To determine the effects of a moderately prolonged exercise on left ventricular systolic performance, 23 healthy male subjects, aged 18 to 51 yr (mean 37 yr) were studied. The subjects exercised first on a treadmill (brief exercise) and completed, on a separate day, a 20-km run. M-mode, two-dimensional, and Doppler echocardiography, as well as calibrated carotid pulse tracings, were obtained at rest and immediately on completion of both brief and prolonged exercise. Left ventricular systolic function was assessed by end-systolic stress-shortening relationships. Heart rate increased similarly after brief and prolonged exercise (+30%). Mean arterial pressure decreased from 99 +/- 7 to 92 +/- 8 mmHg (P less than 0.001) after prolonged exercise, but it remained unchanged after brief exercise. Left ventricular end-diastolic volume was decreased after prolonged exercise (130 +/- 23 vs. 147 +/- 18 ml at rest, P less than 0.01). Both ejection fraction and rate-adjusted mean velocity of fiber shortening decreased after prolonged exercise [from 67 +/- 5 to 60 +/- 6% (P less than 0.001) and from 1.12 +/- 0.2 to 0.91 +/- 0.2 cm/s (P less than 0.001), respectively] despite a lower circumferential end-systolic wall stress (133 +/- 23 vs. 152 +/- 20 g/cm2). The relationship between ejection fraction (or mean velocity of fiber shortening adjusted for heart rate) and end-systolic wall stress was displaced downward on race finish (P less than 0.05). These changes were independent of the changes in left ventricular end-diastolic volume and hence those in preload. The data suggest that moderately prolonged exercise may result in depressed left ventricular performance in healthy normal subjects.