Circulating fatty acids and risk of coronary heart disease and stroke: Individual participant data meta-analysis in up to 16,126 participants

Background: We aimed at investigating the association of circulating fatty acids with coronary heart disease (CHD) and stroke risk.

Methods and Results: Individual-participant meta-analysis of five UK-based cohorts and one matched case-control study. Fatty acids [i.e. omega-3 docosahexaenoic acid (DHA), omega-6 linoleic acid (LA), monounsaturated (MUFA) and saturated (SFA) fatty acids] were measured at baseline using an automated high-throughput serum nuclear magnetic resonance (NMR) metabolomics platform. Data from 3,022 incident CHD cases (13,104 controls) and 1,606 incident stroke cases (13,369 controls) were included. Logistic regression was used to model the relation between fatty acids and odds of CHD and stroke, adjusting for demographic and lifestyle variables only (i.e. minimally adjusted model) or with further adjustment for other fatty acids (i.e. fully adjusted model). Although circulating DHA, but not LA, was related to lower CHD risk in the fully adjusted model (OR: 0.85; 95%CI: 0.76, 0.95 per standard unit of DHA), there was evidence of high between-study heterogeneity and effect modification by study design. Stroke risk was consistently lower with increasing circulating LA (OR for fully adjusted model: 0.82; 95%CI: 0.75, 0.90). Circulating MUFA was associated with higher CHD risk across all models and with stroke risk in the fully adjusted model (OR: 1.22; 95%CI: 1.03, 1.44). SFA was not related to increased CHD risk in the fully adjusted model (OR: 1.01; 95%CI: 0.93, 1.09), or stroke risk.

Conclusions: We found consistent evidence that LA was associated with decreased risk of stroke and that MUFA with increased risk of CHD. The different pattern between CHD and stroke in terms of fatty acids risk profile suggests future studies should be cautious about using composite events. Different study designs are needed to assess which, if any, of the associations observed is causal.