.8), 9.5 (4.5), and 11.1 (4.3) mg/dL for the NIR, BR, and DAM methods, respectively. Values ranged from 0.6 to 62.5, from 1.2 to 18.0 and from 4.2 to 23.3 mg/dL for the NIR, BR, and DAM methods, respectively. Regressions were as follow: NIR-UN = 1.69 (standard error (se) = 0.34) x BR-UN (r2 = 0.49, root MSE = 18.6); NIR-UN = 1.66 (se = 0.27) x DAM-UN (r2 = 0.60, root MSE = 16.5), and BR-UN = 0.84 (se = 0.06) x DAM-UN (r2 = 0.89, root MSE = 3.5). For each regression, the slope was different than 1.0 (P<0.05). Because the true UN in colostrum was not known, the most accurate technique cannot be established. Nevertheless in this trial, the NIR-UN was the least precise. In addition, the NIR-UN was poorly correlated with BR-UN and DAM-UN, but the latter two techniques were in close agreement with each other. All cows were fed the same diet prior to calving, thus the wide variation in colostral UN observed in this trial could not be explained by dietary effects but reflected most likely metabolic differences among cows at parturition.