What "requirements" for essential amino acids in broiler chickens? A critical survey of their determination and of some practical models

To balance the feed levels in essential amino acids, nutritionists use tables containing average requirements stemming from bibliographical surveys. Result of experiments studying essential amino acid requirements, depend both on the criterion to be optimized (growth rate or feed efficiency) and on the methodology adopted for analysing the results. From an experiment to the other, environmental conditions and chicken genotype may also explain a part of the observed variability of the requirements. To cope with this variability, the results from numerous experiments obtained mainly in fast growing broilers raised under optimal conditions are pooled but the results of those grouped analyses are not precise, and this impairs the application of measured amino acid requirements to field conditions. Mechanistic models integrating basic knowledge of the energy and protein metabolism might constitute additional tools to quantify the effects of an essential amino acid level on broiler performances under different conditions. Models not only synthesise the knowledge acquired by various experiments but also take simultaneously into account the multiple interactions among factors of variation. There are some marketed models measuring the effects of the main feed characteristics (essential amino acids, energy etc.) and raising conditions on chicken growth. Their results seem precise enough for an application but they are presented as "black boxes" and the complexities of their functioning make them difficult to update. INAVI is a model fully open to the user that might be more adapted to the diversity of production systems. Prediction of the growth of the chicken by INAVI is based on the effects of environmental and nutritional factors on feed intake regulation.