The aim of this study was to examine the effects of a high carbohydrate diet on glycaemic control, resting muscle glycogen levels and exercise performance in athletes with insulin dependent diabetes (IDDM). Seven trained (mean +/- S.D., VO2max 50.3 +/- 7.4 ml/kg/min) IDDM males consumed a high carbohydrate diet (HCD) or a normal mixed diet (NMD) for 3 week periods in a randomised crossover trial with a one week wash-out. Carbohydrate provided 59% or 50% of total energy intake, respectively, on the two diets. Fasting plasma lipids, mean blood glucose (over 96 h), fructosamine and muscle glycogen were measured and insulin use recorded. Exercise performance was evaluated by a 15 min time trial following a 50 min pre-loading block. Statistical significance was assessed using two tailed paired Student t-tests. Mean blood glucose was 10% higher on HCD than NMD (p = 0.005), fructosamine levels were 375 +/- 54 and 353 +/- 51 (mol/L on HCD and NMD, resp., p = 0.04) and daily insulin requirements were 15% higher on HCD than NMD (p = 0.02). Fasting blood lipids were similar on the two diets. Muscle glycogen was significantly lower on HCD than NMD (88.2 +/- 19.2 and 95.6 +/- 14.6 mmol/kg ww, respectively, p = 0.02). Exercise completed during the time trial was 6% less on HCD than on NMD (p = 0.007). An increased carbohydrate intake for three weeks, in IDDM athletes, is associated with a deterioration in glycaemic control, increased insulin requirements, decreased muscle glycogen and reduced exercise performance. These data do not support recommendations for IDDM athletes to consume a high carbohydrate diet, at least not when glycaemic control worsens upon following this advice, as was observed in this short-term study.