|Byrdwell, W Craig|
|Harnly, James - Jim|
Submitted to: Journal of Food Composition and Analysis
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/17/2008
Publication Date: 5/16/2008
Citation: Phillips, K., Byrdwell, W.C., Exler, J., Harnly, J.M., Holden, J.M., Holick, M., Hollis, B., Horst, R., Lemar, L.E., Patterson, K.K., Tarrango-Trani, M., Wolf, W.R. 2008. Evaluation and harmonization of measurements of vitamin D in selected food materials. Journal of Food Composition and Analysis. 21:527-534.
Interpretive Summary: Data on vitamin D in the literature and in the USDA National Nutrient Database for Standard Reference have been questioned because of the lack of reference standards and reliable methods. As part of a USDA project to improve the data in the database, an analytical committee was formed to develop 5 control materials (canned salmon and vitamin D fortified cereal, orange juice, milk, and cheese) and to compare analytical methods. Initially, the results from 6 participating labs compared poorly. However, after comparing primary standards and eliminating other systematic errors, good agreement as obtained. These control materials will benefit the analytical and database communities by providing a basis for comparing methods and results. This will significantly increase researcher and consumer confidence in the database values.
Technical Abstract: As part of the United States Department of Agriculture's (USDA) National Food and Nutrient Analysis Program (NFNAP), food composition data for vitamin D in the USDA National Nutrient Database for Standard Reference are being updated and expanded, focusing on high priority foods (fish, and vitamin D fortified orange juice, breakfast cereals, fluid milk, margarines, sliced American cheese, yogurt) and validated analytical methodology. A lack of certified reference materials and analytical methods validated for these key foods required the development of 5 matrix-specific control composite materials (CC) (canned salmon and vitamin D fortified cereal, orange juice, milk, and cheese). Each of 6 experienced laboratories (research and commercial) analyzed vitamin D3 in five subsamples of each CC in five separate analytical batches, with one subsample of each material in each run. Research laboratories performed recovery studies, mass spectrometric analysis, and other studies to validate quantitation in each matrix. Initial results showed a wide disparity between the 6 laboratories with relative standard deviations (RSD) of 26-46%. Extensive collaboration resolved several problems related to calibration standards, extraction solvents and the internal standard, achieving final values with RSDs of approximately 10%, validated by mass spectrometry tests that confirmed lack of matrix interferences in these foods.