The objective of this study was to evaluate the effect of benfluorex in type II diabetic patients already treated with sulfonylureas (SU) in a randomized placebo-controlled trial. After a 4-week placebo run-in, 68 patients (49 men and 19 women; age range 40-70 years; known duration of diabetes 0.5-19 years) were randomized to double-blind 12-week treatment with benfluorex (B) or placebo (P). Primary end points were HbA1c and fasting blood glucose (FBG). Secondary end points were glucose tolerance (meal test over 120 min), plasma insulin, C-peptide, and lipid profile. Results were analyzed using both intention-to-treat analysis (ITT) in patients completing at least one treatment visit and per protocol analysis in those completing the whole study. There were no baseline differences between the two groups in any study parameter. Fifty-eight patients completed the study (28 B, 30 P), and 66 patients (33 B, 33 P) were eligible for ITT analysis. Over the 12-week treatment period, FBG decreased by 14.9% in the B group (-1.39 mmol/L, p < 0.001), and 3.2% in the P group (-0.28 mmol/L, NS) according to the ITT analysis and by 17.4% (p < 0.001) and 3.8% (NS), respectively, in the per protocol analysis. The difference in FBG outcome between the two groups was significant (p = 0.009 and p = 0.004, respectively). In patients completing the study, mean HbA1c decreased in the B group (-0.66%, p = 0.005) and remained stable in the P group (+0.14%, NS). HbA1c outcome differed between the two groups (p = 0.007). The decrease in AUCglucose was greater in the B group than in the P group (-210 +/- 220 versus -60 +/- 270 mmol/L x 120 min, p = 0.026). Plasma insulin and C-peptide changes did not differ between the two groups. Low-density lipoprotein (LDL) cholesterol decreased in B patients and was stable in P patients (-0.43 versus -0.05 mmol/L, p = 0.026). Of the 68 randomized patients, six on B and four on P reported at least one adverse event, causing dropout in five and two patients, respectively. In conclusion, benfluorex is an effective agent for combination therapy in type II diabetic patients poorly controlled on SUs alone.