Background. The cardiac output-O2 uptake (VO2) relation, which is thought to be linear and predictable in normal humans, has not been clarified in cardiac patients. We evaluated the relation between cardiac output and VO2 during the incremental exercise test in patients with previous myocardial infarction. Methods and Results. Twenty-two patients (age, 58.1 +/- 8.0 years) with previous myocardial infarction. performed a symptom-limited exercise test on a cycle ergometer. VO2 was calculated from the expired gas analysis, and cardiac output was measured by a computerized cadmium telluride detector every 10 seconds during exercise. The ratio of increase in VO2 to the increase in work rate (DELTA-VO2/DELTA-WR) below and above the anerobic threshold (AT) was 11.1 +/- 3.6 and 11.1 +/- 2.9 ml/min/W, respectively, showing no significant difference. However, the ratio of increase in cardiac output to increase in work rate (DELTA-CO/DELTA-WR) below the AT was 50.1 +/- 26.6 ml/min/W and was significantly decreased to 11.8 +/- 25.3 ml/min/W above the AT (p = 0.0002). The decreased DELTA-CO/DELTA-WR above the AT primarily would be due to silent myocardial ischemia produced by exercise, as there was the presence of Tl-201 redistribution in 15 of 16 patients in whom myocardial Tl-201 scintigraphy with dipyridamole or exercise stress testing was evaluated. DELTA-CO/DELTA-Vo2, which has been reported to be approximately 5.5 in normal subjects, was only 4.4 +/- 2.6 at work rates below the AT and was decreased to 1.1 +/- 2.3 at work rates above the AT. Conclusions. The relation between cardiac output and Vo2 during exercise in patients with previous myocardial infarction differs profoundly from that reported in normal subjects. These findings must be considered when we noninvasively estimate the change in cardiac output during exercise by obtaining Vo2 in patients with coronary artery disease.