Respiratory gas exchange data were collected from 77 men greater than 6 months after acute myocardial infarction. Maximal exercise was performed on an ergometer cycle programmed for a ramp protocol of 15 W/min. The gas exchange anaerobic threshold (ATge) was determined by analysis of the carbon dioxide elimination (VCO2) vs oxygen consumption (VO2) curve below a respiratory exchange ratio of 1.00 using a computerized algorithm. This value was estimated at the inflection of VCO2 from a line with a slope of 1 which intersects the VCO2 vs VO2 curve. The relation of the ATge to the lactate acidosis threshold was studied in 29 patients. The reproducibility of the ATge method was studied in 77 patients. Mean (+/- standard deviation) VO2 for the ATge was 905 +/- 220 vs 866 +/- 299 ml/min for the lactate acidosis threshold (r = 0.86, p less than 0.001). Mean VO2 at the ATge for test 1 was 968 +/- 225 vs 952 +/- 217 ml/min for test 2 (r = 0.71, p less than 0.001). Mean peak VO2 was 1,392 +/- 379 vs 912 +/- 202 ml/min at the ATge (r = 0.76, p less than 0.001). Results demonstrate that this ATge method correlates well with the lactate acidosis threshold, is reproducible, and should be useful as an objective measure of submaximal exercise performance.