Influence of substitution method and of food intake on bioassays to determine metabolisable energy with chickens

1. Several different methods of assaying the metabolisable energy (ME) of ground maize and a maize and glucose mixture were studied. Over a 24‐h period fasted cockerels lost more energy and nitrogen in their excreta than those fed 15 g maize and 15 g glucose monohydrate.

2. Discrepancies in ME values of maize were found between 1) Hill's method (substitution for glucose) and 2) Sibbald's method (the difference between energy ingested and excreted, corrected for endogenous losses by a comparison with fasted controls). These discrepancies were attributed to differences in endogenous energy losses between fed and fasted birds and interactions between the test substance and reference diet.

3. No significant differences could be detected when substitution for glucose was compared to substitution for the entire diet in three experiments.

4. Significant differences in the ME of maize because of the composition of the basal diet were found (semi‐purified: practical; 17·20:14·56 kJ/g; and 50 g/kg added fat: 100 g/kg added fat; 15·94:16·69 kJ/g).

5. Food intakes equal to 70, 60 or 30% of ad libitum did not affect the ME of maize when the substitution for glucose method was used.

6. It is concluded that basal diets for ME determinations should be carefully selected because of possible interactions between dietary components. When ME is determined by feeding small amounts of the test material, endogenous losses may be over‐estimated because the control birds are in a different physiological state (fasted).