Adsa Logo White Adsa Title White

Empirical modelling of vitamin B12 duodenal flow in lactating dairy cows.

V. Brisson

Abstract:

M15
Empirical modelling of vitamin B12 duodenal flow in lactating dairy cows.
V. Brisson*1, C. L. Girard2, J. A. Metcalf3, D. S. Castagnino3, J. Dijkstra4, J. L. Ellis1. 1University of Guelph Guelph, ON, Canada, 2Agriculture and Agri-Food Canada Sherbrooke, QC, Canada, 3Trouw Nutrition Canada Guelph, ON, Canada, 4Wageningen University and Research Wageningen, the Netherlands.

Unlike other B vitamins, vitamin B12 is not found in plants and is produced only by bacteria. Therefore, supply to the dairy cow, unless provided via supplementation, will mainly be the result of B12 manufactured by ruminal microbes. The duodenal flow of B vitamins therefore represents the amount of vitamin available for absorption by the ruminant, which can be used for essential metabolic functions and milk production. However, diet composition may affect ruminal synthesis and the resulting duodenal flow (DF) of vitamin B12, due to alterations to fermentation and ruminal conditions. Therefore, the objective of this study was to conduct a meta-analysis describing how diet composition affects DF of vitamin B12. Data were collected from 340 individual lactating cows involved in 16 published studies. Saved diet and duodenal samples from these studies were subsequently reanalyzed for B vitamin content to create the database used in the present study. Potential driving variables considered included (DM basis) dietary organic matter (%), NDF (%), starch (%), crude protein (%) and DMI (kg/d). The meta-analysis was conducted in 3 steps, followed by statistical evaluation of the resulting empirical models. A Spearman correlation matrix was constructed between all potential driving variables to assess for collinearity between X variables, and guide model creation. Then, using Cook's distance statistic (Proc MIXED), outliers were determined and removed. Finally, a suite of potential models (with study treated as a random effect) were developed in GLIMMIX. Where models were statistically significant, evaluation was completed using root mean square prediction error (RMSPE) and concordance correlation coefficient (CCC) to determine the sources of error. The best performing model was: B12DF (mg/d) = −7.87 (�2.46) + 0.29 (�0.056) � DietNDF(%) + 0.44 (�0.042) � DMI (kg/d); RMSPE: 41.1%, CCC: 0.268. In conclusion, DF of B12 was positively impacted by both the overall DMI and the dietary NDF content of the diet. This information may be used to better understand supply of vitamin B12 to the modern dairy cow, in relation to requirements, to improve milk production efficiency.

Keywords: vitamin B12, meta-analysis, duodenal flow.

Biography: I am a young and determined scientist currently completing my master's degree with the Department of Animal Biosciences at the University of Guelph. I grew up on a dairy farm, and I have come to understand and experience first-hand the crucial impact nutrition can have on a herd, and I therefore aim to return to industry after the completion of my thesis. I firmly believe that my research on B vitamins can be valuable to ensure the sustainability of the dairy industry, and I look forward to networking with both industry and academic researchers to gain exposure and experience.