Finite element models, which can determine the degree of vascular injury caused to a vessel by a stent and hence ascertain the propensity of that stent design to cause instent restenosis, would be a valuable tool for both medical device companies and interventional cardiologists. The basis of this thesis is to establish a preclinical testing methodology for stents, which can determine the degree of mechanical injury caused to a vessel by a particular stent design. The testing technique will enable stents to be designed to prevent or minimise vascular injury and hence restenosis. To achieve this goal, uniaxial and equibiaxial tensile tests on cardiovascular tissue were carried out to develop suitable Mooney-Rivlin constitutive models for arterial tissue. Three-dimensional models of different stent designs were generated and placed in their expanded states within models of coronary arteries. The established Mooney-Rivlin models were used to represent the materials of the artery. The mechanical loading induced within the stented vessels by the different stent designs was calculated to determine the stent design least likely to cause excessive vascular injury and hence instent restenosis.

Fatigue tests were carried out on coronary arterial tissue to determine a measure of damage accumulation within stented vessels at the elevated stresses imposed by stents on coronary arteries. The measured rate of damage accumulation within the tissue was used as a stimulus for in-stent restenosis in a computational model of the vascular healing process. The model was based on the predominant mechanism of the in-stent restenosis, SMC migration and proliferation. The propensity of the different stent designs to cause in-stent restenosis was determined from the computational model of the restenosis mechanism.

The in-stent restenosis model successfially predicted the stent design that is least likely to cause occlusive in-stent restenosis, by comparison to clinical restenosis data. It therefore supports the use of this method to analyse new stent designs. This method could ultimately be used as a preclinical test to determine the optimum bare-metal or drug-eluting stent design to prevent in-stent restenosis.