Antidiabetogenic action of glucagon-like peptide-1 related to administration relative to meal intake in subjects with type 2 diabetes

Objective, To establish the antidiabetogenic effect of glucagon-like peptide-1 (GLP-1) when differently administered relative to meal intake in subjects with type 2 diabetes, Design, The study was a placebo-controlled comparison with random assignment to treatment sequence. A 3-h stepwise infusion of GLP-1 (17 nmol) was started either at the onset of a standard meal (550 kCal) (A) or at 30 min (B) or 60 min (C) after the start of the meal. Setting, The study was conducted at a university hospital. Subjects, Eight patients with type 2 diabetes (four women and four men), age 62 +/- 3.9 years (range 47-74 years), weight 79.8 +/- 5.4 kg (range 62-104 kg), BMI 26.2 +/- 1.3 kg m(-2) (range 21-31 kg m(-2)), diabetes duration 10.5 +/- 2.0 years (range 3-19 years) and HbA1c levels 6.1 +/- 0.3% (range 4.7-7.7%) participated in the study. All patients were treated with oral sulphonylureas. Results. Glucagon-like peptide-1 significantly lowered postprandial glycaemia by a similar degree in all three situations versus the control meal(P < 0.05). Postprandial insulin levels were not different in the four experimental series, whereas the postprandial glucagon levels were significantly lowered by GLP-1 in (A) and (B) (P < 0.03) but not in (C). Gastric emptying, as determined by the paracetamol test, was retarded by GLP-1 only in (A) (P < 0.01), but not affected in (B) or (C). Conclusions. GLP-1 reduced postprandial hyperglycaemia in subjects with type 2 diabetes regardless of administration at the onset of meal intake or at 30 or 60 min after start of meal intake, although the mechanism of the antidiabetogenic action of GLP-1 depended on administration versus meal intake. Thus, when administered at the start of a meal, GLP-1 was antidiabetogenic mainly through retarding gastric emptying, whereas when given at 30 or 60 min after meal ingestion, changes in islet hormone secretion seem to be predominant.