Clinical outcomes with insulin lispro compared with human regular insulin: a meta-analysis.

We performed a meta-analysis to compare insulin lispro and human regular insulin across a range of outcomes common in modern diabetes management to establish a basis for subsequent economic evaluation. We included all identifiable head-to-head randomized controlled trials, pooling dichotomous and continuous outcomes using appropriate statistical methods. Measures associated with various aspects of glycemic control (preparandial and postprandial glycemic control, glucose excursion, and glycated hemoglobin) and with hypoglycemia were evaluated. Results showed significant differences in favor of insulin lispro in the outcomes associated with postprandial glycemic control without an increase in hypoglycemia. Outcomes associated with fasting glycemic control and overall long-term glycemic control were not significantly different between insulin lispro and human regular insulin. Alternative approaches to the meta-analysis were explored but did not alter the conclusions. Thus our meta-analysis supports the existence of significant differences between insulin lispro and human regular insulin in terms of important postprandial outcome measures in diabetes. In addition, there is a practical difference in injection timing relative to meals: human regular insulin should be administered 30 to 45 minutes before eating, whereas insulin lispro can be administered 15 minutes or less before eating. These differences should be the subject of an economic evaluation to assist in determining the place of insulin lispro in diabetes management.