The aim of present study is to compare the efficacy and safety of adding vildagliptin with sulfonylurea dose-increasing as an active comparator in patients who had inadequately controlled type 2 diabetes mellitus (T2DM) using metformin plus sulfonylurea in real clinical practice. Patients using metformin plus sulfonylurea were assigned to either vildagliptin add-on (50 mg twice a day, n=172) or sulfonylurea dose-increasing by 50% (n=172) treatment groups. The primary endpoint was a change in HbA(1c) after 24 weeks. The secondary endpoints were patients achieving HbA(1c)≤7.0% (53 mmol/mol) and changes in the fasting plasma glucose (FPG), 2-h postprandial glucose (2pp), lipid profiles, and urine albumin-to-creatinine ratio. Body weight and hypoglycemia were also investigated. The mean HbA(1c) at baseline was 8.6% (70 mmol/mol) in both groups. At week 24, the adjusted mean HbA(1c) levels decreased by -1.19% (-13.09 mmol/mol) with vildagliptin add-on and -0.46% (-5.06 mmol/mol) with sulfonylurea (P<0.001). Significantly more vildagliptin add-on patients achieved HbA(1c)≤7.0% (53 mmol/mol) than did sulfonylurea patients (40.1% vs. 7.9%; P<0.001). Greater reductions in FPG and 2pp were observed with vildagliptin add-on than with sulfonylurea (P<0.001). The vildagliptin add-on group exhibited no clinically relevant weight gain and had a lower incidence of hypoglycemia compared with the sulfonylurea group. Vildagliptin add-on therapy might be a suitable option for patients with T2DM that is controlled inadequately by metformin and sulfonylurea, based on its greater glucose control and better safety profile (ClinicalTrial.gov: NCT01099137).