Clinical, physiologic, anatomic and procedural factors predictive of restenosis after percutaneous transluminal coronary angioplasty.

In a recent prospective double-blind placebo-controlled trial, a combination of aspirin and dipyridamole was not associated with a reduction in the rate of restenosis within the 1st 6 months after coronary angioplasty. The purpose of this study was to determine whether clinical, anatomic or procedural factors were predictive of the observed restenosis rates in that prospective trial. A total of 247 patients and 280 segments underwent follow-up angiography and quantitative coronary angiographic analysis between 4 and 7 months after coronary angioplasty. Two baseline clinical characteristics--angina class and duration of angina in months--were related to the rate of restenosis by univariate analysis. Patient-related stepwise logistic regression analysis identified severity of angina as the only clinical predictor of restenosis. Three univariate baseline anatomic characteristics--percent diameter stenosis before angioplasty, stenosis greater than 10 mm in length and calcific stenosis--and two early postangioplasty characteristics--residual percent diameter stenosis and residual mean pressure gradient--were predictive of restenosis. Of these, only two--length of stenosis and residual percent diameter stenosis--were independently related to restenosis by multivariate analysis and only the former is identifiable before the procedure. It is concluded that in prospective studies in contrast to retrospective studies, few clinical and anatomic factors appear to be predictive of restenosis after coronary angioplasty.