The effect of small amounts of fructose on net hepatic glucose uptake (NHGU) during hyperglycemia was examined in the presence of insulinopenia in conscious 42 h-fasted dogs. During the study, somatostatin (0.8 µg^.kg^-1.min^-1) was given along with basal insulin (1.8 pmol^.kg^-1.min^-1) and glucagon (0.5 ng^.kg^-1.min^-1). After a control period, glucose (36.1µmol^.kg^-1.min^-1) was continuously given intraportally for 4 h with (2.2 µmol^.kg^-1.min^-1) or without fructose. In the fructose group, the sinusoidal blood fructose level (nmol/ml) rose from <16 to 176 ± 11. The infusion of glucose alone (the control group) elevated arterial blood glucose (µmol/ml) from 4.3 ± 0.3 to 11.2 ± 0.6 during the first 2 h after which it remained at 11.6 ± 0.8. In the presence of fructose, glucose infusion elevated arterial blood glucose (µmol/ml) from 4.3 ± 0.2 to 7.4 ± 0.6 during the first 1 h after which it decreased to 6.1 ± 0.4 by 180 min. With glucose infusion, net hepatic glucose balance (µmol^.kg^-1.min^{- 1}) switched from output (8.9 ± 1.7 and 13.3 ± 2.8) to uptake (12.2 ± 4.4 and 29.4 ± 6.7) in the control and fructose groups, respectively. Average NHGU (µmol^.kg^-1.min^-1) and fractional glucose extraction (%) during last 3 h of the test period were higher in the fructose group (30.6 ± 3.3 and 14.5 ± 1.4) than in the control group (15.0 ± 4.4 and 5.9 ± 1.8). Glucose-6-phosphate and glycogen content (µmol glucose/g) in the liver and glucose incorporation into hepatic glycogen (µmol glucose/g) were higher in the fructose (218 ± 2, 283 ± 25 and 109 ± 26, respectively) than in the control group (80 ± 8, 220 ± 31 and 41 ± 5, respectively). In conclusion, small amounts of fructose can markedly reduce hyperglycemia during intraportal glucose infusion by increasing NHGU even when insulin secretion is compromised.