We have studied the impact of liquid diets formulated for complete or supplemental enteral nutrition of type II, non insulin-dependent (NIDDM) diabetics on carbohydrate homeostasis. To achieve this, liquid formula tolerance tests were performed in NIDDM patients under an oral treatment regimen with a combination of diet plus glibenclamide (7 men, 3 women, age: 56 +/- 11 years; mean body mass index of 26.2 +/- 3.6 kg/m2). After an overnight fast, each patient received the usual morning medication and, thereafter, ingested formula diet in randomized order with 10 day intervals between tests. 500 ml were administered of either a standard liquid diet (Biosorb Sonde), a fibre containing diet (Biosorb Plus Sonde), or a carbohydrate modified, fructose containing "diabetes" diet (Fresubin Diabetes) (carbohydrate contents of approximately equal to 60 g, respectively). Blood samples were collected over 180 min. Considering minor variations in the nutritional values of the diets, IR-insulin, IR-C-peptide, IR-glucose-dependent insulinotropic polypeptide (GIP), and IR-glucagon-like peptide. 1 (GLP-1) in plasma did not significantly differ between the study groups. After ingestion of Biosorb Sonde area under the curve glucose was greater than that seen after uptake of fibre containing or carbohydrate modified, i.e. fructose containing "liquid diabetes diets". All diets challenged the entero-insular axis in non-insulin dependent diabetics to a comparable extent. This data does not support the contention to employ special "diabetes" formulas for enteral nutrition of patients with NIDDM.