Contribution of postprandial versus interprandial blood glucose to HbA(1c)in type 1 diabetes on physiologic intensive therapy with lispro insulin atmealtime

OBJECTIVE - To quantitate the contribution of postprandial blood glucose, w hich improves with the short-acting insulin analog lispro [Lys(B28),Pro(B29 )] in type 1 diabetes, to the overall 24-h blood glucose concentration and the long-term HbA(1c) concentration under conditions of different postabsor ptive blood glucose. RESEARCH DESIGN AND METHODS - A total of 24 type 1 diabetic patients on lon gterm intensive therapy with premeal human regular insulin (Hum-R) and bedt ime NPH were randomly assigned to a continuation of Hum-R (group 1, n = 8), lispro (group 2, n = 8), or lispro + NPH tin variable proportions) adminis tered at mealtime (group 3, n = 8) for 3 months. NPH administered at bedtim e was continued in all three groups. Data from home blood glucose monitorin g were collected, and a 24-h plasma glucose and insulin profile was obtaine d during a 2-day hospital visit to calculate areas under the postprandial g lucose curve (3.5 h after breakfast, 3.5 h after lunch, and 3.0 h after din ner for a total of 10.0 h) and the postabsorptive blood glucose curve (the remaining 14.0 h out of 24.0 h) (AUC). Eight nondiabetic subjects were also studied. RESULTS - The substitution of Hum-R with lispro (group 2) resulted in lower postprandial blood glucose, but greater postabsorptive blood glucose (P < 0.05 vs, group 1). The postprandial blood glucose AUC was lower (161 +/- 19 vs. 167 +/- 20 mg . 100 ml(-1) . h(-1)),but the postabsorptive blood gluco se AUC was greater (155 +/- 22 vs. 142 +/- 19 mg . 100 ml(-1) . h(-1)) (P < 0.05). Therefore, the 24-h blood glucose AUC was no different (NS). Conseq uently, HbA(1c) was no different (NS). This occurred because in group 2, me altime lispro resulted in normal prandial plasma insulin, but also resulted in lower interprandial concentration (P < 0.05 vs, group I). When NPH was added to lispro (30% at breakfast, 40% at lunch, 10% at dinner) in group 3, postabsorptive plasma insulin was similar to group 1 (NS). In group 3, the postprandial blood glucose AUC (153 +/- 17 mg . 100 ml(-1) . h(-1)) was lo wer and the postabsorptive blood glucose AUC was no different, as compared with group 1 (NS). Therefore, the 24-h blood glucose AUC was lower (147 +/- 17 vs. 155 +/- 21 and 158 +/- 20 mg . 100 ml(-1) . h(-1)), and HbA(1c) was lower (6.41 +/- 0.12 vs. 6.84 +/- 0.2 and 6.96 +/- 0.2% (groups 3, 1, and 2 respectively, P < 0.05). Frequency of hypoglycemia was greater in group 2 (P < 0.05),but not in group 3 (NS) vs. group 1. CONCLUSIONS - Lispro administered at mealtime, which improves postprandial blood glucose, should be associated with optimized replacement of basal ins ulin to prevent deterioration of postabsorptive blood glucose. NPH administ ered four times daily normalizes interprandial daily plasma insulin concent ration and decreases mean daily blood glucose and HbA(1c), with no increase in hypoglycemia.