Simultaneous estimates of cardiac output were made during graded upright maximal exercise in 10 male subjects by means of Doppler velocity spectrum of ascending aortic flow, apical two-dimensional echocardiograms, thermodilution, and Fick oximetry. In 15 subjects, aortic annular and root diameters were measured during similar exercise from parasternal two-dimensional echocardiograms. The linear correlation between Doppler, two-dimensional echocardiography, and the invasive estimates ranged from r = .78 to r = .92. Both echocardiographic techniques were able to predict changes in invasive flow estimates with reasonable accuracy. Two-dimensional echocardiographic flow estimates underestimated invasive values by about 60%. The accuracy of Doppler flow estimates varied with the method of estimating aortic cross-sectional area. Greatest accuracy was obtained with areas calculated from diameters measured at the aortic value anulus with the leading edge-to-leading edge method of measurement. Correlation coefficients comparing Doppler and thermodilution flow estimates were generally higher (r = .75 to .96, mean .86) for individuals than for the group, but accuracy of the Doppler estimates in single subjects was quite variable. Aortic diameters did not increase from rest to moderate levels of upright exercise. A 3% to 5% increase in resting aortic diameter was noted in the upright posture as compared with the supine. Doppler flow estimates were obtained in all subjects to maximal exertion but in only a minority of subjects with two-dimensional echocardiography or thermodilution. Thus two-dimensional and Doppler echocardiography offer a noninvasive means of estimating cardiac output during vigorous exercise. The Doppler technique is technically more suitable to the study of exercise than two-dimensional echocardiography. Aortic area estimates for Doppler flow calculations are best made from resting two-dimensional echocardiograms of the aortic anulus by means of the leading edge-to-leading edge method of measurement. There does not appear to be a significant change in aortic diameter during upright exercise, but there may be a postural effect on aortic dimensions. Circulation 76, No. 3, 539-547, 1987. STROKE VOLUME and cardiac output are fundamental descriptors of cardiovascular function. Cardiac output is the primary indicator of the functional capacity of the circulation to meet the increased demands of physical exertion. The relative contributions of changes in stroke volume and heart rate to cardiac output and other factors influencing cardiac output 12 It was selected for this study because it allowed for rapid volume calculation from easily obtained and reproducible measurements. This formula has previously been validated in our laboratory by comparing it to left ventricular volume measured by left ventricular angiography in 20 patients with symmetric left ventricular wall motion. The correlation between echocardiographic and cineangiographic estimates of volume (range 42 to 232 ml) was .92 (echocardiog...