For intensified insulin therapy of type 1 diabetes, bolus injection of regular human insulin 30–15 min before a meal is currently recommended. This randomized study is aimed to determine whether insulin lispro (LIS), a new insulin analog with a rapid onset of action, can provide comparable blood glucose (BG) control by injection after the meal.


Eighteen type 1 diabetic subjects injected regular insulin (REG) at 40, 20, or 0 min before or LIS at 20 or 0 min before or 15 min after the start of a standardized test meal. BG excursions and area under the curve of BG excursions (AUC) at the six visits were compared by analysis of variance. Hypoglycemic events (BG ≤2.78 mmol/1) were evaluated in relation to the achieved postprandial BG control.


Mean AUC values were 2.00, 2.55, and 3.33 mmol · h · 1−1 for REG given 40, 20, and 0 min before the test meal, respectively, and −2.19, −2.15, and 1.98 mmol · h · 1−1 for LIS given 20 and 0 min before and 15 min after the start of the test meal, respectively. LIS injected 20 min (−20) or immediately (0) before the meal was significantly more effective in controlling postprandial BG excursion (P < 0.001) than any REG treatment. Postprandial injection of LIS (15) did not compromise postprandial BG control and resulted in less hypoglycemia. REG −40 and LIS −20 were associated with early hypoglycemia, but other hypoglycemic events were equally distributed among groups.


The optimal time for bolus insulin injection was 20 min before the meal for REG and immediately before the meal for LIS. LIS injected immediately after a standard meal provided postprandial BG control at least as good as REG injected from 40 to 0 min before the meal. Postprandial injection of LIS is an attractive new therapeutic option.

This content is only available via PDF.