Comparative Effectiveness between Dipeptidylpeptidase-4 Inhibitors and Sulfonylureas in Combination with Metformin in Type 2 Diabetes Mellitus Patients

ABSTRACT

OBJECTIVE: Treatment with sulfonylureas in combination with metformin improves glycemic control in type 2 diabetes mellitus (T2DM), but is associated with hypoglycemia and weight gain. This retrospective study aims to compare the effectiveness of dipeptidylpeptidase-4 (DPP-4) inhibitors and sulfonylureas as an add-on therapy to metformin in patients with T2DM. METHODS: Data from medical records of 355 T2DM patients received therapy either DPP-4 inhibitors (DPP-4 inhibitor group) or sulfonylurea (SU group) in combination with metformin from 1 March 2009 to 30 September 2011 were retrospectively reviewed. Of total 355 patients, 231 patients were in DPP-4 inhibitor group and 124 patients were in SU group. Baseline Hemoglobin A1c (HbA(1c)) level in SU group was higher than DPP-4 inhibitor group with a statistically significant difference (8.6% vs. 7.8%). Comparative analysis between DPP-4 inhibitor group and SU group was performed for HbA(1c) values, amounts of HbA(1c) changes, and rates of HbA(1c) changes from baseline at 6-month intervals and incidence rates of major cardiocerebral events. RESULTS: SU group showed larger HbA(1c) changes in both amounts and rates compared to DPP-4 inhibitor group, although statistical significance was not found in all study periods. Proportions of patients with stable HbA(1c) <6.5% or 7% were significantly higher in DPP-4 inhibitor group than SU group (<6.5% 30.4% vs. 13.4%, <7% 72.3% vs. 41.2%). Time to achieve stable HbA(1c) <6.5% was not significantly different, but time to achieve stable HbA(1c) <7% was shorter in DPP4 inhibitor group than SU group with a significant difference. The incidence rate of cardiocerebral events in group of patients with or without previous events was 1.7%, not significantly lower than that in DPP-4 inhibitor group (4.0%). For newly encountered cardiocerebral events during the treatment, incidence rates of two groups did not differ significantly. CONCLUSION: DPP-4 inhibitors were as effective as sulfonylureas in achieving the HbA(1c) goal of less than 6.5% or 7% and cardiocerebral event rates did not differ between the two drugs.