The addition of fructose to natural meals elicits lower serum glucose and immunoreactive insulin responses when compared with that of sucrose and starch meals. Differences in rates of splanchnic glucose appearance and peripheral glucose disposal may be partly responsible. To evaluate the role of both parameters after different carbohydrate-enriched meals, we measured the arterialized venous blood glucose, immunoreactive insulin and gastric inhibitory polypeptide concentrations in seven Type 2 diabetic patients after ingestion of isocaloric test meals. Measurements were made in a random manner on three separate occasions. Fructose, sucrose, and bread supplementation constituted 68% of the total carbohydrate content of each meal. Rates of total glucose appearance, glucose utilization and metabolic clearance rates of glucose were determined by the D3-H-3 glucose prime-continuous infusion technique. The mean fasting glucose levels were similar in the three groups. Mean peak glucose concentrations and integrated incremental areas were significantly lower (p less than 0.02) after the fructose-enriched meals compared with that of either sucrose or bread. The basal arterialized venous blood glucose levels were similar in all three groups. The mean incremental integrated arterialized venous blood glucose area was significantly lower in the fructose group when compared with the sucrose (p less than 0.05) and bread (p less than 0.02) groups. The mean fasting gastric inhibitory polypeptide levels were similar in the three groups. However, the mean incremental integrated gastric inhibitory polypeptide areas were significantly lower in the fructose group compared with the sucrose and bread groups (p less than 0.01 and p less than 0.05 respectively). Basal hepatic glucose outputs were not significantly different in the three groups.(ABSTRACT TRUNCATED AT 250 WORDS)