NIRS AS A TOOL TO PREDICT TRUE ILEAL DIGESTIBLE AMINO-ACID CONTENTS OF FEEDSTUFFS

The nutritional Value of a feedstuff for protein growth has traditiona lly been based on the feedstuffs nitrogen content (multiplied by 6.25 to get an estimate of the protein content). However, since animals are in need of a specific pattern of available amino acids to accrete pro tein, nitrogen is an ineffective indicator of the value of a feedstuff for the animal given that the nitrogen content and the available amin o-acid content of a feedstuff are poorly correlated. Near infrared ref lectance spectroscopy (NIRS) is being employed in the animal feed indu stry since it provides a rapid tool to evaluate feedstuffs; however, c urrently the use of NIRS is limited to the prediction of the nitrogen/ protein content of feedstuffs, and the NIRS results are thus of limite d relevance to a nutritionist. In our laboratory we have evaluated the possibility to use NIRS to predict the true ileal digestible amino-ac id content of feedstuffs as an estimate of amino-acid availability. Th ese tests indicated that monochromators are capable of predicting the digestible essential amino-acid content of animal meals with an accura cy which makes the NIRS interesting as a routine tool for the evaluati on of the nutritional value of feedstuffs in feedmills. For example, d igestible lysine predictions for meat & bone meal were able to explain 77% of the variation within meat & bone meal. In order to successfull y implement (on-line) NIRS technology in the feed industry, it thus se ems that calibration development for (on-line) NIRS should focus on a measure of amino-acid availability rather that nitrogen. (C) 1997 Else vier Science B.V.