We assessed the utility of short-term (30 s) maximal isokinetic cycle ergometry as an additional method of investigating the limitations to exercise in 33 carefully selected patients with documented coronary artery disease. The technique proved safe and reproducible in these patients. In relation to normal standards, performance was better in the maximal isokinetic cycle ergometer test (peak power = 819 +/- 116 W; average power = 532 +/- 72 W; total work = 13.1 +/- 2.1 kJ; 95-101% of predicted) than in the progressive incremental exercise test (VO2 = 1.80 +/- 0.37 l.min-1; power output = 919 +/- 165 kpm.min-1; 70-80% of predicted). Beta blockade did not affect maximal performance during either isokinetic or progressive incremental cycling, although maximal heart rate was significantly lower during both tests in patients on beta blockade. Power output in the progressive exercise was not as strongly related to the indices measured during the 30 s isokinetic test (r = 0.59-0.63) as it was in previous studies of healthy individuals (r = 0.89). The ability to detect individual variations in short-term exercise capacity measured with maximal isokinetic cycle ergometry may have significant potential value 1) as an additional method of determining the limitations to exercise and 2) when executing an exercise prescription in patients with coronary artery disease.