Programming of insulin delivery with meals during subcutaneous insulin infusion.

The serum insulin elevation after subcutaneous (s.c.) infusion of 10 U of neutral regular insulin given over 5 or 30 min has been examined in normal subjects to find a delivery profile that would simulate the normal serum insulin response to a meal. Endogenous insulin release, assessed by measurement of C-peptide, was suppressed by an i.v. infusion of 1 U/h neutral regular insulin, beginning 60 min before the s.c. administration and continuing throughout the study. The increment in serum insulin levels after the 5-min s.c. infusion reached a peak of 62 +/- 9.5 mU/L at 45 min, and the rate of elevation of insulin levels closely approximated the normal insulin response to a 50-g, 500 kcal mixed meal. The increment in serum insulin levels with the 30-min s.c. infusion reached a peak of 41 +/- 1.4 mU/L at 75 min. The incremental insulin response up to 180 min (area under the curve) after the 5-min s.c. infusion was 30% greater than the response to the 30-min infusion. On the basis of these studies, we conclude that insulin delivery for meals during s.c. insulin infusion need not be initiated before the meal, but should be given over a period of approximately 5 min or less. Further study is required to delineate the effects of concentration, local degradation, and rate of delivery on entry of s.c. insulin to the vascular system.