Post-gastric bypass hyperinsulinemic hypoglycemia is a recently described disorder occurring in some patients after gastric bypass surgery for obesity. The pathogenesis is incompletely understood but involves a robust insulin response to ingested carbohydrate. The resultant hyperinsulinemia sometimes produces hypoglycemia with neuroglycopenia, confusion and even loss of consciousness. Various treatments have been recommended including low carbohydrate diets, coingestion of the medication acarbose with carbohydrate containing meals, partial pancreatectomy and even total pancreatectomy. None is completely satisfactory. We propose to test two new potential treatments. Using a design with random assignment of three conditions we plan to compare, in 10 patients with post-gastric bypass hyperinsulinemic hypoglycemia, a high carbohydrate test meal (control condition), a high carbohydrate test meal after pre-treatment with rapid acting aspart insulin (insulin condition), and a high fructose, low glucose test meal with carbohydrate and caloric content similar to the control meal (fructose condition).

Post-gastric bypass hyperinsulinemic hypoglycemia is a recently described disorder occurring in some patients after gastric bypass surgery for obesity. The pathogenesis is incompletely understood but involves a robust insulin response to ingested carbohydrate. The resultant hyperinsulinemia sometimes produces hypoglycemia with neuroglycopenia, confusion and even loss of consciousness. Various treatments have been recommended including low carbohydrate diets, coingestion of the medication acarbose with carbohydrate containing meals, partial pancreatectomy and even total pancreatectomy. None is completely satisfactory. We propose to test two new potential treatments. Using a design with random assignment of three conditions we plan to compare, in 10 patients with post-gastric bypass hyperinsulinemic hypoglycemia, a high carbohydrate test meal (control condition), a high carbohydrate test meal after pre-treatment with rapid acting aspart insulin (insulin condition), and a high fructose, low glucose test meal with carbohydrate and caloric content similar to the control meal (fructose condition). The hypothesis to be tested are 1) pretreatment with aspart insulin will prevent, or at least reduce, the occurrence of hypoglycemia and 2) substitution of fructose for glucose in the test meal will prevent, or at least reduce, the occurrence of hypoglycemia. Plasma glucose and serum insulin will be sampled before and for four hours after the three test conditions. The primary study endpoint will be the occurrence or not of plasma glucose < 60 mg/dL after the test meals. The control meal will be compared to the insulin pre-treated test meal and, in a separate comparison, to the fructose test meal. Secondary endpoints will be comparisons between the control and active treatments in peak postprandial serum insulin, peak postprandial plasma glucose, nadir postprandial plasma glucose, and the 4-hr longitudinal course of plasma glucose measurements.

Inclusion Criteria:

• Participants must be at least 21 years of age
• History of postprandial hypoglycemia with neuroglycopenia occurring one year or more after gastric bypass surgery
• History of spontaneous correction of hypoglycemia
• Normal fasting plasma glucose and serum insulin after a carbohydrate containing mixed meal, demonstration of serum insulin > 50u/UL and plasma glucose < 50mg/dL

Exclusion Criteria:

• Under 21 years of age