Since randomized trials comparing the results of coronary artery bypass surgery (CABG) to the results obtained with percutaneous transluminal coronary angioplasty (PTCA) will not be completed for some time, the objective of this study was to compare available exercise test data gathered before and after both of these revascularization procedures. The design was a retrospective assessment of veterans being treated at Long Beach VA Medical Center, Long Beach, California. All patients identified as having undergone exercise testing before and after PTCA and CABG were considered for selection according to medication status and timing of exercise tests. Twenty-eight patients formed the CABG group and 38 patients formed the PTCA group. Since the timing of the tests was according to usual clinical practices, the exercise tests were performed an average of 2.5 weeks after PTCA and 5 months after CABG. The medication status was comparable, but there were significantly more patients with multivessel disease in the CABG group than in the PTCA group (P < 0.001). CABG was found to be significantly more effective in decreasing sings (P < 0.01) and symptoms (P = 0.04) of ischemia than PTCA but there was no significant differences in estimated aerobic capacity (1.8 METs vs 2.2 METs, NS). These results indicate a more effective decrease in signs and symptoms of ischemia after CABG compared to PTCA. Though no clinical selective bias could be indentified, these findings remain hypothetical until the results of randomized studies are available.