Peripheral vascular endothelial function testing for the diagnosis of coronary artery disease

Background Abnormalities in endothelium-dependent vasodilation may be detected in arteries before the development of overt atherosclerosis, and their presence may predict stress-induced ischemia as assessed by ST-segment depression and/or perfusion defects. Brachial artery ultrasound during reactive hyperemia is a noninvasive method of assessing peripheral vasomotion, measured by flow-mediated vasodilation (FMD). The purpose of the current study was to assess whether endothelium-dependent FMD of the brachial artery, by ultrasound imaging, predicts the presence of angiographically assessed coronary artery disease (CAD). Methods One hundred ninety-eight in-hospital patients (age, 59 ± 9 years; 78 women) with chest pain syndrome and without previous myocardial infarction or revascularization procedures were enrolled in the present study. All of the patients, at testing time, were not receiving nitrate therapy and underwent, on different days, coronary angiography and endothelium-dependent FMD testing of the brachial artery by high-resolution ultrasound. The result of the flow-mediated dilation (%FMD) is defined as the percent change in the internal diameter of the brachial artery during reactive hyperemia related to baseline. A coronary vessel was considered to have a significant obstruction if its diameter was narrowed by 50% or more on quantitative computer-assisted analysis. A prognostically validated angiographic Duke score (from 0 = normal to 100 = severe left main disease) was calculated. Results The %FMD was lower in patients with (n = 69) compared with those without (n = 129) CAD (4.64% ± 4.36% vs 7.39% ± 5.68%; P = .01). By multivariate analysis, the %FMD (P = .01; odds ratio [OR], 1.13; 95% confidence interval [CI], 1.05 to 1.23), male sex (P = .01; OR, 3.47; 95% CI, 1.64 to 7.36), and cigarette smoking habit (P < .01; OR, 4.00; 95% CI, 2.50 to 6.35) were independent predictors of CAD. %FMD was poorly albeit significantly correlated with the severity of CAD (%FMD Duke score, P < .01, r = −0.25). The receiver operator characteristic curve showed the %FMD optimal cutoff value as ≤8.84, with sensitivity of 90%, specificity of 37%, negative predictive value of 90%, and positive predictive value of 43%. Conclusions In patients with chest pain, a depressed FMD of the brachial artery was a sensitive indicator of CAD, but it showed poor specificity, and it appeared to be unable to predict both the extent and the severity of angiographically assessed CAD.