A real world comparison of sulfonylurea and insulin vs. incretin-based treatments in patients not controlled on prior metformin monotherapy

AIMS: Metformin is the first line drug for patients diagnosed with type-2 diabetes; however, the impact of different treatment escalation strategies after metformin failure has thus far not been investigated in a real world situation. The registry described herein goes some way to clarifying treatment outcomes in such patients. METHODS: DiaRegis is a multicentre registry including 3,810 patients with type-2 diabetes. For the present analysis we selected patients being treated with metformin monotherapy at baseline (n = 1,373), with the subsequent addition of incretin-based drugs (Met/Incr; n = 783), sulfonylureas (Met/SU; n = 255), or insulin (n = 220). RESULTS: After two years 1,110 of the initial 1,373 patients had a complete follow-up (80.8%) and 726 of these were still on the initial treatment combination (65.4%). After treatment escalation, compared to Met/Incr (n = 421), Met/SU (n = 154) therapy resulted in a higher HbA1c reduction vs. baseline (−0.6 ± 1.4% vs. −0.5 ± 1.0%; p = 0.039). Insulin (n = 151) resulted in a stronger reduction in HbA1c (−0.9 ± 2.0% vs. −0.5 ± 1.0%; p = 0.003), and fasting plasma glucose (−24 ± 70 mg/dl vs. −19 ± 42 mg/dl; p = 0.001), but was associated with increased bodyweight (0.8 ± 9.0 kg vs. −1.5 ± 5.0 kg; p = 0.028). Hypoglycaemia rates (any with or without help and symptoms) were higher for patients receiving insulin (Odds Ratio [OR] 8.35; 95% Confidence Interval [CI] 4.84-14.4) and Met/SU (OR 2.70; 95% CI 1.48-4.92) versus Met/Incr. While there was little difference in event rates between Met/Incr and Met/SU, insulin was associated with higher rates of death, major cardiac and cerebrovascular events, and microvascular disease. CONCLUSIONS: Taking the results of DiaRegis into consideration it can be concluded that incretin-based treatment strategies appear to have a favourable balance between glycemic control and treatment emergent adverse effects. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12933-015-0172-9) contains supplementary material, which is available to authorized users.