Impact of chemical composition on metabolizable energy and its prediction models in brewer's spent grains for broilers at different ages.

This study aimed to investigate the relationship between chemical composition and metabolizable energy (ME) in brewer's spent grains (BSG), and to develop ME prediction models for fast-growing white feathered broilers at different ages. Broilers aged 11 to 14 days or 25 to 28 days were fed 11 experimental diets (one basal diet and 10 diets supplemented with BSG) in a 2 × 11 completely randomized factorial arrangement. Each experimental diet was assigned to 6 replicates of 4 broilers, either 11 to 14 days or 25 to 28 days of age. The total excreta collection method was employed to determine the ME of the experimental diet for the calculation of the ME of BSG. Correlations between the chemical composition and ME of BSG were analyzed, and stepwise regression models for apparent metabolizable energy (AME) and nitrogen-corrected AME (AME) were established for broilers aged 11 to 14 days and 25 to 28 days. Substantial variations in crude protein (CP; 18.63% to 38.94%, CV=27.0%) and crude fiber (CF; 11.97% to 32.15%, CV=31.2%) contributed to a wide range of ME values (667 to 3392 kcal/kg DM) across the 10 sources of BSG. Principal component analysis of the chemical composition clearly showed substantial variability in the quality of the BSG samples, as reflected by the dispersed scores of chemical components. Eight of 10 BSG samples had close ME values in broilers aged 11 to 14 days and 25 to 28 days. However, notable differences in ME were observed between these age groups for BSG samples 1 and 5 (P<0.05). The AME and AMEn of BSG were positively correlated with CP, ether extract (EE), and gross energy (GE; P<0.05), but negatively correlated with CF, ash, and acid detergent fiber (ADF) in both age groups (P<0.05). For broilers aged 11 to 14 days, the AME and AMEn of BSG were best predicted by CP and EE contents (R≥0.902, RMSE≤292.9 kcal/kg DM, P<0.01). For broilers aged 25 to 28 days, the AME and AMEn of BSG were best predicted by GE and EE contents (R≥0.926, RMSE≤217.7 kcal/kg DM, P<0.01). This study developed robust, age-specific prediction equations for ME in broilers based on the chemical composition of BSG. Overall, ME values for BSG can be reliably predicted from its CP, EE and GE contents.