Predictive value of carotid artery ultrasonography for the risk of coronary artery disease

PURPOSE: To assess carotid intima-media thickness (IMT), carotid plaques, and cardiovascular risk factors in patients with suspected coronary artery disease (CHD) to determine their association and predictive value for CHD. METHODS: We performed duplex Doppler ultrasonography of the carotid arteries and coronary angiography or CT in 480 patients with suspected CHD, and investigated their personal and medical histories. Patients were then assigned to the CHD or the control group depending on the presence of coronary lesions. Ultrasonography was performed the morning after admission prior to any treatment, coronary angiography, or CT. RESULTS: Carotid plaques were mainly distributed in the common carotid artery bifurcation, with a significant difference between the CHD and control groups. Plaque incidence (80%) and IMT were significantly higher (P < .001 and P = .012, respectively) in the CHD (80% and 0.84 ± 0.21 mm) than in the control group (49% and 0.76 ± 0.18 mm). The factors significantly associated with CHD were introduced into a multivariate regression model. Male subject (OR = 1.569, 95%CI 1.004-2.453; P = .048) and plaque burden (OR = 0.457, 95%CI 0.210-0.993; P = .048) were significant predictors for CHD occurrence. The presence of carotid plaques performed significantly better than IMT and the Framingham risk score for predicting CHD lesions (P < .001 for both). CONCLUSIONS: CHD patients showed higher percentage of clinical (plaques) or subclinical (IMT) carotid artery wall change, and the presence of carotid plaques showed better predictive value than IMT and Framingham risk score for the presence of coronary artery lesions.