High radial force has become an essential feature of new coronary stents. These stents are expected to exhibit sufficient radial force and minimum radial recoil when deployed in atheromatous lesions of various morphologies. The aim of this study was to compare the radial force of 17 coronary stents in vitro (Anglodynamics, ACS Multilink, AVE Micro II and GFX, Bard XT, Biocompatibles PC, Cook GR II, Cordis Crossflex, Hexacath Freedom, Johnson & Johnson PS 153'PS 154, and Crown, Medtronic Wiktor and BeStent BEL-15, Saint-Come SC 1616, Scimed Nir 7 and 9). Two in vitro mechanical tests were performed to assess the stent radial force. The first test measured the deformation of coronary stents in a V-stand by using a deformation controlled by a dynamometer based on a longitudinal generatrix. The precision was of 1 micron. This comparative test highlighted the appreciable variation in behavior of the range of tested stents in terms of resistance to local compression. The line graphs obtained show a very small (and in some cases nonexistent) purely elastic behavior area in the stents, with this result reflecting a small (<0.4 Newtons) range of forces applied to all stents. Above this value, the stents did not return to their size before compression and the line graph remains approximately linear for a long period. It was therefore possible to distinguish between two families of stents on the basis of the presence or absence of this elastic area. In the second test, the stent was deployed in a 3.0-mm elastic tube and a pressure gradient created between the interior and exterior of the tube. The precision was of 50 microns. A typical line graph of the pressure-diameter relationship was recognizable for a given stent. No difference in behavior between the different coronary stents was noted up to 0.3 x 10(5) Pa; subsequently, after 0.3 x 10(5) Pa, sizable deformation differences were visible between the least resistant stent, the BeStent BEL-15, and the most resistant stent, the Crossflex.