Validation of an analytical method to determine the content of fumonisins in baby food, breakfast cereals and animal feed : Report on the collaborative trial "Determination of fumonisins B1 and B2 in baby food, breakfast cereals and animal feed by immunoaffinity column clean-up with high performance liquid chromatography and fluorimetric detection"

An inter-laboratory comparison was carried out to evaluate the effectiveness of a method based on immunoaffinity column clean-up followed by derivatisation and high performance liquid chromatography with fluorimetric quantification (HPLC-FL). The method was tested for the determination of Fumonisins B1 and B2 (FB1 & FB2) in baby food, breakfast cereals and animal feed to monitor compliance with limits according to Regulation 1881/2006/EC and Recommendation 576/2006/EC. The test portion of the sample was extracted with methanol:water. The sample extract was filtered, diluted, passed over an immunoaffinity column for cleanup and evaporated. The redissolved and purified eluate was separated and determined by reverse-phase high performance liquid chromatography (HPLC) and fluorescence detection after the fumonisins had been derivatised to their fluorescent isoindols in the presence of o-phthaldehyde and a thiol-coupling reagent with either pre- or postcolumn derivatisation. Baby food, breakfast cereal and animal feed samples, both blank and naturally contaminated with FB1 and FB2, were sent to 40 laboratories from 19 EU Member States, and a laboratory in Uruguay. For recovery determination extra test portions of the blank samples were to be spiked by the participants at levels of 135 μg/kg for the sum of FB1 and FB2 in baby food, 400 μg/kg in breakfast cereals, and 3700 μg/kg in animal feed. All samples were sent as blinded duplicates. Mean recoveries were calculated as 71 % for baby food, and 87 % for breakfast cereals. Based on results for the spiked and naturally contaminated samples the relative standard deviations for reproducibility (RSDR) in baby food were 31 % at a spiked level of 135 μg/kg, 44 % at a natural contamination level of 267 μg/kg, and 33 % at a natural contamination level of 501 μg/kg. For breakfast cereal these figures were 15 % at a spiked level of 400 μg/kg, and 33 % at a natural contamination level of 1034 μg/kg. The values for RSDr in those materials ranged from 5 to 29 % in baby food and 12 to 14 % in breakfast cereal. For animal feed the recovery was 68 % and RSDR values were 84 % at a spiked level of 3700 μg/kg, and for naturally contaminated samples 68 % at 2730 μg/kg, 88 % at 3695 μg/kg, and 49 % at 10037 μg/kg. The values for RSDr values ranged from 6 to 49 %. European Commission Regulation 401/2006/EC lays down performance criteria that must be met by a method to determine fumonisin FB1 and FB2 in food. These criteria have been met by this method for the baby food and the breakfast cereal, whereas for animal feed the determined performance criteria failed to comply with legal requirements