Relationships Between Apparent and True Metabolizable Energy and the Effects of a Nitrogen Correction

Abstract

Some relationships among four common estimates of bioavailable energy (BE), namely apparent (AME) and true (TME) metabolizable energy and their nitrogen-corrected equivalents (AMEn, TMEn) are discussed. Data from four poultry experiments are used to illustrate the relationships. In terms of accuracy, both AME and AMEn substantially underestimate BE at low feed intakes. Within the range of practical ad libitum feed intakes, AME<AMEn when retained nitrogen (RN) <0, and AME> AMEn when RN>0. At low intakes, TME overestimates BE but the difference is less than the underestimation of BE either AME or AMEn. The bioassay for TMEn provides a good estimate of BE at all levels of intake. The precision of the BE estimators tends to increase with feed intake and is usually greater when a nitrogen correction is applied. The variance associated with the excreta energy of fasted control birds causes the TME and TMEn estimates to be less precise than the corresponding AME and AMEn values at a specific level of feed intake. Theoretical considerations and experimental evidence indicate that TMEn is the most useful estimate of BE for practical purposes.