Risk Factors for Progression of Common Carotid Atherosclerosis:
The ARIC Study 1987-1998
Intima-media thickness (IMT) of the carotid arteries, as measured by B-mode ultrasound, is a marker of atherosclerosis, and as such has been shown to be positively associated with both prevalent and incident coronary heart disease (CHD) and incident stroke, and with risk factors for CHD. We examined the association of baseline risk factors or change in those risk factors with change in carotid IMT over an average nine years’ follow-up (1987-98) in the bi-ethnic population-based cohort of the Atherosclerosis Risk in Communities (ARIC) study (45-64 years old at baseline, n = 15,792). We found common carotid IMT to progress at 6 - 10 micrometers/ year, which is similar to other studies. We found, similar to other large epidemiologic studies, sizeable measurement error in IMT, which can cause large bias in estimating risk factor associations with change in IMT if adjusting for baseline IMT using standard methods. We concluded that adjustment for the potential confounding of baseline IMT, with proper correction for measurement error, generally had only a small effect on estimates of the risk factor associations. Either not adjusting for baseline IMT at all, or if doing so also correcting for measurement error by regression calibration, we found statistically significant (p < 0.05) associations of IMT change with baseline diabetes, current smoking, HDL cholesterol, Lp(a), pulse pressure, white blood cell count, and fibrinogen, all in the expected direction except Lp(a). The associations were of the same order of magnitude as anticipated from our cross-sectional findings. The association with baseline triglycerides was marginally significant, while those for systolic and diastolic blood pressure, hypertension, BMI, and LDL cholesterol were not significant. We examined the associations of IMT change with change in HDL cholesterol, LDL cholesterol, triglycerides, and BMI and with onset of diabetes and hypertension, and found statistically significant associations with all but HDL cholesterol. Modifiable, established CHD risk factors are associated with the rate of change of subclinical atherosclerosis, as indexed by carotid IMT.