Myocardial oxygen requirements are met by coronary blood flow and oxygen extraction from the arterial blood. The myocardial oxygen extraction coefficient (1) expresses the fraction of available oxygen in arterial blood removed by the myocardium: coronary A-Vo2/Ao2 X 100 = myocardial extraction percentage. Myocardial oxygen extraction during rest is approximately 70 to 75 per cent, exceeding that of all other organs (1, 2). This degree of extraction is equalled only by exercising skeletal muscle, in which increased oxygen requirements are met both by additional blood flow and by increased oxygen extraction. Earlier workers have noted a relative constancy of myocardial oxygen extraction under the widely varying conditions of rest, anemia, hypoxia, and increased left ventricular work, suggesting that coronary blood flow is altered commensurate with myocardial oxygen requirements (1, 3, 4). Case, Berglund and Sarnoff demonstrated that when coronary blood flow reserve was limited by experimental coronary constriction, myocardial extraction increased due to the combined stress of anemia and increased left ventricular work (3).The effect of exercise on coronary hemodynamics has received little attention. Lombardo and co-workers reported observations in 13 human subjects with various forms of heart disease (5). As part of another study, Regan and associates reported on the effects of exercise on the coronary circulation in nine normal subjects (6). This report will present myocardial oxygen extraction data from 108 patients at rest, and from 79 subjects during mild exercise. The physiological significance of myocardial oxygen extrac-* This work was supported by grants from the U. S. (7), and routine cardiac catheterization findings. The exercise electrocardiograms were interpreted according to the criteria of Mattingly, Fancher, Bauer and Robb (8).Four clinical states were selected for correlation with myocardial oxygen extraction patterns. The patients were grouped clinically according to the following criteria.Control. Patients having no symptoms or electrocardiographic, radiologic or physical findings suggesting coronary artery disease or congestive heart failure served as controls. Pulmonary capillary pressures were normal in all subjects except those with mild mitral stenosis. The range of diagnoses included functional systolic murmurs, patent ductus arteriosus, ventricular and atrial septal defects, mild aortic insufficiency, noncritical mitral stenosis (valve area greater than 1.5 cm'), tetralogy of Fallot, and mild coarctation of the aorta. Two had atypical chest pain without evidence to suggest coronary insufficiency.Coronary insufficiency. Patients judged to have coronary insufficiency on the basis of classical angina pectoris, documented myocardial infarction, and positive Master's test were placed in this group. In several cases the coronary vessels were assessed at operation, by coronary arteriography (9), or at necropsy. Several patients had coexisting congestive heart failure.Congestive heart failure. Th...