Between January 1991 and June 1993, a total of 128 patients underwent coronary artery bypass grafting employing multiple autologous arterial conduits, including 157 internal mammary arteries, 69 inferior epigastric arteries, 44 gastroepiploic arteries, and 72 radial artery grafts. Their mean age was 61.4 years (range 29 to 82 years). The patients were divided into 2 groups: group A, consisted of 63 patients (mean age 60.3 years), in whom multiple arterial conduits were used exclusively (no vein grafts); group, included 53 patients (mean age 62.7 years) in whom, in addition to multiple arterial conduits, 89 saphenous vein grafts were used concomitantly. The mean number of grafts was 3.1 and 3.7, for groups A and B, respectively. The preoperative leftventricular function, and the prevalence of unstable angina, a recent myocardial infarction, and diabetes, were not significantly different between both groups. Our series included 11 ''redo'' operations (8 in group A, and; 3 in group B). There were 6 early deaths (4.7% mortality) (1 in group A, and 5 in group B), and 4 perioperative myocardial infarctions (1 in group A, 3 in group B). During a mean follow-up of 12.3 months (range 1 to 28 months) there were no late deaths or reoperations in any group. All patients in group A are free of symptoms. In group B, 2 patients have recurrent angina, and 1 had a late myocardial infarction, in the distribution of a vein graft. A myocardial SPECT scan with exercise revealed new perfusion defects in 4 of 49 patients (1 in group A, 3 in group B), studied 1 year after surgery. In summary, multiple arterial conduits can be used safely for coronary bypass surgery, and ''redo'' operations, with no increased surgical risk, and excellent results after 2 years. A long-term follow-up will be necessary to determine if a complete myocardial revascularization without vein grafts will effectively decrease the incidence of recurrent symptoms, late events, or the heed for reoperation.