As part of a multicenter trial, 70 individuals with insulin-dependent diabetes were randomized to either conventional insulin therapy (CIT) or continuous subcutaneous insulin infusion (CSII). In order to standardize patient selection in the six participating centers, one of the eligibility criteria was the demonstration that each patient had no residual endogenous insulin secretion as assessed by plasma C-peptide determinations. The patients were of average (± SEM) age of 33.0 ± 1.6 yr, had had diabetes for a mean (±SEM) duration of 17.4 ± 1.1 yr, and had both fasting and postglucagon stimulation C-peptide values of <0.1 pmol/ml, consistent with clinically insignificant endogenous insulin secretion. There was no change in C-peptide response at 4 or 8 mo compared with baseline values, whether or not the patient's glucose control remained unchanged (CIT group) or significantly improved to near-normoglycemia (CSII group). In a subgroup of 34 patients at three centers, the 24-h mean glucose concentration was significantly lower in the CSII group compared with the CIT group at 4 mo (126 ± 10 versus 176 ± 14 mg/dl) and at 8 mo (121 ± 5 versus 183 ± 15 mg/dl) (P < 0.005). Although the 24-h mean serum free immunoreactive insulin levels were shown to be no different at baseline (27.4 ± 3.8 versus 26.2 ± 3.1 μU/ml) or after 4 mo (22.5 ± 3.2 versus 25.6 ± 3.2) or 8 mo (26.5 ± 3.4 versus 28.8 ± 3.4) of CIT or CSII therapy, respectively, the mean increase of free insulin concentrations in relation to the main meals was greater in the CSII group (P < 0.05). The mean serum free insulin levels achieved with both CIT and CSII were significantly greater than the 24-h mean free insulin level obtained in 12 normal subjects (12.2 ± 1.4 μU/ml) (P < 0.001).

This content is only available via PDF.