Gastric inhibitory polypeptide in obesity and diabetes mellitus

Gastric inhibitory polypeptide (GIP) concentrations may be influenced by obesity, diabetes, and glucagon deficiency and be under feedback inhibition by insulin. To assess these factors, insulin-dependent diabetic, totally pancreatectomized diabetic, and lean and obese noninsulin-dependent diabetic patients were studied twice, once during partial insulin withdrawal and again when euglycemia was achieved before and after mixed meal ingestion, using an artificial endocrine pancreas. The results were compared to those from weight-matched lean and obese nondiabetic subjects. No significant differences in postprandial GIP responses were found between lean and obese nondiabetic subjects. Despite basal and postprandial hyperglycemia, the GIP responses to the mixed meal were not significantly different between insulin-deficient (insulin-dependent and totally pancreatectomized) patients and lean nondiabetic subjects. In addition, there were no significant differences in postprandial GIP responses between insulin-dependent and totally pancreatectomized patients. In contrast, lean and obese noninsulin-dependent diabetic patients had reduced GIP responses compared to weight-matched nondiabetic subjects (mean +/- SE, 37.9 +/- 5.4 vs. 67.1 +/- 10.8 ng ml-1 240 min-1, respectively; P less than 0.05). This difference was entirely due to the reduced GIP responses in obese noninsulin-dependent diabetic patients compared to those in obese nondiabetic subjects (32.1 +/- 7.9 vs. 76.9 +/- 18.2 ng ml-1 240 min-1, respectively; P less than 0.05); the postprandial GIP responses were not significantly different between lean noninsulin-dependent diabetic patients and lean nondiabetic subjects. Insulin infusion by an artificial endocrine pancreas resulted in postprandial insulin and glucose profiles that approximated those of nondiabetics, but did not significantly alter GIP responses to the mixed meal (48.2 +/- 5.5 ng ml-1 240 min-1) in the 18 diabetic patients compared to results obtained with sc insulin treatment (42.2 +/- 5.2 ng ml-1 240 min-1). In conclusion, postprandial GIP responses are normal in obese nondiabetic subjects and insulin-deficient diabetic patients and are blunted in obese, but not in lean, noninsulin-dependent diabetic patients. In addition, GIP does not appear to be under feedback inhibition by insulin or influenced by glucagon deficiency in diabetes.