Back ground : Multi-detector row Computed Tomography (MDCT) permits non-invasive visualization of the coronary arteries. 64-slice MDCT is improving accuracy in coronary plaque detection and offers a better opportunity for plaque characterization, but its ability of exact separation of lumen, plaque, and vessel wall has not been evaluated. Objective : We evaluated the ability of 64-slice MDCT, compared with intravascular ultrasound(IVUS), to detect inner vessel wall calcified and noncalcified coronary atherosclerotic plaque volume in patient with significant coronary artery stenosis. Method and result : In 15 patients, the contrast-enhanced MDCT (0.625-mm collimation, 350-ms rotation) and IVUS were performed. In signigicant coronary stenosis ( > 50 % stenosis and >2.5 mm in diameter ), 50 segments were obtained by IVUS and MDCT. For calcified plaque, MDCT sensitivity of exact separation between vessel, surrounding tissue plaque, and lumen is 68 % (34 of 50) and MDCT overestimated plaque volume per segment as compared with IVUS (49 짹 18 mm쨀 versus 37 짹 13 mm쨀). For uncalcified plaque, MDCT substantially underestimated plaque volume per segment as compared with IVUS (34 짹 13 mm쨀 versus 43 짹 14 mm쨀). Conclusion : Our data suggests that 64-slice MDCT has the technical limitations to prevent an exact separation of lumen, plaque, and vessel wall. As a consequence, we observed a high variability for 64-slice CT measurements and an only moderate concordance to IVUS measurements. In calcified plaque, MDCT Show a marked tendency to overesitmate plaque volume and weakly predictable to detect inner vessel wall calcified plaque. In uncalcified plaque, MDCT is disposed to underestimate plaque volume as compared with IVUS.