The development of rebound hypoglycaemia has been reported after pre-exercise carbohydrate (CHO) ingestion in some studies but not in others. Differences in the experimental design and factors such as the exercise intensity are likely to be responsible for the discrepancies between these studies. Exercise intensity might be a crucial factor since it affects both insulinaemia and glucose uptake. Therefore the aim of the present study was to compare the glycaemic and insulinaemic responses to exercise at different intensities after ingestion of a standardized pre-exercise CHO load. Eight moderately trained subjects consumed 75 g of glucose 45 min prior to 20 min of exercise at 40%, 65% or 80% maximal power output. Blood samples were collected before glucose ingestion, at 15 min intervals at rest and 5 min intervals during exercise. During exercise, measurements of heart rate and breath-by-breath analysis of expired gas were performed continuously. The trials were performed at [mean (SEM)] 55 (1), 77 (1) and 90 (1) percentages maximal oxygen uptake. At the onset of exercise, plasma glucose concentration returned to pre-ingestion levels, while the insulin concentration was more than three times higher than at rest [on average 57 (7) compared to 16 (1) microU.ml(-1)]. During exercise, plasma glucose concentrations decreased during the first 5 min of exercise and then stabilized in all trials at concentrations that would not be considered to be hypoglycaemic. There were no significant differences in glucose or insulin concentrations between the three trials during exercise. These data suggest that the glycaemic response to ingestion of 75 g of CHO 45 min pre-exercise is similar during exercise of different intensities.