Background and purpose : Aortic plaque has been recognized as a potential source of embolism, for which diagnostic transesophageal echocardiography (TEE) is widely used in a clinical setting. TEE is not always performed in patients with definite embolic sources, including atrial fibrillation, because of the semi-invasive nature of the examination. Accordingly, the indications for TEE may vary among hospitals, resulting in underestimation of the clinical significance of aortogenic embolism. This study investigated the features predictive of aortic plaque on TEE, in order to establish systematic indications for TEE even in patients with an evident cardiogenic embolic source. Methods : The subjects comprised 55 patients with acute ischemic stroke who were examined by TEE. Plaque 24 mm in diameter, or showing ulceration or mobile components, was diagnosed as complex plaque (CP). Multivariate logistic regression analyses and receiver operating characteristic (ROC) curve analysis were used to reveal factors predictive of the presence of CP. Results : CP was demonstrated in 25 patients (45.5%). The maximum intima-media thickness (maxIMT) of the carotid arteries was significantly associated with CP (p = 0.005). The odds ratio and 95% confidence interval of the third tertile of maxIMT (1.8-6.1 mm) as a reference for the first tertile (0.6-0.8 mm) was 5.568 (1.131-27.410, p = 0.035). When the cutoff value of maxIMT revealed by TEE was set at 1.8 mm, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for diagnosis of CP were 52.0%, 80.0%, 50.0%, 6.7%, and 73%, respectively. Conclusions : MaxIMT of > 1.8 mm for the carotid arteries determined by TEE was predictive of aortic plaque as a potential embolic source. A prospective clinical evaluation of carotid maxIMT determined by TEE, irrespective of other embolic sources, is warranted.
Background : Stroke can be a devastating complication after coronary artery bypass grafting (CABG). Silent cerebral infarction (SCI) is common in elderly patients, and – like previous cerebrovascular disease (CVD)– is a risk factor for stroke. We examined the ultrasonographic features of the carotid arteries as predictors of SCI in patients undergoing CABG. Methods : Data for 633 CABG patients aged ≥ 60 yr were collected, and the patients were divided into three groups : control (n = 348), SCI (n = 175), and CVD (n = 110). We compared the images of carotid plaque, and the risk factors and incidence of stroke among the three groups, and calculated the probability of SCI. Results : In the SCI group, the plaque score (PS) was 7.7 ± 4.7, the incidence of carotid stenosis (> 75%) was 13%, that of echolucent plaque was 37%, that of cerebral artery stenosis (> 50%) was 26%, and that of severe aortic atherosclerosis was 19%, being almost equal to the data for the CVD group. The incidence of stroke was 6.4% in the CVD group, 4.5% in the SCI group, and 1.4% in the control group (P < 0.05). Univariate analysis revealed that 5 factors were correlated with SCI : PS > 5.1, stenosis > 75%, echolucent plaque, peripheral artery disease (PAD), and creatinine ≥ 1.5 mg/dL. The probability of SCI in patients with all 5 risk factors was 0.77. Conclusions : Ultrasonographic features combined with PAD and creatinine would be useful for prediction of SCI.