Accuracy of diagnosis of coronary artery disease by radionuclide measurement of left ventricular function during rest and exercise

Rest and exercise radionuclide angiocardiographic measurements of left ventricular function were obtained in 496 patients who underwent cardiac catheterization for chest pain. Two hundred forty-eight of these patients also had an exercise treadmill test. An ejection fraction less that 50% was the abnormality of resting left ventricular function that provided the greatest diagnostic information. In patients with normal resting left ventricular function, exercise abnormalities that were optimal for diagnosis of coronary artery disease were an ejection fraction at least 6% less than predicted, an increase of greater than 20 ml in end-systolic volume and the appearance of an exercise-induced wall motion abnormality. The sensitivity and specificity of the test were lower in patients who were taking propranolol at the time of study and in patients who failed to achieve an adequate exercise end point. In the 387 patients with an optimal study, the test had a sensitivity of 90% and a specificity of 58%. Radionuclide angiocardiography was more sensitive and less specific than the exercise treadmill test. The high degree of sensitivity of the radionuclide test suggests that it is most appropriately applied to patient groups with a high prevalence of disease, including those considered for cardiac catheterization.