Plasma glucose within the normal range is not associated with carotid atherosclerosis - Prospective results in subjects with normal glucose tolerancefrom the Bruneck Study

OBJECTIVE - There is substantial evidence that glucose intolerance is assoc iated with an increased risk of cardiovascular disease. However, it is not well established whether plasma glucose is independently related to atheros clerosis when glucose tolerance is normal and, if so, to which stage of the complex atherosclerotic process. RESEARCH DESIGN AND METHODS - We prospectively examined the status of carot id arteries in 625 subjects aged 40-79 years who were randomly selected fro m the general population and had normal glucose tolerance (according to Wor ld Health Organization criteria) both at baseline and at 5 years of follow- up. All subjects had high-resolution echo-duplex evaluation of the common a nd internal carotid arteries (eight regions of interest on both sides) in 1 990 and 1995 to detect the change in carotid status over time. The occurren ce of new plaques in previously normal segments was termed "incident nonste notic" or "early atherosclerosis," and the occurrence of stenosis in >40% o f previously normal segments was termed "incident stenotic" or "advanced at herosclerosis." In addition, we evaluated the changes in the atherosclerosi s score (the sum of all plaques) during the follow-up, and we measured inti mal-medial thickening (IMT) in the common carotid artery in 1995. In all su bjects, several candidate risk factors were assessed: sex, age, BMI, waist- to-hip ratio, glucose, HbA(1c), insulin, urate, lipids, apolipoproteins Al and B, blood pressure, lipoprotein(a), fibrinogen, antithrombin III, factor V Leiden mutation, ferritin, leukocyte count, smoking, alcohol intake, phy sical activity, and socioeconomic status. Fasting plasma glucose (FPG), pla sma glucose 2 hr after the glucose load (2-h PG), and HbA(1c) concentration s in 1990 and 1995 were averaged in each subject to obtain an estimate of l ong-term glucose exposure of the arterial wall. RESULTS - Linear or logistical regression analyses indicated that neither b aseline glucose and HbA(1c) levels nor mean FPG, mean 2-h PG, or mean HbA(1 c) in 1990 and 1995 were independently related to IMT, a 5-year change in t he atherosclerotic score, incident nonstenotic (early) atherosclerosis, or incident stenotic (advanced) atherosclerosis. Likewise, subjects with FPG l evels above the median and subjects in the new category of "impaired fastin g glucose" did not have an increased occurrence or progression of atheroscl erosis. All results were consistent before and after adjustment for other v ascular risk factors and possible confounders. CONCLUSIONS - These results suggest that plasma glucose levels within the n ormal range (<7.8 mg/dl both at FPG and 2-h PG) are not independently relat ed to any stage of atherosclerosis.