Development of an LC-MS/MS analytical method for the simultaneous measurement of aldehydes from polyunsaturated fatty acids degradation in animal feed Caroline Douny,* Pinar Bayram, François Brose, Guy Degand and Marie-Louise Scippo Knowing that polyunsaturated fatty acids can lead to the formation of potentially toxic aldehydes as secondary oxidation prod- ucts, an analytical method using liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) detection was devel- oped to measure the concentration of eight aldehydes in animal feed: malondialdehyde (MDA), 4-hydroxy-2-nonenal (4-HNE), 4-hydroxy-2-hexenal (4-HHE), crotonaldehyde (CRT), benzaldehyde (BNZ), hexanal (HXL), 2,4-nonadienal, and 2,4-decadienal. The developed method was validated according to the criteria and procedure described in international standards. The evaluated parameters were specificity/selectivity, recovery, precision, accuracy, uncertainty, limits of detection and quantification, using the concept of accuracy profiles. These parameters were determined during experiments conducted over three different days with ground Kellogg’s® Corn Flakes® cereals as model matrix for animal feed and spiked at different levels of concentration. Malondialdehyde, 4-HHE, 4-HNE, crotonaldehyde, benzaldehyde, and hexanal can be analyzed in the same run in animal feed with a very good accuracy, with recovery rates ranging from 86 to 109% for a working range going from 0.16 to 12.50 mg/kg. The analysis of 2,4-nonadienal and 2,4-decadienal can also be performed but in a limited range of concentration and with a limited degree of accuracy. Their recovery rates ranged between 54 and 114% and coefficient of variation for the intermediate precision between 11 and 25% for these two compounds. Copyright © 2016 John Wiley & Sons, Ltd. Keywords: aldehydes; liquid chromatography; tandem mass spectrometry; animal feed; validation Introduction Nowadays, many nutritional and health studies recommend a higher consumption of fat composed of polyunsaturated fatty acids (PUFA), mainly n-3 polyunsaturated fatty acids. [1,2] Increasing the amount of n-3 fatty acids in animal feed is a way to increase human intake of those compounds through the consumption of food from animal origin other than fatty fish, the major natural dietary source of long chain n-3 fatty acids. [3–5] Consequently, many products enriched with n-3 fatty acids can be found in the market: meat, milk and dairy products, eggs, etc. A good way to increase the amount of n-3 fatty acids in animal feed is to use linseed (also known as flax or Linum usitatissimum) for its high α-linolenic acid content (more than 50% of the fatty acids) [3,6,7] or fish oil and microalgae for their eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) content. [8,9] Unfortunately, at these high polyunsaturated fatty acids levels there is rapid oxidation to other less desirous productsd [10–12] ; consequently, it is important to monitor this potential oxidation pathway in animal feed because it can lead to less polyunsaturated fatty acids in meat, milk, or eggs than expected. The formed aldehydes, including malondialdehyde (MDA), 4-hydroxy-2-nonenal (4-HNE), 4-hydroxy-2-hexenal (4-HHE), crotonaldehyde (CRT), benzaldehyde (BNZ), hexanal (HXL), 2,4-nonadienal and 2,4-decadienal can be relatively stable and some of them have been shown to be cytotoxic and genotoxic by reacting with proteins and nucleic acids. [13–16] What is more, in 2011, in an advisory note from the Belgian Superior Health Council, MDA, crotonaldehyde, and 4-HNE have been considered of major concern for human health. [17] Until now, no maximum levels have been established for aldehydes in food or animal feed, mainly because of the lack of data in the literature about toxic concentra- tions, quantities contained in food or feed, and quantities of these oxidation products actually ingested. Most of the published methods described to measure aldehydes in animal feed use a colourimetric method called the TBARS (thiobar- bituric acid reactive species) method. [18–20] However, this method lacks specificity because it measures the total content of aldehydes able to react with thiobarbituric acid, expressed in malondialdehyde content, instead of measuring each compound individually. Eight aldehydes that can be formed after PUFA oxidation were chosen to evaluate the lipid oxidation from linseed in animal feed (MDA, 4-HNE, 4-HHE, CRT, BNZ, HXL, 2,4-nonadienal and 2,4-decadienal (Figure 1)) and a liquid chromatography coupled * Correspondence to: Caroline Douny, Department of Food Sciences, Laboratory of Food Analysis, FARAH - Veterinary Public Health, University of Liège, Quartier Vallée 2, Avenue de Cureghem 10, Sart Tilman B43bis - 4000 Liège, Belgium. E-mail: cdouny@ulg.ac.be Department of Food Sciences, Laboratory of Food Analysis, FARAH - Veterinary Public Health, University of Liège, Quartier Vallèe 2, Avenue de Cureghem 10, Sart Tilman B43bis - 4000 Liège, Belgium Drug Test. Analysis 2016,8, 458–464 Copyright © 2016 John Wiley & Sons, Ltd. Research article Drug Testing and Analysis Received: 17 August 2015 Revised: 18 December 2015 Accepted: 7 April 2016 Published online in Wiley Online Library (www.drugtestinganalysis.com) DOI 10.1002/dta.2013 458