Prediction of digestibility of nutrients and energy values of pig diets from chemical analysis

Abstract

The DE and ME values and digestible nutrient contents of 114 diets were measured in 45-kg growing pigs (four to five animals per diet) fed approximately 500 kcal of ME/kg BW.60. Diets differed widely with regard to their chemical characteristics and their ingredients. Chemical composition of each diet was measured by at least four laboratories. The results were used to establish prediction equations of DE or ME values, digestible nutrient contents, and digestibility coefficients of energy and nutrients from chemical characteristics. Digestibility coefficients of energy (range: 65 to 95%) and CP (range: 64 to 94%) were highly dependent on dietary fiber and mineral contents. The digestibility coefficient of ether extract increased curvilinearly (from 2 to 84%) with the dietary fat content. The digestibility of fiber was lower (45% for NDF) than for the other chemical constituents. The ME:DE ratio averaged 96.3% and was negatively correlated to the dietary protein content. The DE and ME values could be accurately predicted (R2 > .90 and CV < 2%) from chemical characteristics; the best equations were obtained when the following predictors were combined in a linear model: ash, ether extract, crude protein, and an estimate of dietary fiber. The accuracy of the prediction was higher with NDF than with ADF or Weende crude fiber. The results suggest that even a rather large proportion of dietary fiber (approximately 50%) is degraded in the digestive tract, the amount of available energy from fiber digestion is negligible in connection with increased endogenous protein and fat losses. The equations obtained in the present study represent a basis for the prediction of the energy values of mixed diets with a composition of unknown ingredients.