Amino acid analysis of feeds in The Netherlands: four-year proficiency study

To improve accuracy and precision of amino acid analysis in 12 Dutch feed laboratories, a proficiency study was organized twice annually over a 4-year period. The method used included reflux acid hydrolysis for 22 h followed by evaporation, separation on a cation-exchange resin in an amino acid analyzer, and photometric detection after post-column derivatization with ninhydrin. For the determination of sulfur-containing amino acids, samples were oxidized prior to hydrolysis. For the determination of tryptophan, samples underwent an alkaline hydrolysis excluding oxygen, were separated by liquid chromatography on a Hypersil ODS analytical column, and were assayed by UV or fluorescence detection. The average relative standard deviations within (CVr) and between (CVR) laboratories were 3 and 7%, respectively. For some mixed feed samples, the results from the proficiency study were compared with those obtained by the International Analytical Group. For those samples, the relative standard deviation of the difference between IAG and Dutch groups was only 1.9%. For samples that were analyzed twice during this 4-year period, relative standard deviation of between-series differences was only 2.1%.