Abstract P269: Circulating Fatty Acids, Determinants and Use as Biomarkers for Dietary Intake: The CoDAM and Hoorn Study
Objective: To increase the understanding of circulating fatty acids (FA) as biomarkers of FA intake, we investigated (1) determinants of circulating proportions of linoleic acid (LA), alpha-linolenic acid (ALA), eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA); and (2) the effect of demographic, lifestyle, and health factors on the relation between circulating and self-reported intake of LA, ALA, EPA, and DHA.
Hypothesis: We hypothesized that demographic, lifestyle, and health factors influence the relation between circulating and self-reported intake of FA.
Methods: Cross-sectional analysis in combined data from the CoDAM (n=472) and Hoorn study (n=708). Dietary FA intakes (% of total FA) were calculated from a validated 79-item semi-quantitative food frequency questionnaire. Fasting total fatty acids (% of total FA) in plasma (CoDAM) or serum (Hoorn) were measured by gas liquid chromatography. The variation in circulating proportions of FA explained by demographic, lifestyle and health factors was calculated by multivariable linear regression analysis. Correlation coefficients between circulating proportions of FA and self-reported FA intakes were calculated by standardized multivariable linear regression analysis adjusted for demographic, lifestyle, and health factors. To assess the influence of specific factors on correlations, stratified analyses were performed and interactions were calculated.
Results: Self-reported intakes were the primary determinants of circulating proportions of LA (partial R2: 7%), ALA (2%), EPA (9%) and DHA (16%). Standardized regression coefficients between circulating and self-reported FA were: LA β=0.280 (95% Confidence Interval: 0.227-0.333), ALA β=0.130 (0.071-0.188), EPA β=0.338 (0.281-0.395), and DHA β=0.450 (0.397-0.503). Other determinants of circulating FA were the use of lipid lowering drugs, waist circumference and sex for LA; prevalence of type 2 diabetes mellitus, age, sex, and alcohol intake for EPA; and age for DHA. The correlation between circulating and dietary LA was stronger among people with a lower vs higher waist circumference and higher vs lower alcohol intake (interaction: p<0.05). In women, the correlation between circulating and dietary EPA and DHA was weaker than in men, and the correlation between circulating and dietary DHA was higher with higher alcohol intake. Underreporting of energy intake did not affect the correlations.
Conclusion: Self-reported intake of FA is the primary, but not the only determinant of circulating proportions of LA, ALA, EPA and DHA. This analysis indicates that demographic, lifestyle, and health characteristics may influence the relation between circulating proportions and self-reported intake of FA. Improved understanding is needed of factors determining circulating FA and the implication for their use as biomarkers of dietary intake in different subgroups.
Author Disclosures: A.J. Wanders: A. Employment; Significant; Employment Unilever R&D. S.E.M. De Hoon: A. Employment; Significant; Employment Unilever R&D. M. Alssema: A. Employment; Significant; Employment Unilever R&D. E.J.M. Feskens: None. G.J. Van Woudenbergh: None. C.J. Van Der Kallen: None. G. Nijpels: None. P.L. Zock: A. Employment; Significant; Employment Unilever R&D. H. Refsum: None. C.A. Drevon: None. A. Elshorbagy: None. C.G. Schalkwijk: None. C.D.A. Stehouwer: None. J.M. Dekker: None. M.J. Van Greevenbroek: None.
- © 2016 by American Heart Association, Inc.