To determine the impact of both short- and longterm near-normoglycaemia on insulin resistance in Type 1 (insulin-dependent) diabetes hepatic glucose production (mg · kg^–1 · min^–1) and peripheral glucose utilisation (M-value, mg · kg^–1 · min^–1) were estimated during an euglycaemic hyperinsulinaemic clamp (10 mU · kg · min) in patients with either good (HbA_1c<5.8%, groups A and B) or poor (HbA_1c>7.5%, groups C and D) long-term metabolic control (time > 12 months) and in healthy subjects (HbA_1c: 5.08±0.20%; n=8). To this end blood glucose was stabilized at 6.7 mmol/l by overnight (t=12 h) i.v. regular insulin in groups (n=8 each) A (HbA_1c: 5.49±0.46%) and C (HbA_1c: 8.83±1.20%),while groups B (HbA_1c:5.55±0.19%) andD (HbA_1c: 8.51±1.09%) were kept overnight on long-acting insulin without feed-back control of blood glucose before euglycaemic clamping. Thereby, pre-equilibration of blood glucose at 6.7 mmol/l was shown to normalize basal hepatic glucose production (A: 2.27±0.48; C 2.50±0.57 mg · kg^–1 · min^–1) despite different HbA_1c values, whereas basal hepatic glucose production stayed elevated in groups B (3.09±0.38 mg · kg^–1 · min^–1) and D (3.21±0.58 mg · kg^–1 · min^–1) with poor actual glycaemia (B: 10.9±4.6; D: 12.1±4.6 mmol/l). To restitute peripheral glucose utilisation close to normal (healthy subjects: 13.99±2.13; A: 12.12±2.67; B: 8.72±3.0; C: 10.27±1.69; D: 7.10±2.31 mg · kg^–1 · min^–1; healthy subjects vs A: NS; healthy subjects vs B, C, D: p<0.05) both long-term (HbA_1c<5.8%) and acute nearnormoglycaemia by 12-h i. v. insulin pre-treatment were required (group A). We conclude that good long-term glucose control per se is unable to normalize hepatic and peripheral glucose metabolism in Type 1 diabetic patients unless actual near-normoglycaemia is provided consistently, e.g. by i.v. overnight infusion of regular insulin.