ABSTRACT
The Bypass Angioplasty Revascularization Investigation (BARI) randomized 1,829 patients to percutaneous transluminal coronary angioplasty (PTCA) or coronary artery bypass grafting (CABG). Clinical site angiographers categorized lesions of > or = 50% diameter stenosis (n = 4,977) as clinically significant (86.4%) or nonsignificant (13.6%), and as favorable or nonfavorable for PTCA or CABG. More lesions were considered favorable for revascularization by CABG than by PTCA (91.5% vs 78.4%; p <0.001), particularly in the subgroup of 99% to 100% lesions (77.6% for CABG vs 21.9% for PTCA, p <0.001). Lesion features, characterized by the BARI core laboratory, were correlated with clinical site angiographers' assessment of clinical importance and suitability for PTCA or CABG. By multivariate analysis, positive predictors of clinical importance for 50% to 95% stenoses were greater stenosis severity, more jeopardized myocardium, larger reference diameter, and proximal vessel location. For 99% to 100% occlusions, predictors were shorter duration of occlusion and more jeopardized myocardium. PTCA suitability for 50% to 95% stenoses was inversely related to lesion length, ostial location, location on a bend, difficult access, and age, and was directly associated with greater Thrombolysis in Myocardial Infarction (TIMI) trial flow rate and more jeopardized myocardium. Predictors of PTCA suitability for 99% to 100% lesions were a lower American College of Cardiology/American Heart Association class and higher TIMI grade. Predictors for 50% to 95% stenoses were more jeopardized myocardium, larger reference diameter, and more proximal vessel location, and for 99% to 100% occlusions, more jeopardized myocardium and shorter duration of occlusion. Suitability for PTCA depended on lesion potency (<99%) and multiple morphologic characteristics that contrasted with the few angiographic features that adversely affect CABG suitability.
