Testing for mycotoxins involves in general three different steps: sampling, sample preparation
and the analytical procedure. Sampling comprises the selection of a sample of a given
size from a bulk lot, grinding and taking a representative sub-sample of ground material.
Because contaminated particles may not be distributed uniformly throughout the lot, the
sample should be an accumulation of small portions taken from many different locations
(Schmitt and Hurburgh, 1989). Sample preparation consists of several processes, i.e. the
test samples are usually ground and sub-sampled, the mycotoxin is solvent-extracted from
the sub-sample, and the extract is purified before the mycotoxin in the solvent is quantified.
Some methods, such as antibody-based rapid test systems or more elaborate methods
such as liquid-chromatography with (multiple) mass detection, might not require clean-up
in case of simple and/or thoroughly validated matrices. The mycotoxin value, measured
in the analytical step is eventually used to estimate the lot concentration or is compared
to a maximum limit in order to classify the lot as acceptable or unacceptable. This means
that a very small quantity of the lot is used in the final quantification step to estimate the
mycotoxin concentration of the entire lot. Because of the associated uncertainty the true
mycotoxin concentration of a bulk lot cannot be determined with 100% certainty, or can
100% of lots sampled be correctly classified into “good” or “bad” categories.