Aims/hypothesis We compared in vivo changes in liver glycogen concentration during exercise between patients with type 1 diabetes and healthy volunteers. MethodsWe studied seven men with type 1 diabetes (mean ± SEM diabetes duration 10±2 years, age 33±3 years, BMI 24±1 kg/m 2 , HbA 1c 8.1±0.2% and VO 2 peak 43±2 ml [kg lean body mass] −1 min −1 ) and five non-diabetic controls (mean ± SEM age 30±3 years, BMI 22±1 kg/m 2 , HbA 1c 5.4±0.1% and VO 2 peak 52±4 ml [kg lean body mass] −1 min −1 , before and after a standardised breakfast and after three bouts (EX1, EX2, EX3) of 40 min of cycling at 60% VO 2 peak. 13 C Magnetic resonance spectroscopy of liver glycogen was acquired in a 3.0 T magnet using a surface coil. Whole-body substrate oxidation was determined using indirect calorimetry.Results Blood glucose and serum insulin concentrations were significantly higher (p<0.05) in the fasting state, during the postprandial period and during EX1 and EX2 in subjects with type 1 diabetes compared with controls. Serum insulin concentration was still different between groups during EX3 (p<0.05), but blood glucose concentration was similar. There was no difference between groups in liver glycogen concentration before or after the three bouts of exercise, despite the relative hyperinsulinaemia in type 1 diabetes. There were also no differences in substrate oxidation rates between groups. Conclusions/interpretation In patients with type 1 diabetes, hyperinsulinaemic and hyperglycaemic conditions during moderate exercise did not suppress hepatic glycogen concentrations. These findings do not support the hypothesis that exercise-induced hypoglycaemia in patients with type 1 diabetes is due to suppression of hepatic glycogen mobilisation.