Comparison of an oral mixed meal plus arginine and intravenous glucose, GLP-1 plus arginine to unmask residual islet function in longstanding type 1 diabetes.

Aims: Residual beta cells are present in most patients with longstanding type 1 diabetes but it is unknown whether these beta cells react normally to different stimuli. Moreover a defect in proinsulin conversion and abnormal alfa cell response are also part of the islet dysfunction. Methods: A three-phase (euglycemia, hyperglycemia and hyperglycemia + glucagon-like peptide 1) clamp was performed in patients with longstanding type 1 diabetes. Intravenous arginine boluses were administered at the end of each phase. On another day a mixed meal stimulation test with a subsequent intravenous arginine bolus was performed. Results: C-peptide was detectable in a subgroup of subjects at baseline (2/15) or only after stimulation (3/15). When detectable, C-peptide increased 2.9 fold [95% CI: 1.2-7.1] during the hyperglycemia phase and 14.1 fold [95% CI: 3.1-65.2] during the hyperglycemia+GLP-1 phase and 22.3 [95% CI: 5.6-89.1] fold during hyperglycemia+GLP-1+arginine when compared to baseline. The same subset of patients with a C-peptide response were identified during the mixed meal stimulation test as during the clamp. There was an inhibition of glucagon secretion (0.72-fold, [95% CI: 0.63-0.84]) during the glucose clamp irrespective of the presence of detectable beta cell function. Proinsulin was only present in a subset of subjects with detectable C-peptide (3/15) and proinsulin mimicked the C-peptide response to the different stimuli when detectable. Conclusion: Residual beta cells in longstanding type 1 diabetes respond adequately to different stimuli and could be of clinical benefit.