Dietary glycaemic index (GI) in relation to metabolic risk fators in a Dutch population at risk for diabetes and cardiovascular disease

Background: Little information is available regarding the association between the glycemic index (GI) and the consumption of individual foods in the general Dutch population. Furthermore, the association between GI and metabolic factors is not fully elucidated yet. Objective: To examine the association of GI with food intake and levels of metabolic factors including glucose, insulin and lipid profiles and an indicator of low –grade inflammation, CRP. Research design and methods: We analysed data from the CODAM study and the Hoorn follow-up study, both consisting of a sample of the general population over - represented by subjects with impaired glucose metabolism and type 2 diabetes. Cross - sectional associations were assessed using Spearman correlation and multiple linear regression. Results: The combined study population included 976 subjects who had a mean (SD) GI of 58 (3.7). GI was most strongly correlated with the consumption of dairy products (energy-adjusted r (radj)=-0.62), fruits (radj=-0.41), cereals (radj=0.39) and potatoes (radj =0.38). These four groups together explained 65% of the variation in GI. After adjustment for potential confounders including study, age, gender, BMI, smoking, total energy intake and dietary fibre etc., GI was positively associated with fasting insulin (P=0.02), HOMAIR (P=0.01) and CRP (P=0.01). A positive association between GI and fasting glucose was found but among never smokers only (P=0.002). Conclusion: In this Dutch population, a low GI diet, high in dairy and fruits but low in potatoes and cereals, is associated with a favourable diabetic and cardiovascular risk profile as indicated by low levels of fasting insulin, HOMA -IR and CRP.