We aimed to prospectively assess the optimal cutoff value for a minimum lumen cross-sectional area (CSA) on a 64-slice multidetector computed tomography (MDCT) compared with an intravascular ultrasound (IVUS). In 39 patients with 43 stents, the minimum lumen diameter, stent diameter, diameter stenosis, minimum lumen CSA, stent CSA, and area stenosis at the narrowest point were measured independently on 64-slice MDCT and IVUS images. For the assessment of diameter and CSA, 64-slice MDCT showed good correlations with IVUS (r = 0.82 for minimum lumen diameter, r = 0.66 for stent diameter, r = 0.79 for minimum lumen CSA, and r = 0.75 for stent CSA, respectively, P < 0.0001). For the assessment of diameter and area stenoses, a 64-slice MDCT showed good correlations with IVUS (r = 0.89 and 0.91, respectively, P < 0.0001). The overall sensitivity, specificity, positive predictive value, and negative predictive value to detect in-stent area restenosis (≥ 50 % area stenosis) of a 64-slice MDCT were 77, 100, 100, and 91 %, respectively. The cutoff value of a 64-slice MDCT, determined by receiver operator characteristic (ROC) analysis, was 5.0 mm(2) with 76.5 % sensitivity and 92.3 % specificity for significant in-stent area restenosis; the area under the ROC curve was 0.902 (P < 0.0001). A good correlation was found between a 64-slice MDCT and the IVUS, regarding the assessment of diameter and area stenoses of coronary stents in selected patients implanted with stents of more than 3 mm in diameter. Optimal cutoff value for the minimum lumen CSA of coronary stents on the 64-slice MDCT is 5 mm(2) to predict a CSA of 4 mm(2) on IVUS.