Studies on accuracy of trichothecene multitoxin analysis using stable isotope dilution assays.

Critical parameters in mycotoxin analysis were examined by using stable isotope-labelled tricho-thecenes. Sample weight was downsized to 1 g without loosing precision when sufficiently homogenized samples were taken for analysis. Complete extraction of trichothecenes could be achieved with a solvent mixture of acetonitrile+water (84+16; v+v) even without the use of stable isotope labelled standards. However, in particular for the analysis of deoxynivalenol the absolute amount of water in the solvent volume used for extraction appeared critical. Depending on the matrix a low water amount resulted in too low quantitative values when no stable isotope-labelled standards are applied to correct for incomplete extraction. In this case the used extraction volume had to be at least 10 ml for 1 g sample when acetonitrile + water (84+16; v+v) was used as extraction solvent.Losses during sample preparation using two different clean-up columns were not observed. On the contrary, matrix suppression in the ESI-interface of the LC-MS equipment was found to be a serious problem. Depending on the matrix, the latter effect resulted in considerably lower values for trichothecenes when no stable isotope-labelled standards were used to counterbalance this suppression.