Metabolic effects of continuous subcutaneous insulin infusion: evidence that a rise and fall of portal vein insulin concentration with each major meal facilitates post-absorptive glycemic control

Eighteen lean adult volunteers with insulin-requiring diabetes mellitus attempted to achieve normoglycemia using continuous subcutaneous insulin infusion (CSII) or conventional insulin therapy (CIT) in a randomized crossover trial of 68 +/- 2.5 weeks (mean +/- SEM) duration. As reported (Diabetes Care 8: 447-55, 1985) the group with absent to low beta-cell function (C-peptide negative, n = 11) attained mean post-absorptive normoglycemia only during CSII vs CIT (p less than 0.05). Only following CSII was this without change in post-absorptive serum triglyceride concentrations (-4 +/- 5.6 vs 12 +/- 4.7 mg/dl; -0.04 +/- 0.6 vs 0.14 +/- 0.05 mM, p less than 0.05) or body weight (0.01 +/- 0.02 vs 0.05 +/- 0.01 kg/week, p less than 0.05). In the group with glucagon stimulated serum C-peptide 100-400 pmol/L (C-peptide positive) responses to CSII or CIT were equal. As total daily insulin dosage (0.05 +/- 0.04 U/kg/day) was the same under all conditions, to explain the efficacy of CSII, glucoregulatory hormone responses were examined. Pre- and post-test breakfast serum free immunoreactive insulin and plasma glucagon concentrations were essentially unaffected by C-peptide or treatment status. Erythrocyte 125I-insulin binding was decreased in the C-peptide negative group only during CSII (8.6 +/- 0.5 vs 10.1 +/- 0.7%, p less than 0.005); C-peptide positive group receptor binding was consistently low (8.2 +/- 0.8, 8.4 +/- 0.9%). During CIT using intermediate-acting insulin post-lunch peripheral venous insulin failed to rise (p less than 0.05), but in the C-peptide positive group, on the basis of C-peptide responses to breakfast an undetected rise and fall of portal venous insulin was assumed to coincide with each meal. Thus, only during CIT in the C-peptide negative group, which received on average 6.4/wk/subject fewer pre-meal regular insulin boluses (p less than 0.01), was the frequency of meal-related change in portal insulinemia decreased. Consistent meal-related fluctuations in portal insulinemia inherent in CSII hepatocytes sensitized by a post-receptor mechanism to the suppressive effects of insulin on glucose output and thus were indirectly responsible for the observed improvement in glycemic control and lipid metabolism in the C-peptide negative group.