In 41 consecutive patients with 49 stenoses of aorto-coronary venous bypass (ACVB) grafts percutaneous transluminal angioplasty (PTA) was attempted. PTA was successful, i.e., the percent area stenosis was reduced by > 20% to < 70% (quantitative measurement with a precision magnifying lens from two orthogonal angiographic views) in 46 stenoses (94%) of 38 patients (93%). In 35 patients (92%) with 42 stenoses control coronary angiography was performed after a mean interval of 189 +/- 186 days. Recurrence, defined as an increase of percent area stenosis to greater-than-or-equal-to 70%, was found in nine stenoses (21%) of nine patients (26%). Recurrence correlated with a stenosis length > 10 mm before PTA (5/8 vs 4/32 stenoses; p < 0.01). In recurrent stenoses, the average diameter of the grafted native coronary artery was significantly smaller than in recurrence-free stenoses (1.92 +/- 0.52 mm vs 2.45 +/- 0.50 mm; p < 0.01). Clinical data were collected from all 38 patients with successful PTA after an average of 30 +/- 17 months following PTA. In this interval, 11 patients had undergone reangioplasty and eight patients were re-operated; in addition, there were three cardiac deaths. In the 27 surviving patients without reoperation (71%), angina pectoris had improved from a mean of 3.0 +/- 0.7 before PTA to 1.8 +/- 1.0 (CCS-classification) (p < 0.001). In 19 of the 27 patients (70%) the exercise stress test was negative, in contrast to only three patients (11%) before PTA. Thus, in the majority of patients PTA of ACVB-graft stenoses improves quality of life. The caliber of the grafted coronary artery and the length of the critical stenosis are parameters for recurrence of stenosis and should be considered in the selection of patients for PTA of ACVB-graft stenoses.