Peak oxygen consumption during progressive exercise is of clinical relevance in the functional evaluation of the cardiac patient. The use of cardiopulmonary exercise testing for the evaluation of the efficacy of therapeutic intervention requires that the methods used yield reproducible results. This study compared the results of two consecutive, symptom-limited, maximal exercise tests in 170 men following confirmed myocardial infarction. On-line, real-time respiratory gas exchange was measured on a breath-by-breath basis. The data were processed by the system using a 9 s moving average filter and the peak values were determined as averaged over a representative 20-s interval during the final 1 min of the test. The mean (+/- SD) total exercise times for the two tests were 635 (+/- 109) vs. 652 (+/- 112) (r = 0.946). The mean (+/- SD) peak VO2 values were 1480 (+/- 337) vs. 1495 (+/- 350) ml min-1 (r = 0.923). Performance could not be predicted by routine assessment of infarct type or size. This study demonstrates that maximal cardiopulmonary exercise testing in men following myocardial infarction yields highly reproducible results.