Addition of biphasic insulin aspart 30 to optimized metformin and pioglitazone treatment of type 2 diabetes mellitus: The ACTION Study (Achieving Control Through Insulin plus Oral ageNts)

Aim: Efficacy and safety of biphasic insulin aspart (BIAsp 30, 30% short-acting and 70% intermediate-acting insulin aspart) added to an optimized treatment of metformin and pioglitazone (met/pio) were compared with treatment with optimized met/pio in type 2 diabetes patients. Methods: This randomized, 34-week, parallel-group study enrolled insulin-naive, type 2 diabetes patients (HbA_1c 7.5-12.0%) previously using two oral antidiabetic (OAD) agents. During an 8-week run-in period, treatment was changed to met/pio and doses were adjusted up to 2500mg/day and 30 or 45 mg/day respectively. Subjects either continued met/pio alone or added BIAsp 30 initiated at 6units twice daily and titrated to target plasma glucose (PG) (4.4-6.1 mmol/l). Results: At end-of-study, subjects treated with BIAsp 30+met/pio (n = 93) had a mean (± s.d.) HbA_1c reduction significantly greater than treatment with met/pio (n = 88) (1.5% ± 1.1 vs. 0.2% ± 0.9, p < 0.0001 between groups). Subjects treated with BIAsp 30+met/pio were more likely to reach The American Association of Clinical Endocrinologists and European Association for the Study of Diabetes/American Diabetes Association HbA_1c targets of ≤6.5 and <7.0%, respectively, than with met/pio only (HbA_1c ≤6.5%: 59 vs. 12%; HbA_1c <7.0%: 76 vs. 24%). At end-of-study, self-monitored glucose profile values at all eight daily time points were significantly less for the BIAsp 30+met/pio group compared with the met/pio group, and minor hypoglycaemia (defined as PG<3.1 mmol/l) was more frequent (8.3 vs. 0.1 events/year, p < 0.001). Both groups gained weight during treatment (BIAsp 30+met/pio, 4.6 ± 4.3 kg; met/pio, 0.8 ± 3.2 kg; p < 0.001). Conclusion: Addition of insulin in type 2 patients treated with met/pio is an effective way to achieve glycaemic targets. Treatment with BIAsp 30+met/pio achieved significantly greater reduction in HbA_1c, as compared with met/pio alone. In patients with type 2 diabetes poorly controlled by 2 OADs, more achieved glycaemic targets using BIAsp 30+met/pio than using met/pio alone.