Background: Type 2 diabetes mellitus is twice as prevalent in African Americans and Hispanic Americans as in non-Hispanic whites. However, the effectiveness and safety profile of rosiglitazone maleate used as combination therapy with sulfonylureas in the management of diabetes and its effect on cardiovascular disease (CVD) biomarkers/parameters have not been studied in these populations. Objective: The purpose of this study was to determine the efficacy and tolerability of the addition of rosiglitazone to a regimen of glyburide once daily in African American and Hispanic American patients with type 2 diabetes previously inadequately controlled with sulfonylurea monotherapy. Methods: This randomized, double-blind, placebo-controlled, parallel-group study was conducted at 38 centers in the United States. Eligible patients were aged ≤21 years, had type 2 diabetes, a fasting plasma glucose (FPG) level ≥140 mg/dL, and a glycosylated hemoglobin (HbA 1c) value ≥7.5%, and had been treated with sulfonylurea monotherapy for at least 2 months before screening. Patients were assigned to receive treatment with glyburide 10 or 20 mg/d plus rosiglitazone 8 mg (GLY+RSG) or placebo (GLY+PBO) PO (tablets) QD for 24 weeks. The primary efficacy end point was the change from baseline in HbA 1c after 24 weeks of treatment. Secondary end points included change in FPG; proportion of patients achieving HbA 1c targets (<7.0% and <6.5%); and changes in biomarkers for CVD risk, including C-reactive protein (CRP), plasminogen activator inhibitor (PAI)-I activity, fibrinogen, tissue plasminogen activator (tPA) antigen, von Willebrand factor (vWF), soluble vascular cell adhesion molecule (sVCAM), lipoprotein-associated phospholipase A 2 activity, and urinary albumin/creatinine ratio (UACR). Tolerability was assessed using physical examination, including vital-sign measurement, clinical laboratory tests, and adverse event (AE) reports collected at each study visit. Results: A total of 245 patients (101 African American and 144 Hispanic American) were enrolled. Demographic characteristics were comparable between the GLY+RSG and GLY+PBO groups: mean (SD) age (52 [11.9] vs 53 [10.4] years), HbA 1c (9.2% [1.3%] vs 9.4% [1.4%]), sex (men/women, 45.3%/54.7% vs 48.3%/51.7%), race (African American/Hispanic American, 43.6%/56.4% vs 37.9%/62.1%), and mean (SD) weight (86.3 [18.8] vs 88.3 [19.4] kg). In the overall study population, treatment with GLY+RSG was associated with a significantly greater mean (95% CI) reduction from baseline in HbA 1c compared with GLY+PBO (between-group Δ, −1.4% [−1.7% to −1.1%]; P < 0.001). When assessed by ethnicity, HbA 1c values were significantly reduced with GLY+RSG compared with GLY+PBO in African American patients (between-group Δ, -1.4%) and in Hispanic American patients (between-group Δ, −1.5%) (both, P < 0.001), as were FPG levels (between-group Δs, −3.1 mmol/L [57 mg/dL] and −3.8 mmol/L [−69 mg/dL], respectively; both, P < 0.001). With GLY+RSG, 9151 (17.6%) African American patients and 17/66 (25.8%) Hispanic American patients achieved HbA 1c <7%, compared with 2/44 (4.5%) and 1/72 (1.4%) patients, respectively, who achieved this goal with GLY+PBO. Homeostasis model assessment estimates of insulin sensitivity and β-cell function were significantly improved with GLY+RSG compared with GIX+PBO (between-group Δs, 29.3% and 78.4%, respectively; both, P < 0.001). With regard to CVD biomarkers, there were potentially deleterious changes compared with baseline in the GLY+PBO group for CRP (+29.4%; P = 0.042), PAI-1 activity (+27.0%; P = 0.006), fibrinogen (+15.7%; P = 0.007), and sVCAM (+7.0%; P = 0.035), whereas there were no significant increases in these factors in the GLY+RSG group. In the GLY+RSG group, there were significant improvements in tPA (−17.8%; P < 0.001), vWF (−11.3%; P = 0.019), and UACR (−17.2%; P = 0.028) over 24 weeks' treatment, whereas there were no significant changes in any of these factors in the GLY+PBO group. As a result, significant treatment effects were observed for CRP (−29.2%; P = 0.019), tPA (−18.4%; P < 0.001), vWF (−12.9%; P < 0.015), and UACR (−26.7%; P = 0.006) with GLY+RSG compared with GLY+PBO. The most frequently reported AEs with GLY+RSG were edema and weight increase (both 121121 [9.9%] patients) and with GLY+PBO were upper respiratory tract infection (18/124 [14.5%] patients). AEs were reported in 83/121 (68.6%) patients in the GLY+RSG group, of which 6/121 (5.0%) were assessed as severe, compared with 70/124 ( 56.5 % ) patients who received GLY+PBO, of which 31124 (2.4%) were assessed as severe. Conclusion: Add-on rosiglitazone administered for 24 weeks was effective and well tolerated in these African American and Hispanic American patients with type 2 diabetes previously inadequately controlled on sulfonylurea monotherapy.