To prospectively investigate the relation of plasma cholesterol ester (CE) and phospholipid (PL) fatty acid (FA) composition with incidence of coronary heart disease (CHD).

3 591 white participants in the Minneapolis field center of the Atherosclerosis Risk in Communities Study, aged 45–64 years, were studied. Plasma FA composition of CEs and PLs was quantified using gasliquid chromatography and expressed as percentage of total FAs. Incident CHD was identified during 10.7 years of follow-up. In both CE and PL fractions, the proportions of stearic (18:0) acid, dihomo-γ-linolenic (20:3n6) acid and total saturated fatty acids (SFAs) were significantly higher while arachidonic (20:4n6) acid and total polyunsaturated fatty acids (PUFAs) were significantly lower among participants who developed incident CHD (n=282). After adjusting for age, gender, smoking, alcohol drinking, sports activity, and non-FA dietary factors, the incidence of CHD was significantly and positively associated with the proportion of dihomo-γ-linolenic acid but inversely associated with arachidonic acid. The multiply-adjusted rate ratios (RRs) of CHD incidence for the highest versus the lowest quintile were 1.31 in CE and 1.44 in PL for dihomo-γ-linolenic acid (p for trend: 0.05 and 0.017, respectively), 0.59 in CE and 0.65 in PL for arachidonic acid (p: 0.016 and 0.024, respectively). Also significantly and positively associated with incident CHD were PL stearic acid and CE linolenic (18:3n3) acid. Only a borderline significant positive association was observed for total SFAs in CE (multivariate RRs across quintiles: 1.00, 1.15, 1.40, 1.62, 1.32; p=0.07). Total PUFAs or monounsaturated FA were not independently associated with CHD.

Our study found a weak positive association of SFAs with incident CHD. Our findings also confirm that FA metabolism in the body, such as the activity of Δ-5 desaturase, which converts dihomo-γ-linolenic acid to arachidonic acid, may affect the development of CHD.