The purpose of this study was to investigate the effects of exercise on myocardial glucose uptake and whether the pattern of glucose uptake is the same as in skeletal muscle. Glucose uptake was measured using positron emission tomography (PET) and 2-[18F]fluoro-2-deoxy-D-glucose ([18F]FDG). Twelve healthy men were studied during rest, while 14 subjects were studied after 35 min of bicycle exercise corresponding to 30, 55 and 75 % of maximal oxygen consumption (*VO2,max)) on three separate days. [(18)F]FDG was injected 10 min after the start of exercise and exercise continued for a further 25 min. Myocardial and skeletal muscle PET scanning was commenced directly after the completion of the exercise bout. As compared to the resting state, exercise doubled myocardial glucose uptake at the 30 % (P = 0.056) and 55 % intensity levels (P < 0.05), while at the 75 % intensity level glucose uptake was reduced significantly compared to the lower exercise intensities. There was no significant difference between the highest intensity level and the resting state (P = 0.18). At rest and during low-intensity exercise, myocardial glucose uptake was inversely associated with circulating levels of free fatty acids. However, during higher exercise intensities when plasma lactate concentrations increased significantly, this association disappeared. In contrast to myocardial responses, skeletal muscle glucose uptake rose in parallel with exercise intensity from rest to 30 % and then 55 % *VO2,max) (P < 0.001) and tended to increase further at the intensity of 75 % *VO2,max) (P = 0.065). In conclusion, these results demonstrate that myocardial glucose uptake is increased during mild- and moderate-intensity exercise, but is decreased during high-intensity exercise. This finding suggests that the increased myocardial energy that is needed during high-intensity exercise is supplied by substrates other than glucose.