Spontaneous or propranolol-induced hypoglycemia can occur in uremic humans. We studied glucose kinetics (using [3-3H]glucose) in five uremic humans 24 h after hemodialysis and in seven normal controls. The effect of glucagon infusion at rates of 3, 6, 12, and 18 ng X kg-1 X min-1 at 60-min intervals was compared with either saline or beta-adrenergic blockade (propranolol infusion). In uremics, plasma glucose increased by 20-25% and by 40-50% at the 3 and 6 ng X kg-1 X min-1 glucagon doses, respectively, with no further increases at higher infusion rates. Glucose production increased transiently and in tandem with glucose uptake at each glucagon increment (P less than 0.0001). During beta-adrenergic blockade, the effect of glucagon in stimulating glucose production was blunted by 14-24% at the 6-18 ng X kg-1 X min-1 doses (P less than 0.05). During saline infusion, plasma insulin concentrations increased progressively to peak levels fourfold above basal at the 18 ng X kg-1 X min-1 dose. This increase in plasma insulin was virtually abolished by concomitant beta-adrenergic blockade (P = 0.0002). In contrast to uremic subjects, normal controls exhibited lesser degrees of hyperglycemia and hyperinsulinemia at all glucagon infusion rates. Propranolol infusion had no effect on the increments in glucose production and uptake nor on the plasma insulin response. These results suggest that in uremic humans propranolol independently reduces the hepatic response to glucagon and the insulin secretory response to hyperglycemia and/or hyperglucagonemia. These observations provide a possible mechanism for the adrenergic regulation of glucose homeostasis in uremia.