Methods to create thermally oxidized lipids and comparison of analytical procedures to characterize peroxidation.

Abstract

The objective of this experiment was to evaluate peroxidation in 4 lipids, each with 3 levels of peroxidation. Lipid sources were corn oil (CN), canola oil (CA), poultry fat, and tallow. Peroxidation levels were original lipids (OL), slow-oxidized lipids (SO), and rapid-oxidized lipids (RO). To produce peroxidized lipids, OL were either heated at 95°C for 72 h to produce SO or heated at 185°C for 7 h to produce RO. Five indicative measurements (peroxide value [PV], p-anisidine value [AnV], thiobarbituric acid reactive substances [TBARS] concentration, hexanal concentration, 4-hydroxynonenal [HNE] concentration, and 2,4-decadienal [DDE]) and 2 predictive tests (active oxygen method [AOM] stability and oxidative stability index [OSI]) were performed to quantify the level of oxidation of the subsequent 12 lipids with varying levels of peroxidation. Analysis showed that a high PV accurately indicated the high level of lipid peroxidation, but a moderate or low PV may be misleading due to the unstable characteristics of hydroperoxides as indicated by the unchanged PV of rapidly oxidized CN and CA compared to their original state (OL). However, additional tests, which measure secondary peroxidation products such as AnV, TBARS, hexanal, HNE, and DDE, may provide a better indication of lipid peroxidation than PV for lipids subjected to a high level of peroxidation. Similar to PV analysis, these tests may also not provide irrefutable information regarding the extent of peroxidation because of the volatile characteristics of secondary peroxidation products and the changing stage of lipid peroxidation. For the predictive tests, AOM accurately reflected the increased lipid peroxidation caused by SO and RO as indicated by the increased AOM value in CN and CA but not in poultry fat and tallow, which indicated a potential disadvantage of the AOM test. Oxidative stability index successfully showed the increased lipid peroxidation caused by SO and RO in all lipids, but it too may have disadvantages similar to AnV, TBARS, hexanal, DDE, and HNE because OSI directly depends on quantification of the volatile secondary peroxidation products. To accurately analyze the peroxidation damage in lipids, measurements should be determined at appropriate time intervals by more than 1 test and include different levels of peroxidation products simultaneously.