Background Intravascular ultrasound (IVUS) imaging can be used to optimize implantation of intracoronary stents; the variability of the measurements, however, remains unclear. Our aim in this study was to determine the intraobserver and interobserver variability of IVUS measurements after stent implantation. Methods Ninety-four patients underwent implantation of 100 Palmaz-Schatz stents in 98 lesions (79 de novo and 19 restenotic). IVUS measurements (3.5F, 30 MHz) of proximal and distal reference sections and of the smallest stent lumen were performed by 2 investigators. Results Intraobserver and interobserver correlations, respectively, were r = 0.96 and 0.93 for the proximal reference, r = 0.94 and 0.92 for the distal reference, and r = 0.97 and 0.97 for minimal stent lumen. Stent expansion (minimal lumen in the stent/mean reference area) showed a variability of r = 0.80 and 0.70. Taking a cutoff point of 90% for adequacy of stent expansion, observers agreed in only 77% whether the stent was adequately or inadequately expanded. Conclusions IVUS enables reproducible lumen measurements in stents and reference sections. The degree of stent expansion, however, underlies a high measurement variability that can lead to different therapeutic strategies. (Am Heart J 1999;137:368-71.)