Abstract P064: Dietary Glycemic Index and Glycemic Load and Risk of Coronary Heart Disease in a Prospective Study Among US Male Health Professionals

Abstract

Introduction: The associations between dietary glycemic index (GI) and glycemic load (GL) with subsequent risk of coronary heart disease (CHD) are inconclusive among men. The association is further complicated by the potential biological interactions between the carbohydrate quality of the diet and factors that may influence underlying insulin resistance. Hypothesis: We hypothesized that long-term exposure to a diet with high GI and GL would be associated with increased CHD risk in a cohort of US men, and the association would be further modified by fiber intake, alcohol consumption and BMI. Methods: We included 37,736 men aged 40–75 years from the Health Professional Follow-up Study, with no previous diagnosis of CHD, cancer, or type 2 diabetes. We confirmed 3,121 total incident CHD cases during 22 years of follow-up. Cox proportional hazard models were used to adjust for covariates. Results: After adjusting for lifestyle and dietary covariates, the hazard ratio (RR) and 95% confidence intervals comparing men in the highest vs. the lowest quintile was 1.16 (1.04, 1.31; p for trend=0.03) for dietary GI, and 1.09 (0.94, 1.27; p for trend =0.16) for GL. We found a significant effect modification by fiber intake (p=0.02); The associations between GL and CHD risk were strongest among participants in the lowest tertile of fiber intake (RR= 1.00, 1.00, 0.93, 1.11 and 1.16 with increasing quintiles of GL; RR=1.00, 0.68, 0.79, 0.73 and 0.76 for participants in the highest tertile of fiber intake). The association between GL and CHD was stronger among men with body mass index (BMI) greater than 25 kg/m2 than normal weight men, even though the test for interaction was only marginally significant (RR=1.00, 1.03, 1.05, 1.10, and 1.16 for increasing quintles of GL among men with BMI≥25 kg/m2 and RR= 1.00, 0.92, 0.96, 1.09 and 1.04 for men with BMI<25 kg/m2, p for interaction=0.09). Alcohol intake did not modify the association of GL with CHD (p for interaction=0.44). Conclusion: We observed a significantly increased risk of CHD with high GI diet and a modestly elevated association between GL and CHD among men with low fiber intakes or who were overweight or obese. No effect modification by alcohol was observed, but we did find that the association between GL and CHD was more pronounced among those with lower fiber intake or higher BMI.