Background: Restenosis is a major limitation of angioplasty. In this analysis we assessed the effects of lesion site and quality of dilatation on restenosis rate in the Coronary Angioplasty versus Bypass Revascularization Investigation population who underwent angioplasty. Methods: The angiographic quality of the successful angioplasty revascularization at each site was assessed, and the subsequent restenosis rate was determined. Restenosis was defined by the need for a second angioplasty at the initial site or by surgical coronary bypass grafting at or distal to the initial site. Results: The restenosis rate was unaffected by quality of dilatation but was significantly more common in the proximal left anterior descending artery compared with other sites, whether or not optimal dilatation had been achieved (relative risk 2.0 and 1.9, respectively). Conclusion: Revascularization strategies in multivessel disease should consider the presence or absence of a proximal left anterior descending artery target. Furthermore in studies in which restenosis is an outcome of interest, an allowance should be made for the distribution of target disease. (Am Heart J 1998;135:703-8.)