The increase of myocardial 99mTc-MIBI accumulation during exercise was evaluated quantitatively, and the feasibility of response rate as a noninvasive marker of coronary stenosis was tested. Myocardial perfusion imaging with 99mTc-MIBI during exercise and at rest was performed in patients with suspected coronary artery disease. A dose of 296 MBq of 99mTc-MIBI was injected intravenously at maximal treadmill exercise, and myocardial image was obtained 90 min later (1st image). Then, 740 MBq of 99mTc-MIBI was administered at rest, and myocardial image was repeated (2nd image). These images were corrected for a decay and injected dose, and the 1st image was subtracted from the 2nd image to obtain the rest image. An increase of myocardial accumulation of 99mTc-MIBI during exercise was defined as (exercise image-rest image) x 100/rest image (response rate). A response rate of a patient with normal coronary artery was 102%, whereas a response rate in the area of severe coronary stenosis was 21% in a patients with angina pectoris. After successful PTCA to a stenosed coronary artery, a response rate increased to 75% in this patient. Coronary perfusion reserve during exercise can be assessed noninvasively by 99mTc-MIBI. Response rate of 99mTc-MIBI provides additional information to conventional perfusion imaging and may be a new marker of severity of coronary artery disease.