BACKGROUND:
Dietary guidelines support intake of polyunsaturated fatty acids (PUFAs) in fish and vegetable oils. However, some controversy remains about benefits of PUFAs, and most prior studies have relied on self-reported dietary assessment in relatively homogeneous populations.

METHODS AND RESULTS:
In a multiethnic cohort of 2837 US adults (whites, Hispanics, African Americans, Chinese Americans), plasma phospholipid PUFAs were measured at baseline (2000-2002) using gas chromatography and dietary PUFAs estimated using a food frequency questionnaire. Incident cardiovascular disease (CVD) events (including coronary heart disease and stroke; n=189) were prospectively identified through 2010 during 19 778 person-years of follow-up. In multivariable-adjusted Cox models, circulating n-3 eicosapentaenoic acid and docosahexaenoic acid were inversely associated with incident CVD, with extreme-quartile hazard ratios (95% CIs) of 0.49 for eicosapentaenoic acid (0.30 to 0.79; Ptrend=0.01) and 0.39 for docosahexaenoic acid (0.22 to 0.67; Ptrend<0.001). n-3 Docosapentaenoic acid (DPA) was inversely associated with CVD in whites and Chinese, but not in other race/ethnicities (P-interaction=0.01). No significant associations with CVD were observed for circulating n-3 alpha-linolenic acid or n-6 PUFA (linoleic acid, arachidonic acid). Associations with CVD of self-reported dietary PUFA were consistent with those of the PUFA biomarkers. All associations were similar across racial-ethnic groups, except those of docosapentaenoic acid.

CONCLUSIONS:
Both dietary and circulating eicosapentaenoic acid and docosahexaenoic acid, but not alpha-linolenic acid or n-6 PUFA, were inversely associated with CVD incidence. These findings suggest that increased consumption of n-3 PUFA from seafood may prevent CVD development in a multiethnic population.