ABSTRACT
An interlaboratory study was conducted to assess the applicability of a previously validated method for the analysis of deoxynivalenol (DON) in cereal and cereal products to soft wheat in the range of >0.1-3.0 microg/g. The study evaluated a generic method to determine DON at levels that bracket the existing Canadian guidelines for DON in soft wheat destined for use in baby foods and nonstaple foods. Collaborators selected one of 2 approved brands of DON immunoaffinity column for cleanup and their choice of qualified C18 liquid chromatographic (LC) column. Separation was by LC with UV detection. Blind duplicates from 5 levels of naturally contaminated wheat and a pair of spiked wheat samples were successfully analyzed by 12 laboratories in 8 countries. For samples naturally contaminated with DON from <0.1-2.2 microg/g, the relative standard deviation of repeatability (RSDr) ranged from 3.1 to 14.8%. For reproducibility, the RSDR ranged from 21.0 to 32.9% and the HorRat range was 1.0 to 1.9. Recoveries of 0.5 microg/g DON spiked into wheat ranged from 66 to 98%, with an average of 84%. The RSDr was 5.4%, the RSDR was 12.6%, and the HorRat value was 0.7.
