Ethnicity, plasma phospholipid fatty acid composition and inflammatory/endothelial activation biomarkers in the Multi-Ethnic Study of Atherosclerosis (MESA)

BACKGROUND/OBJECTIVES: It has been recognized that certain long-chain polyunsaturated fatty acids (LC-PUFAs) are involved in inflammation and its resolution. It has also been shown that ethnicity may be a factor in affecting systemic inflammation, and limited evidence suggests it may influence plasma LC-PUFA composition. Given the links among these three factors, we aim to determine ethnicity-based differences in plasma LC-PUFA composition among White, Black, Hispanic and Chinese participants, and whether such differences contribute to variations in markers of inflammation and endothelial activation in a sub-cohort of the Multi-Ethnic Study of Atherosclerosis (MESA). SUBJECTS/METHODS: Plasma phospholipid LC-PUFAs levels (%) were determined in 2848 MESA participants using gas chromatography-flame ionization detection. Enzyme immunoassays determined inflammatory markers levels for high-sensitivity C-reactive protein (n = 2848), interleukin-6 (n = 2796), soluble tumor necrosis factor-alpha receptor type 1 (n = 998), and endothelial activation markers soluble intercellular adhesion molecule-1 (n = 1192) and soluble E-selectin (n = 998). The modifying influence of ethnicity was tested by linear regression analysis. RESULTS: Chinese adults were found to have the highest mean levels of plasma eicosapentaenoic acid (EPA, 1.24%) and docosahexaenoic acid (DHA, 4.95%), and the lowest mean levels of gamma-linolenic (0.10%), dihomo-gamma-linolenic (DGLA, 2.96%) and arachidonic (10.72%) acids compared with the other ethnicities (all P <= 0.01). In contrast, Hispanics had the lowest mean levels of plasma EPA (0.70%) and DHA (3.49%), and the highest levels of DGLA (3.59%; all P <= 0.01). Significant differences in EPA and DHA among ethnicities were attenuated following adjustment for dietary non-fried fish and fish oil supplementation. Ethnicity did not modify the associations of LC-PUFAs with markers of inflammation or endothelial activation (all P-interaction > 0.05). CONCLUSIONS: The absence of a modifying effect of ethnicity indicates that the putative benefits of LC-PUFAs with respect to inflammation are pan-ethnic. Future longitudinal studies may elucidate the origin(s) of ethnicity-based differences in LC-PUFA composition and whether certain patterns, that is, high plasma levels of DGLA and low levels of EPA/DHA, contribute to inflammation-associated health outcomes.