Abstract
BACKGROUND AND AIMS: It is not known if ultrasound carotid plaque features are associated with cardiovascular disease (CVD) risk factors or if they predict future CVD events. METHODS: We measured total carotid plaque area (TPA) and grayscale plaque features (grayscale median, black areas, and discrete white areas) by B-mode carotid ultrasound among 2205 participants who participated in the first (baseline) visit of the Multi-Ethnic Study of Atherosclerosis. Multivariable linear regression was used to examine relationships between ultrasound plaque features and CVD risk factors at baseline. Cox proportional hazards models were used to assess if TPA, grayscale features, and carotid plaque score (number of arterial segments with a plaque) could predict incident coronary heart disease and cerebrovascular disease events over a mean follow-up of 13.3 years. RESULTS: Participants were mean (standard deviation [SD]) 65.4 (9.6) years, 49% male, 39% White, 11% Chinese, 28% Black, and 22% Hispanic. Mean TPA 27.7 (24.7) mm(2), but no grayscale plaque features, was associated with CVD risk factors. In fully adjusted models, TPA but no grayscale features predicted incident coronary heart disease (CHD) events (HR 1.23; 95%CI 1.11-1.36; p<0.001), however, C-statistics for CHD were similar to carotid plaque score but less than for coronary artery calcium (CAC) scoring. Neither TPA nor grayscale features independently predicted cerebrovascular events. CONCLUSIONS: In middle-aged adults free of known cardiovascular disease, TPA but not grayscale plaque features was associated with CVD risk factors and predicted incident CHD events. For CHD, prediction indices for TPA were similar to carotid plaque score but less than for CAC.