This investigation was performed to determine (i) whether 31P spatially localized 31P NMR spectroscopy could be utilized to determine the transmural distribution of 2-deoxyglucose (2DG) uptake in the in vivo canine heart and (ii) whether transmural 2DG uptake would be affected by a preceding ischemic insult. 2DG was infused and the accumulation of 2-deoxyglucose-6-phosphate (2DGP) was monitored (by means of spatially localized 31P NMR) in control hearts, in pharmacologically hyperperfused hearts, and in hearts subjected to four (5 min) occlusions of the left anterior descending coronary artery. Myocardial blood flow was measured with radioactive microspheres. In control hearts, subendocardial (ENDO) 2DGP contents were significantly higher than those in the subepicardium (EPI) being 3.8 +/- 0.3 and 2.2 +/- 0.2 mumol/g, respectively; the ENDO/EPI ratio of 2DGP was 1.70 +/- 0.21. During hyperperfusion blood flow increased approximately four-fold but 2DGP accumulation was not altered. ATP levels in post-ischemic myocardium were significantly decreased (ENDO more than EPI) and 2DGP accumulation in each layer was increased (p < 0.01 vs control); however, the ENDO/EPI ratio of 2DGP was not altered. 2DG infusion induced a marked elevation of blood insulin and norepinephrine levels. These data demonstrate that in the presence of high blood levels of 2DG and insulin: (i) 2DGP accumulation can be measured in the in vivo canine heart; (ii) in normal hearts 2DG uptake is more pronounced in the inner layers of the left ventricular wall (this transmural 2DG uptake gradient is not due to subendocardial hypoperfusion); and (iii) 2DG uptake is greater in the post-ischemic heart but the ENDO/EPI gradient of 2DG uptake is not altered indicating that the more severe ischemic insult in the subendocardium does not result in a disproportionate increase in 2DG uptake in that region of the myocardium. Although 2DG uptake patterns in this model most probably reflect those of glucose (at comparable glucose and insulin levels), quantitative extrapolations with regard to the rate of glucose uptake are not possible from the present data.