The goal of treatment of type I glycogen storage disease (GSD-I) is to prevent hypoglycemia and its biochemical consequences. In seven patients with GSD-I with a mean age of 19.5 y (range: 18.8-21.7 y), we compared the biochemical effects of isoenergetic amounts of uncooked cornstarch (UCS; 1.76 +/- 0.41 g/kg) given in random order on consecutive nights either as a single dose at 2100 (time 0) or as equally divided doses at 2100 and 0200. Over the 10-h period of observation there were significant regimen-by-time interactions for plasma glucose, serum insulin, and blood lactate concentrations. Mean time-averaged plasma glucose (5.8 +/- 0.5 compared with 4.9 +/- 0.9 mmol/L) and serum insulin (244 +/- 93 compared with 151 +/- 57 pmol/L) concentrations from 0 to 360 min were significantly higher after the single dose; blood lactate and serum fatty acid concentrations were not significantly different. At 360 min, mean plasma glucose (4.8 +/- 1.2 compared with 4.7 +/- 1.6 mmol/L) and serum insulin (138 +/- 76 compared with 136 +/- 116 pmol/L) concentrations were virtually identical. After a single dose, plasma glucose concentrations were > or = 3.9 mmol/L for 7 h in five of seven subjects; three subjects were treated for hypoglycemia after 7-9.5 h. With divided doses, plasma glucose concentrations were > or = 3.9 mmol/L for 9 h in six of seven subjects; hypoglycemia occurred at 6 h in one subject. A single dose (1.76 +/- 0.41 g/kg) of UCS at bedtime maintains plasma glucose concentrations > or = 3.9 mmol/L for > or = 7 h in most young adults with GSD-I.