Efficiency of energy utilization for maintenance in growing cattle may be greater than current recommendations: A nonparametric modeling approach

Abstract Text: Energy requirement recommendations for cattle are based on estimates of partial efficiencies of utilizing dietary energy for maintenance (k_m) and growth (k_g) and also maintenance requirements of net (NE_m) and metabolizable energy (ME_m). In growing cattle, an exponential heat production model is traditionally used to estimate NE_m and an indirect estimate of k_m is derived through the ratio of NE_m and estimated ME_m. Conversely, direct estimates of energetic efficiencies may be derived by the utilization of a broken line energy retention model. However, the suitability of the broken line model in describing energy retention as a function of energy intake has been challenged. Consequently, diminishing returns and sigmoidal types of relationship have also been proposed for the analysis of energy balance data. The use of nonlinear response functions implies the assumption of distinct biological principles, for instance, different models are required for varying ranges of intake. The objective of this study was to propose a nonparametric approach for analyzing energy balance data which does not rely on the specification of a particular energy response function. Results of this nonparametric approach were compared with estimates of NE_m from an exponential heat production model. Results were also compared with estimates of k_m derived through the ratio of NE_m and an estimate of ME_m from a broken line model, as suggested by the current US feeding system for beef cattle. Models were implemented in a Bayesian framework using 719 growing cattle energy balance records from calorimetric studies. The smooth nonparametric energy retention curve was estimated through a mixed model representation of penalized splines. ME_m was similar between the two approaches (0.53 MJ/kg BW^0.75 vs. 0.52 MJ/kg BW^0.75 for the nonparametric and conventional approaches, respectively). However, NE_m and k_m were substantially larger in the nonparametric approach. Specifically, the estimates of NE_m in the nonparametric and conventional approaches were 0.49 (SD = 0.05) and 0.38 MJ/kg BW^0.75 (SD = 0.02). Similarly, k_m estimates in the nonparametric approach was substantially larger (92% vs. 73%). Both approaches provided a good fit to the data, suggesting that major differences in NE_m and k_m estimates are obtained when using energy retention and heat production models. The nonparametric approach provides an alternative method for analyzing energy balance data when identification of a specific parametric model for the energy trait is challenging and needs to be considered in updates of national recommendations for beef.

Keywords: energy, efficiency, beef cattle