The aim of this study was to assess regional and global variability of insulin-stimulated myocardial glucose uptake in healthy elderly subjects and to evaluate potentially responsible factors. Twenty men with a mean age of 64 years, no history of cardiovascular disease, and normal blood pressure, bicycle exercise test, electrocardiogram and echocardiography were studied [P(coronary artery disease) <5%]. Whole-body insulin sensitivity and insulin-stimulated myocardial glucose uptake were measured during hyperinsulinaemic euglycaemic glucose clamp with fluorine-18 fluorodeoxyglucose, and myocardial rest and hyperaemic blood flow during dipyridamole infusion were measured with nitrogen-13 ammonia and positron emission tomography in 16 left ventricular myocardial segments. Intra-individual and inter-individual variability of insulin-stimulated myocardial glucose uptake [relative dispersion = (standard deviation/mean)] was 13% and 29% respectively. Although inter-individual variability of glucose uptake and blood flow at rest was of the same magnitude, no correlation was found between these measures. Regional and global insulin-stimulated myocardial glucose uptake correlated linearly with whole-body insulin sensitivity (r=0.51, P<0.05 and r=0.56, P<0.01). The strongest independent association by multivariate linear regression analysis was found between myocardial glucose uptake and hyperaemic blood flow (r=0.63, P<0.005). We conclude that in healthy elderly subjects, insulin-stimulated myocardial glucose uptake is homogeneous throughout the left ventricle, but has moderate inter-individual variability. Inter-individual variability of insulin-stimulated myocardial glucose uptake is primarily explained by variability in coronary vascular reactivity and tissue insulin sensitivity.