Since April in 1988 coronary artery bypass grafting (CABG) was performed in 177 patients and the clinical results were compared in SVG-group (n = 37) in which only saphenous vein graft (SVG) was used as a graft conduit and AG-group in which one arterial graft (AG-1 group, n = 92) or more than two arterial grafts were used (AG-2 group, n = 48). In SVG-group the age was older than that in AG-group, while coronary vessels were most involved in AG-2 group and the mean number of distal anastomosis increased more in AG-group. The total pump time and aortic cross clamping time showed no significant differences among the groups. Furthermore, the occurrence of perioperative myocardial infarction, requirement of intra-aortic balloon pumping, frequency of re-chest open for hemorrhage, or sternal infection showed no significant differences among the groups. Postoperative hospital deaths were in four (three cardiac deaths) of SVG-group and four (one cardiac and two graft-versus-host disease) of AG-group. These results suggest that the use of AG does not increase surgical risk and AG can be positively used for CABG.