The detection of coronary artery disease with radionuclide techniques: a comparison of rest-exercise thallium imaging and ejection fraction response.

Fifty-two patients with suspected coronary artery disease underwent coronary angiography, thallium-201 myocardial imaging, and ECG-gated blood pool ventriculography at rest and at maximal exercise. In 11 patients without coronary artery disease, all thallium images were normal. The resting ejection fraction (EF) was normal in all 11 patients and increased during exercise in six, was unchanged in three, and decreased in two. Of the 41 patients with coronary artery disease, the thallium image was normal at rest and at exercise in six (15%), whereas the exercise EF was abnormal in these six. A new (18 patients) or enlarged (11 patients) defect appeared on the thallium image in 29 of 41 patients (71%) with coronary disease. Six of 41 patients (15%) had an abnormal rest thallium image that was unchanged with exercise. An abnormal rest and/or exercise image defect identified 35 of 41 patients (85%) with coronary artery disease. The resting EF was normal in 26 of the 41 patients (63%) and in 24 patients demonstrated an abnormal response to exercise. Fifteen patients (37%) had an abnormal resting EF, and 14 of these 15 demonstrated persistent abnormalities. Thus, an abnormal exercise EF response identified 38 of 41 (93%) patients with coronary disease. The specificity of the thallium image was 100% and for the exercise EF determination, 54% (p < 0.02). We conclude that an abnormal exercise EF response and the rest-exercise thallium image have similar sensitivities for detecting coronary disease; however, an abnormal exercise ejection fraction was significantly more sen- sitive than was a new thallium abnormality alone (93% vs 71%). Combined, the two studies detected all patients with coronary disease.