BACKGROUND AND OBJECTIVES: Nitroglycerin-mediated arterial dilation (NMD) was shown to be preserved in most previous studies, and this is possibly due to using a single high dose of nitroglycerin (NTG), which causes maximal arterial dilation. We sought to evaluate the clinical factors of flow-mediated dilation (FMD) and NMD at different doses of NTG in the patients with coronary artery disease (CAD). SUBJECTS AND METHODS: Thirty-two consecutive patients (mean age: 61 years old, 18 males) with angiographically proven CAD underwent FMD and NMD at total cumulative doses of 25microgram, 175microgram and 325microgram with using high-resolution ultrasound for the imaging. RESULTS: The FMD, NMD (25microgram), NMD (175microgram) and NMD (325microgram) were 4.72+/-1.82%, 7.08+/-3.02%, 13.33+/-6.14% and 15.89+/-7.24%, respectively (p<0.001 compared with each other). Univariate analysis showed that the FMD is associated with the serum homocysteine level, the NMD (25microgram) is associated with the body mass index, the NMD (175microgram) is associated with the fasting blood sugar and the ejection fraction, and the NMD (325microgram) is associated with the fasting blood sugar, while there was no significant difference of the FMD and NMD according to the presence of CAD risk factors. Multivariate analysis disclosed that the independent factors of FMD were the serum homocysteine and triglyceride levels, and those of NMD (25microgram) were hypertension, a low ejection fraction and severe coronary angiographic findings, while there was no independent factor for NMD (175microgram) and NMD (325microgram). CONCLUSION: This study suggests that hypertension, a low ejection fraction and significant stenotic coronary lesion may be associated with endothelium-independent smooth muscle dysfunction at low dose NTG, while the serum homocysteine and triglyceride levels are associated with endothelium-dependent endothelial dysfunction in the patients with CAD. Using low-dose NTG is important when measuring the NMD.