RESUMO
<p><b>BACKGROUND</b>Glycemic control prevents onset and progression of diabetes-related long-term complications. The objective of this study was to demonstrate that twice daily insulin lispro low mix 25 is noninferior to twice daily human insulin mix 30/70 in achieving glycemic control as measured by hemoglobin A1c (HbA1c), from baseline to endpoint, in patients with type 1 or 2 diabetes.</p><p><b>METHODS</b>In this phase IV, crossover, open-label, multicenter study, 117 Chinese patients with diabetes were randomly assigned to one of two treatment sequence groups. One group received 12-week treatment with twice daily human insulin mix 30/70 followed by 12-week treatment with twice daily insulin lispro low mix 25, while the other group received the reverse treatment sequence. HbA1c, baseline-to-endpoint change in HbA1c, proportion of patients achieving target HbA1c <or= 7% and <or= 6.5%, fasting blood glucose, and daily insulin doses were measured for each period. Safety and tolerability were also assessed.</p><p><b>RESULTS</b>A statistically significant reduction (P <or= 0.0001) of HbA1c was achieved after each treatment (human insulin mix 30/70: mean HbA1c = 7.91% (95%CI: 7.67%, 8.15%); insulin lispro low mix 25: mean HbA1c = 7.96% (95%CI: 7.72%, 8.20%)). The 95%CI (-0.20, 0.10) of the difference between the two treatments satisfied the prespecified noninferiority margin of 0.3% (lower limit of 95% CI > -0.3%). No statistically significant differences between treatments were observed for any of the secondary efficacy measures. The incidence of treatment-emergent adverse events and hypoglycemia between the two treatments and treatment sequence groups was similar. Three serious adverse events were reported (human insulin mix 30/70 group: 2 patients (1.7%, hypoglycemic coma and cardiac failure); insulin lispro low mix 25 group: 1 patient (0.9%, stroke)). All serious adverse events were resolved and no patients died during the study.</p><p><b>CONCLUSION</b>The results support noninferiority of twice daily insulin lispro low mix 25 versus twice daily human insulin mix 30/70 in HbA1c control in Chinese patients with type 1 or 2 diabetes.</p>