Résumé
Objective To investigate the efficacy of biphasic insulin aspart 50(BIAsp50)twice daily(bid) versusbiphasichumaninsulin50(BHI50)(bid)plusmetforminonbloodglucosecontrolfollowingastandardmealtest in Chinese patients with type 2 diabetes mellitus(T2DM). Methods A randomized, open-label, 2-sequence, crossover trial for two 4-week treatment periods was conducted in 14 Chines institutes. Eligible subjects inadequately controlled with BHI50(bid)plus metformin were randomized to two sequences in a 1 : 1 ratio(A:BIAsp50-BHI50, B:BHI50-BIAsp50 ) . Standard meal tests were performed at baseline and the ends of two periods within 4 weeks. Primary endpoint was 2h postprandial plasma glucose ( PPG) increment following standard meal test, with insulin dose standardized at 0. 3 IU/kg. Results A total of 161 subjects were randomized into two sequences(81 to sequence A, and 80 to sequence B) and finally analysed. After 4 weeks of treatment, mean 2h PPG increment with BIAsp50 was lower than that with BHI50 [ treatment difference of BIAsp50 vs BHI50: -1. 12 mmol/L ( 95% CI-1. 66,-0. 58), P<0. 01], suggesting superiority of BIAsp50 over BHI50. Incremental area under the curve for PPG(0-2 h)with BIAsp50 was lower than that with BHI50 [treatment difference:-38. 8 mmol·L-1·min-1(95%CI-77. 3,-0. 26), P=0. 049], as was the mean 2h PPG [treatment difference:-0. 58 mmol/L(95% CI -1. 13,-0. 03), P=0. 040]. The FPG value with BIAsp50 was higher than that with BHI50 [treatment difference:0. 52 mmol/L(95%CI 0. 18, 0. 86), P=0. 003]. The rate of nocturnal hypoglycemia with BIAsp50 was lower than that with BHI50(1. 13 vs 2. 86 events per subject year, P<0. 01). Conclusion In patients with T2DM inadequately controlled with BHI50 plus metformin, BIAsp50 was proven to be well-tolerated with improved postprandial glucose control compared with BHI50.