Training effects on peak oxygen consumption (VO2), specific to the mode of movement, are well-known in exercise training of young, healthy adults. However, these specific training effects were never studied in patients with coronary artery disease, but may be important in the evaluation of training effects of cardiac rehabilitation programs. Exercise training programs dominated by, for example, cycling might improve peak VO2, measured during cycling, more than during treadmill testing. Therefore, the effects of an exercise training program dominated by cycling and of a program with both cycling and walking/jogging during a 6-weeks cardiac rehabilitation program were evaluated on both cycle ergometer and treadmill. Male patients (aged between 35 and 70 years) with coronary artery disease (history of myocardial infarction and/or angina pectoris and/or coronary artery bypass surgery) were randomly assigned to either a program dominated by cycling (Group I: n=18,mean age 53+/-6.7) or a program with both cycling and jogging (Group II: n=20, mean age 48+/-9.1). Before and after the program peak VO2 was measured on both cycle ergometer and treadmill. At baseline peak VO2 on treadmill was significantly greater than on cycle ergometer in both groups. Peak VO2 (both cycle and treadmill) increased highly significantly during both programs; in group I the increase of peak VO2 on cycle ergometer was greater than on treadmill (respectively, 28.1% versus 18.8%; p<0.05), in contrast to group II (respectively, 22.8% and 16.6%; n.s.). As a result, the difference between peak VO2 on treadmill and cycle ergometer decreased significantly more during the program in group I (p<0.05). These results suggest specific training effects in patients with coronary artery disease and should be considered outcome assessment and exercise prescription of cardiac rehabilitation programs.