|Van Der Kamp, Henk|
Submitted to: Journal of Agricultural and Food Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/31/2011
Publication Date: 4/1/2011
Citation: Lehotay, S.J., Koesukwiwat, U., Van Der Kamp, H., Mol, H.G., Leepipatpiboon, N. 2011. Qualitative aspects in the analysis of pesticide residues in fruits and vegetables using fast, low-pressure gas chromatography - time-of-flight mass spectrometry. Journal of Agricultural and Food Chemistry. 59:7544-7556. Interpretive Summary: For monitoring purposes, a wide range of pesticide residues in numerous food commodities are routinely analyzed in many regulatory and private contract laboratories worldwide. Not only must the concentrations of detected pesticide residues be determined (quantification), but modern analytical methods must also be able to accurately identify which pesticides are present and which are absent. This study tests the ability of a fast detection method to adequately identify pesticide residues in fruits and vegetable samples using automated software or human judgment when making decisions. An outcome of this study will be the use of better software and more sophisticated decision-making to obtain more accurate results in the routine monitoring of pesticide residues for international trade and food safety purposes.
Technical Abstract: Assessment of qualitative results in analytical methods is needed to estimate selectivity and devise criteria for chemical identification, particularly for mass spectrometric analysis. Low-pressure gas chromatography - mass spectrometry (LP-GC/MS) has been demonstrated to increase the speed of analysis for GC-amenable residues in various foods and provide more advantages over the traditional GC-MS approach. We used LP-GC/MS on a time-of-flight (ToF) instrument, which provided high sample throughput with <10 min analysis time. In this study to assess qualitative performance, we analyzed 90 samples in total of strawberry, tomato, potato, orange, and lettuce extracts from the QuEChERS sample preparation approach. The extracts were randomly spiked with different pesticides at different levels, both unknown to the analyst, in the different matrices. We compared automated software evaluation with human assessments in terms of false positive and negative results. Among the 13,590 possible permutations with 696 blind additions made, the automated software approach yielded 1.2% false presumptive positives with 23% false negatives whereas the analyst achieved 0.8% false presumptive positives and 17% false negatives for the same analytical data files. The false negative rate was reduced to 5-10% if some problematic analytes were excluded. Despite its somewhat better performance in this study, the analyst approach was extremely time-consuming and would not be practical in high sample throughput applications for so many analytes in complicated matrices.