A study was performed to test the hypothesis that Doppler echocardiographic measurement of ascending aortic blood flow can detect exercise induced changes in left ventricular performance during exercise in patients suspected of having ischaemic heart disease. Acceleration and peak velocity of flow and stroke volume were determined by non-imaging Doppler echocardiography in the suprasternal notch in 38 patients as they underwent simultaneous exercise radionuclide ventriculography. The patients were divided into four groups: group 1 had resting ejection fractions greater than or equal to 50% and increased their ejection fractions greater than or equal to 5% during exercise; group 2 had resting ejection fractions of greater than or equal to 50% but the ejection fraction either fell or rose less than 5% during exercise; group 3 had resting ejection fractions less than 50% but the ejection fraction rose greater than or equal to 5% during exercise; and group 4 had resting ejection fractions less than 50% and the exercise ejection fraction either fell or rose less than 5% during exercise. Acceleration, velocity, and stroke volume all rose significantly during exercise in group 1. Acceleration also increased in group 2 but to a lesser extent; velocity and stroke volume did not increase. In group 3 acceleration and velocity increased but to a lesser extent than in group 1; stroke volume did not increase. In group 4 velocity increased slightly during exercise but acceleration and stroke volume were unchanged. Doppler echocardiography thus appears capable of detecting exercise induced changes in left ventricular performance and can identify normal and abnormal responses, as defined by radionuclide ventriculography.