First colombian interlaboratory study for the determination of aflatoxin B1 in yellow corn

Abstract

An interlaboratory study for the determination of aflatoxin B1 in yellow corn was conducted with 16 laboratories that analyze for aflatoxins in Colombia. Naturally contaminated ground yellow corn (Monsanto DK 4004) with an assigned reference value of 26.3 μg/kg aflatoxin B1 was distributed as double blind duplicates in sachets of 55 g. z-Scores were computed for each of the results; repeatability of the two replicate analysis was also calculated. Four of the participating laboratories used HPLC, seven used TLC, one used fluorometry, one ELISA and three a semi quantitative analytical technique (Aflacard®). Only 10 of the 26 quantitative results (39%) had satisfactory z-scores, two scores were questionable (8%) and 14 of the 26 results had unsatisfactory z-scores (54%). A total of 8 laboratories had satisfactory repeatability (62%), and 5 had unsatisfactory repeatability (39%). The present study indicates that only about one third of the results for aflatoxin reported by Colombian laboratories have good accuracy (as measured by the z-score of the result), although satisfactory precision (measured as repeatability) is achieved by about two thirds of the laboratories. These results indicate that an improvement in quality assurance is needed in Colombian laboratories. The routine use of reference standards and reference materials is highly recommended.