Coronary angioplasty was attempted in 1,074 consecutive patients, including 169 patients with total (100%) occlusion (group 1), 102 patients with functional total (99%) occlusion (group 2) and 711 patients with conventional (70 to 95%) stenoses (group 3). After exclusion of 92 patients with acute myocardial infarction, the mean age of the patients was 57 +/- 12 years, including 727 men (74%) and 255 women (26%). Although there were no differences between groups with respect to anginal symptoms or extent of coronary artery disease, the primary success rate (by lesion) varied according to lesion severity, and was 63%, 78% and 90% for groups 1, 2 and 3, respectively (p less than 0.001). Only 3 patients (1%) with chronic coronary artery occlusion (groups 1 and 2) required emergency surgery because of side-branch occlusion in 2 patients and guidewire fracture in 1 patient, whereas emergency surgery was required in 14 patients (2%) in group 3. Long-term follow-up averaging 19 +/- 11 months was available for 95% of patients. Although the combined likelihood of death or nonfatal myocardial infarction was below 6% for each group at 2 years (difference not significant), the likelihood of death, nonfatal infarction, coronary bypass surgery or repeat percutaneous transluminal coronary angioplasty was higher (41%) in group 1 than in groups 2 (28%) or 3 (28%) (p less than 0.001). Thus, coronary angioplasty can be performed safely and effectively in patients with chronic total occlusion, although neither the primary success rate nor the long-term follow-up are as favorable as in patients with conventional stenoses.