Estimation of dietary metabolizable energy content in dairy cattle using the measured dietary content of digestible nutrients: A meta-analysis

Le résumé fourni par la source

An accurate estimation of metabolizable energy (ME) content is critical for optimizing dairy cattle nutrition and improving feed efficiency.This meta-analysis used data from studies conducted in Wageningen (the Netherlands) using Holstein cows and in Nebraska using Jersey cows to refine predictive models of dietary ME content (kJ/kg DM) based on total-tract digestible nutrients.In the Wageningen studies (except for one study), apparent total-tract digestibility of nutrients was estimated using chromium oxide as an external marker included in the concentrate.In contrast, the Nebraska studies determined digestibility by difference, based on measured nutrient intake and total fecal output.Importantly, the digestibility values for nutrients used in the models' development were directly measured from total-tract digestibility data, rather than estimated values.Models varied from a simple model including a single total-tract digestible organic matter (dOM) fraction as explanatory variable to complex models in which dOM was fractionated into total-tract digestible crude protein (dCP), fat (ether extract or fatty acids; dFAT), starch (dSTA), neutral detergent fiber (dNDF), sugars (SUG), and a residual fraction (dOM -dFAT -dCP -dNDF -dSTA -0.97 × SUG).Compared with the simplest model, which predicted ME content based on dOM (with a corrected Akaike information criterion [AICc] of 9,570, R 2 of 0.859, and root mean square prediction error [RMSPE] of 3.69% of the observed mean).The complex model with the various digestible components improved model fit by reducing the AICc by 102 units to 9,468, increasing R 2 to 0.878, and lowering the RMSPE to 3.43% of observed mean, thereby achieving an improvement of ~7% in estimation accuracy.Including feeding level (FL; calculated as ME intake [kJ/d/kg BW 0 • 75 ] divided by 562, assuming the ME requirement for maintenance is 562 kJ/d/kg BW 0 • 75 ) further improved model performance, with ME content increasing by ~412 kJ/kg DM per unit rise in FL, underscoring the impact of intake levels on dietary energy estimates.Correcting dFAT and dCP for basal endogenous losses did not further improve model performance.Additionally, location had a significant effect on ME estimation, likely reflecting differences in breed (Holstein vs. Jersey), diet composition, and methodological approaches (e.g., respiration chambers vs. head boxes, marker vs. total collection).Findings from this study provide insight into within-study relationships between observed digestible nutrient fractions and dietary ME content, which may be used for improving energy evaluation systems for dairy cattle.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.