Abstract P369: Higher Diet quality is Prospectively Associated with Markers of Endothelial Dysfunction: The Coronary Artery Risk Development in Young Adults Study
Introduction We examined the prospective association between diet quality and circulating soluble intercellular adhesion molecule 1 (sICAM-1) and p-selectin, higher values of both of which indicate endothelial dysfunction, a stage in the development of atherosclerosis. We hypothesized that higher diet quality relates inversely with these biomarkers of endothelial dysfunction.
Methods We studied 2,569 black and white men and women in the Coronary Artery Risk Development in Young Adults (CARDIA) study, aged 18-30 at year 0 (1985-86). Diet was assessed at years 0 and 7; sICAM-1 and p-selectin were assayed at year 15. Principal component analysis yielded a diet pattern score rich in refined grain, butter and red meat (meat diet), higher scores reflecting poorer diet quality, and one rich in plant foods (fruit and vegetable (FV) diet), higher scores reflecting better diet quality. Linear regression models predicted sICAM-1 and p-selectin from the diet pattern scores, adjusting for sex, race, study center, age, total energy, smoking status, educational attainment and physical activity and in a separate model additionally adjusted for BMI (all regressors average of year 0 and 7).
Results Concentrations of sICAM-1 and p-selectin (year 15 mean and standard deviation (SD)) were 153.4±44.1 ng/mL and 36.7±11.3 ng/mL, respectively. Both diet pattern scores were associated with sICAM-1 (table), most strongly the meat diet. The mean sICAM-1 was 17.3 ng/mL lower in the lowest than the highest meat diet quintile, amounting to an effect size of 39% of an SD for sICAM-1 (17.3/44.1 ng/mL). P-selectin was positively associated with the meat diet, but was not significantly associated with the FV diet. Associations were little affected by adjustment for BMI.
Conclusion The association of higher diet quality with biomarkers of endothelial dysfunction should be considered as a possible pathway through which diet may affect the development of cardiovascular disease.
|Quintile of Dietary Pattern Consumption|
|sICAM-1 (153.4±44.1 ng/mL)|
|Model 1||b (se)||0||3.5 (2.5)||10.2 (2.9)||17.5 (3.6)||17.3 (5.3)||0.001|
|Model 2||b (se)||0||3.0 (2.5)||8.4 (2.8)||15.4 (3.5)||16.5 (5.2)||0.0023|
|Model 1||b (se)||0||−1.8 (2.7)||−3.1 (2.9)||−5.8 (3.0)||−8.4 (3.3)||0.011|
|Model 2||b (se)||0||−1.5 (2.7)||−3.0 (2.8)||−5.6 (2.9)||−8.2 (3.3)||0.0066|
|P-selectin (36.7±11.3 ng/mL)|
|Model 1||b (se)||0||0.6 (0.7)||1.8 (0.8)||3.1 (0.9)||5.6 (1.4)||0.0018|
|Model 2||b (se)||0||0.5 (0.7)||1.6 (0.8)||2.9 (0.9)||5.6 (1.4)||0.0026|
|Model 1||b (se)||0||0.5 (0.7)||0.5 (0.8)||−0.1 (0.8)||−0.7 (0.9)||0.144|
|Model 2||b (se)||0||0.5 (0.7)||0.5 (0.8)||0.0 (0.8)||−0.6 (0.9)||0.1321|
Each row represents a separate regression model. Model 1 adjusted for race, sex, center, age, education, energy intake and physical activity, model 2 additionally adjusted for BMI. Each cell provides for a given quintile of the particular diet pattern score the number at risk (N) or the regression coefficient and standard error (b (se)) comparing mean sICAM-1 or p-selectin to its value in the lowest diet pattern score quintile.
- © 2012 by American Heart Association, Inc.