The magnitude of the counterregulatory response to insulin-induced hypoglycemia is primarily determined by the degree of hypoglycemia. We examined whether the route of acute insulin delivery (portal or peripheral venous) is also important in determining the magnitude of the counterregulatory response to hypoglycemia in nine healthy nondiabetic men. Pancreatic insulin secretion, stimulated by an i.v. tolbutamide infusion (portal insulin study), was matched with an exogenous insulin infusion into the peripheral vein 4-6 weeks later (peripheral insulin study). Each study consisted of a 150-min baseline tracer equilibration period, a 180-min euglycemic hyperinsulinemic (portal or peripheral insulin delivery) period, a 60-min hypoglycemic period in which insulin secretion diminished during tolbutamide or was reduced during exogenous insulin, and a 30-min recovery period. Peripheral venous glucose concentrations were well matched in the portal and peripheral studies during euglycemia and hypoglycemia (glucose nadir, 2.9 +/- 0.1 mmol/L in the portal and 2.7 +/- 0.1 mmol/L in the peripheral; mean +/- SEM; P = NS), and insulin concentrations were about 1.5-fold higher throughout the experiment in the peripheral vs. the portal insulin study due to the first pass extraction of insulin in the portal study. There was a much greater increment (P < 0.0001) in FFA in the portal vs. the peripheral study (area under the curve: portal, 19.5 +/- 3.9 mmol/L x 90 min; peripheral, 3.3 +/- 1.1 mmol/L x 90 min), whereas plasma glucagon and GH were higher in the peripheral study (P = 0.01 for glucagon; P = 0.015 for GH). There was no significant difference between studies in epinephrine and norepinephrine responses to hypoglycemia or stimulation of endogenous glucose production (area under the curve: portal, 636 +/- 103 micromol/kg x 90 min; peripheral, 705 +/- 69 micromol/kg x 90 min; P = NS). In summary, we have shown that the glucagon, GH, and FFA responses to hypoglycemia during insulin dissipation are affected by the route of insulin delivery and are not controlled exclusively by the nadir blood glucose level. The clinical importance of these observations in diabetic subjects as they relate to route of insulin delivery (portal or peripheral) during insulin dissipation remains to be determined.