Measurements of stroke volume (SV) by echo-Doppler techniques (2D-echocardiography, continuous and pulsed Doppler) were compared to those obtained by thermodilution in 116 critically ill patients. The best correlation was obtained with pulsed Doppler positioned in the left ventricular (LV) outflow (r = 0.78, p less than 0.001). In a subgroup of 12 patients who had LV ejection fraction less than 60% and LV end-diastolic pressure greater than 15 mmHg, we also measured the peak aortic blood velocity (PABV) and acceleration (PABA) by continuous and pulsed Doppler as indexes of myocardial contractility before coronary artery bypass graft. When compared to healthy volunteers, these patients had decreased PABV and PABA, measured by continuous Doppler (91 +/- 16 vs 79 +/- 9 cm/s, p less than 0.05 and 913 +/- 202 vs 660 +/- 149 cm/s/s, p less than 0.05, respectively). These data therefore indicate that in critically ill patients pulsed Doppler placed in the LV outflow is an adequate echo-Doppler technique to measure SV, while continuous Doppler placed in the suprasternal notch can be used to assess LV performance.