We documented stability of plasma glucose concentrations and glucose production and utilization rates, and levels of other metabolic substrates and regulatory factors, during the islet clamp (somatostatin infusion with glucagon and insulin replacement) in the absence of an intervention in five normal humans and further applied this technique to the study of glucoregulation during moderate exercise. Based on previous evidence that sympathochromaffin activation plays a primary role in the prevention of hypoglycemia during exercise, the role of adrenomedullary catecholamines was assessed by exercise (60% of maximum oxygen consumption for 60 min) studies in four bilaterally adrenalectomized, epinephrine-deficient humans under two conditions: control (saline infusion) and islet clamp. Increased glucose utilization and production rates were matched and plasma glucose was unchanged during exercise under both conditions. Thus adrenomedullary catecholamines including epinephrine are not critical to glucoregulation during moderate exercise in humans even when changes in insulin and glucagon are prevented. These findings provide further support for the suggestion that sympathetic neural norepinephrine is the operative catecholamine in the prevention of hypoglycemia during exercise in humans.

During mild or moderate nonexhausting exercise, glucose utilization increases sharply but is normally matched by increased glucose production such that hypoglycemia does not occur. To test the hypothesis that redundant glucoregulatory systems including sympathochromaffin activation and changes in pancreatic islet hormone secretion underlie this precise matching, eight young adults exercised at 55-60% of maximal oxygen consumption for 60 min on separate occasions under four conditions: (a) control study (saline infusion); (b) islet clamp study (insulin and glucagon held constant by somatostatin infusion with glucagon and insulin replacement at fixed rates before, during and after exercise with insulin doses determined individually and shown to produce normal and stable plasma glucose concentrations prior to each study); (c) adrenergic blockage study (infusions of the a-and ,8-adrenergic antagonists phentolamine and propranolol); (d) adrenergic blockade plus islet clamp study. Glucose production matched increased glucose utilization during exercise in the control study and plasma glucose did not fall (92±1 mg/dl at base line, 90±2 mg/dl at the end of exercise). Plasma glucose also did not fall during exercise when changes in insulin and glucagon were prevented in the islet clamp study. In the adrenergic blockade study, plasma glucose declined initially during exercise because of a greater initial increase in glucose utilization, then plateaued with an end-exercise value of 74±3 mg/dl (P < 0.01 vs. control). In contrast, in the adrenergic blockade plus islet clamp study, exercise was associated with glucose production substantially lower than control and plasma glucose fell progressively to 58±7 mg/dl (P < 0.001); end-exercise plasma glucose concentrations ranged from 34 to 72 mg/dl. Thus, we conclude that: (a) redundant glucoregulatory systems are involved in the precise matching of increased glucose utilization and glucose production that normally prevents hypoglycemia during moderate exercise in humans. (b) Sympathochromaffin activation, perhaps sympathetic neural norepinephrine release, plays a primary glucoregulatory role by limiting glucose utilization as well as stimulating glucose production. (c) Changes in pancreatic islet hormone secretion (decrements in insulin, increments in glucagon, or both) are not normally critical but become critical when catecholamine action is deficient. (d) Glucoregulation fails,