The occurrence of coronary artery disease following mediastinal radiation for malignancies has long been debated. However, the development of coronary pathology in young individuals following radiation who lack risk factors for atherosclerosis is highly suggestive of a cause-and-effect relationship. By far the most convincing pathologic changes are adventitial scarring and medial atrophy with severe intimal atherosclerotic disease consisting of necrotic core, fibrous tissue, and calcification. Initial clinical studies in patients with coronary atherosclerosis treated with intraluminal radiation following stenting hold great promise in the treatment and prevention of restenosis. There are little or no data, however, on long-term effects of intra-coronary radiation therapy in man. Therefore, it may be important to study the chronic effects of radiation in animal models in order to predict what is likely to occur in humans. We evaluated the effects of varying doses (0.15-23.0 microCi of 32P) of beta-particle-emitting radioactive stents in pig coronary arteries at 1 and 6 months. At 1 month, there were dose-dependent changes in the morphology of the intima and media. High activities (>3 microCi) caused medial necrosis with fibrin deposition in the media and intima, with interspersed red cells most marked in regions surrounding the stent struts. Only rare smooth muscle cells (SMCs) and inflammatory cells were seen away from the stent struts. In the intermediate (1 microCi) stent activity group, the neointima was expanded by SMCs and a proteoglycan-rich matrix with focal endothelialization of the luminal surface. Neovascular capillaries and extravascular red cells were present adjacent to stent struts. At low activities (<0.5 microCi), the neointima was significantly smaller than control stents and consisted of SMCs and matrix with complete endothelialization of the luminal surface. The neointimal cell density of the media and intima decreased with increasing doses of radiation. In pigs 6 months after radioactive stenting (activities ranging from 0.5-12 microCi 32P), >3.0 microCi radioactive stents induced marked neointimal thickening, with changes similar to atherosclerosis, consisting of necrotic debris containing cholesterol clefts surrounded by macrophage collections, fibrosis, and focal calcification. There was increased adventitial thickening in the radiated vs non-radiated arteries. The intermediate stent activity (1.0 microCi) also showed greater neointimal thickening (vs control stents) and consisted mostly of SMCs in a proteoglycan-rich matrix. At <1.0 microCi, there were minimal differences seen between radiated and control non-radiated stented arteries. The media was unevenly injured in all stent activities and varied from less than to significantly greater than controls. These data suggest that radiation-induced coronary atherosclerosis seen in man is partially simulated in normal porcine coronary arteries 6 months following high-dose beta-particle-emitting radioactive stent placement. There is greater fibrosis and thickness of the adventitia and focal attenuation of the media in man and severe luminal narrowing in pig coronary arteries treated with high doses of radiation. Only long-term clinical follow up and careful autopsy studies will determine if endoluminal or intra-arterial radiation is a viable means of reducing restenosis in man.