We evaluated the effect of antecedent hypoglycemia on glucose counterregulation during hypoglycemia in non-diabetic human subjects. In single hypoglycemia studies, glucose production [( 3H]3-glucose) and counterregulatory hormone concentrations were measured (after a 3.5-h baseline period of euglycemia) during 120 min of hypoglycemia (glucose clamped at 3.0 mmol/L). During the final 60 min of hypoglycemia, counterregulation resulted in significant increments in glucose production (12.88 +/- 0.83 mumol/kg.min), and plasma glucagon (IRG; 185 +/- 22 ng/L), GH (29.3 +/- 7.0 micrograms/L), cortisol (630 +/- 100 nmol/L), epinephrine (3.44 +/- 0.76 nmol/L), and norepinephrine (2.02 +/- 0.21 nmol/L). In the recurrent hypoglycemia experiment, an antecedent period of identical hypoglycemia was induced. Glucose counterregulation during the second of two periods of hypoglycemia (HYPO 2) was then compared to that in single hypoglycemia studies. During HYPO 2, there were decreased responses in Ra (by 32%; P less than 0.03), GH (by 67%; P less than 0.05), F (by 41%; P less than 0.03), and norepinephrine (by 20%; P = 0.03) compared to those in the single hypoglycemia study. In contrast, plasma IRG values were similar in the single hypoglycemia studies and HYPO 2, but were reduced relative to those during the first hypoglycemic period of recurrent hypoglycemia (IRG, 263 +/- 18 ng/L; P less than 0.025 vs. HYPO 2 and P less than 0.05 vs. single hypoglycemia). Our results suggest that 1) antecedent hypoglycemia may alter glucose counterregulation during hypoglycemia; and 2) recurrent hypoglycemia may result in alterations in reduction of hepatic glucose production.

Counterregulatory hormone responses were evaluated in a 37-yr-old woman before and after removal of a benign insulin-producing islet cell tumor. Counterregulatory hormone concentrations were measured during a glucose clamp with graded reductions of plasma glucose from 5.2 to 2.6 mmol/L. In the study before surgery, the increase in plasma epinephrine concentration was markedly blunted (by greater than 90%) compared to that in the study after surgery. The peak plasma norepinephrine concentration was similarly reduced by 71%, and plasma cortisol by 63%. In addition, the glycemic thresholds for secretion of the counterregulatory hormones were lower before removal of the tumor. Peak plasma GH responses were equivalent before and after surgery, but the threshold for GH secretion was 21% lower in the first hypoglycemia study. We conclude 1) that there is evidence for abnormal glucose counterregulatory hormone secretion in this patient, which may contribute to the pathogenesis of hypoglycemia seen in patients with insulinoma; 2) the reversal of reduced counterregulatory hormone secretion after tumor resection suggests that these defective hormonal responses may be related to recurrent hypoglycemia, persistent hyperinsulinemia, or both; and 3) that abnormal glucose counterregulation may exist in the absence of type 1 diabetes.

We evaluated the effect of previous experimental hypoglycemia on counterregulatory responses to hypoglycemia in 13 IDDM patients. These patients had defects in counterregulatory responses to hypoglycemia compared with 7 nondiabetic control subjects. Plasma EPI and glucagon responses to hypoglycemia in IDDM patients were approximately 60% of levels in nondiabetic subjects (P less than 0.02 and P less than 0.001, respectively). Hepatic glucose output ([3-3H]glucose) was reduced by approximately 60% of normal (P less than 0.005), and the glucose infusion rate required to maintain plasma glucose was correspondingly greater in people with IDDM (P less than 0.001). With a modified glucose clamp (plasma insulin approximately 330 pM), the diabetic subjects underwent two sequential 120-min periods of hypoglycemia (approximately 3.0 mM) with an intervening 60-min euglycemic recovery period. In the IDDM patients, there were 30-50% decreases in plasma GH (P less than 0.005) and cortisol (P less than 0.001) responses during the second hypoglycemic period compared with the first. In addition, glucose output, already defective compared with that in nondiabetic subjects, was further reduced by 33% (P = 0.03) during the second period of experimental hypoglycemia. There was no effect of repeated hypoglycemia on the responses of plasma glucagon, EPI, or NE, though plasma EPI was correlated directly with glucose output (P less than 0.001) and inversely with glucose uptake (P less than 0.05). There was no correlation between the rise in glucose output during hypoglycemia and antecedent glycemic control as measured by HbA1.(ABSTRACT TRUNCATED AT 250 WORDS)

Our conclusions are as follows: 1) symptoms of moderate hypoglycemia occur at plasma glucose levels averaging approximately 5-15 mg/dl lower than the plasma glucose concentrations required to trigger counterregulatory hormone release; 2) after acute antecedent hypoglycemia, glucagon, EPI, and GH secretion occur at higher plasma glucose concentrations and NE is released at lower plasma glucose concentrations; and 3) there may be CNS adaptation to prior hypoglycemia reflected in preservation of logical memory function at plasma glucose levels of approximately 50 mg/dl. These findings suggest that thresholds for hormone secretion and for changes in cognitive function can be altered very acutely by foregoing hypoglycemia in healthy humans.

1) Brief exposure to even high physiological levels of insulin do not alter the magnitude of counterregulatory hormone secretion during hypoglycemia; 2) prolonged hyperinsulinemia results in a selective blunting of the plasma glucagon response to hypoglycemia, perhaps due to a direct suppressive effect of insulin on alpha-cell secretion.