The effectiveness of a cardiovascular stent depends on many factors, such as its ability to sustain the compression applied by the vessel wall, minimal longitudinal contraction when it is expanded, and its ability to flex when navigating tortuous blood vessels. The long-term reaction of the tissue to the stent is also device dependant; in particular some designs provoke in-stent restenosis (i.e., regrowth of the occlusion around the stent). The mechanism of restenosis is thought to involve injury or damage to the vessel wall due to the high stresses generated around the stent when it expands. Because of this, the deflection of the tissue between the struts of the stent (called prolapse or "draping") has been used as a measure of the potential of a stent to cause restenosis. In this paper, uniaxial and biaxial experiments on human femoral artery and porcine aortic vascular tissue are used to develop a hyperelastic constitive model of vascular tissue suitable for implementation in finite-element analysis. To analyze prolapse, four stent designs (BeStent 2, Medtronic AVE; NIROYAL, Boston Scientific; VELOCITY, Cordis; TETRA, Guidant) were expanded in vitro to determine their repeating-unit dimensions. This geometric data was used to generate a finite element model of the vascular tissue supported within a repeating-unit of the stent. Under a pressure of 450 mm Hg (representing the radial compression of the vessel wall), maximum radial deflection of 0.253 mm, 0.279 mm, 0.348 mm and 0.48 mm were calculated for each of the four stents. Stresses in the vascular wall were highest for the VELOCITY stent. The method is proposed as a way to compare stents relative to their potential for restenosis and as a basis for a biomechanical design of a stent repeating-unit that would minimize restenosis.