Several polypeptides with the potential to cause blood vessel growth (angiogenesis) via endothelial cell proliferation and migration have been sequenced and synthesized during the last few years. Our ultimate goal is to utilize these angiogenic agent(s) to facilitate myocardial revascularization in patients with coronary heart disease. The specific purpose of this investigation is to utilize fibroblast growth factor (FGF) to effect myocardial angiogenesis in a canine model, and to direct this process to ameliorate myocardial ischemia. This polypeptide, available to us in large quantities, is a potent stimulator of angiogenesis in vitro. In our experimental model, the left circumflex coronary artery (LCX) of dogs is occluded gradually over a 2 to 3 week period by an ameroid constrictor applied to the proximal vessel. As a result of LCX occlusion, the territory it supplies becomes dependent on collateral vessels, but myocardial infarction generally does not occur as long as the occlusion is gradual. After 5 weeks, regional myocardial blood flow will be quantitated wi h radiolabeled microspheres under basal conditions, and during pharmacologically induced coronary vasodilatation. Dogs will then be randomized to receive FGF or placebo directly into the LCX. After 5 weeks, resting and maximal myocardial blood flow will again be quantitated. Thus, collateral function can be compared before and after treatment in both groups. Vessels will be ex in morphometrically and various hematologic, biochemical, and immunologic meters will be assessed in the 2 groups. Dogs will be monitored for any potential adverse effects of FGF.