Identification and confirmation of chemical residues in food by chromatography-mass spectrometry and other techniques

Abstract

A quantitative answer cannot exist in analysis without a qualitative component to give enough confidence that the result meets the analytical needs (i.e. the result relates to the analyte and not something else). Just as a quantitative method must typically undergo an empirical validation process to demonstrate that it is fit for purpose, qualitative methods should also empirically demonstrate that they are suitable to meet the analytical needs. However, thorough qualitative method validation requires analysis of a great number of samples (possibly more than can be reasonably done), which is generally avoided due to the time and the effort involved. Instead, mass spectrometry (MS) is generally assumed to be the gold standard for qualitative methods, and its results are typically unquestioned. For example, a system was developed by European regulators of veterinary drug residues in food animals (2002/657/EC), in which the number of identification points given in MS analyses depends on the general degree of selectivity of the MS technique used. This well-defined approach gives a definite answer for decision-makers, so it has grown in popularity. However, the identification-points system is not scientific. The reality is that each situation requires information gathering and careful deductive thinking on the part of the analyst to make MS identifications. Rather than devise arbitrary requirements that need to be met by an unthinking analyst, we remind the analytical community that confirmation can be given only if two or more independent analyses are in agreement, preferably using orthogonally selective (independent) chemical mechanisms. In this article, we discuss the proper use of terminology, highlight the identification power of various MS techniques, demonstrate how MS identifications can fail if precautions are not taken, and re-assert the value of basic confirmation practices, qualitative method validation, information checklists, routine quality-control procedures, and blind proficiency-test analyses.