All rosiglitazone doses significantly reduced fasting plasma glucose (Fig. 1) and peak postprandial glucose concentrations compared with baseline. Concomitantly, C-peptide and serum insulin concentrations (Fig. 1) as well as non-esterified fatty acid levels were significantly reduced from baseline in all rosiglitazone groups. Rosiglitazone at 4 and 6 mg twice daily significantly decreased fructosamine levels (Fig. 1) and prevented the increase in HbA1c observed in the placebo group.
Whatever the parameter considered, the glucose-lowering effect of the 4 mg twice-daily dose of rosiglitazone was similar to that of 6 mg twice daily. Triglyceride levels did not change, while total, LDL and HDL cholesterol concentrations increased significantly in the rosiglitazone treatment groups, with no significant change in the total cholesterol/HDL ratio. The proportion of patients with one or more adverse events was similar in all four treatment groups. No patients showed evidence of hepatotoxicity. Dose-dependent small decreases in haemoglobin and haematocrit and increases in body weight were observed in the rosiglitazone-treated groups.
In conclusion, rosiglitazone given twice daily significantly reduced fasting and postprandial glucose concentrations, C-peptide, insulin and non-esterified fatty acids in Type 2 diabetic patients, with a maximum clinical dose of 4 mg twice daily.

Comment
This dose-ranging study clearly shows that rosiglitazone given as monotherapy in doses of 2, 4 and 6 mg twice daily reduces fasting and postprandial glucose concentrations in overweight/obese Type 2 diabetic patients. Metabolic improvement was already observed after 4 weeks of treatment and progressed until the end of the study period at 8 weeks. Improved glycaemic control occurred in rosiglitazone-treated patients without an accompanying increase in fasting or postprandial insulin concentrations; in contrast, significant reductions in insulin and C-peptide levels were observed. This finding is consistent with the concept that rosiglitazone improves peripheral insulin sensitivity, presumably by increasing glucose uptake by skeletal muscle. This effect may be at least partly related to the significant decrease in the circulating levels of non-esterified fatty acids observed during rosiglitazone therapy.
Despite the rather short duration of the study, significant changes in fructosamine and HbA1c levels were observed in patients receiving rosiglitazone 4 or 6 mg twice daily. The effects of the 4 mg twice-daily dose were greater than those of the 2 mg twice-daily dose, but similar to those of the 6 mg twice-daily dose. These observations suggest that the top of the dose-response curve had been attained and that 4 mg twice daily should be the maximum clinical dose. The overall improvement in metabolic control observed with rosiglitazone 4 mg twice daily was similar to that previously reported with troglitazone 400–800 mg per day [1, 2]. These clinical observations support previous biochemical data demonstrating that rosiglitazone is 100 times more potent than troglitazone in terms of its activation of peroxisome proliferator-activated receptor-gamma.
Troglitazone has been recently withdrawn from the market because of severe liver toxicity in some patients. Although rosiglitazone and troglitazone (and pioglitazone as well) share a common thiazolidine-2-4-dione structure, differences in the side chain result in differences in bioavailability, metabolism and antihyperglycaemic action. No signs of liver toxicity were detected in the present short-term (8-week) study of some 220 patients receiving rosiglitazone. Other long-term (6- to 12-month) clinical trials performed on several hundreds of diabetic patients also failed to evidence any clinically significant alterations in liver enzymes, suggesting that rosiglitazone, in contrast to troglitazone, has no hepatotoxicity.
Other concerns have been raised about treatment with compounds of the glitazone family, among them the occurrence of mild anaemia (due to haemodilution), oedema (due to fluid retention which may trigger cardiac insufficiency in some at-risk patients), weight gain (of a few kilograms during the first months of therapy) and unwanted changes in lipid profile (such as a mild increase in total and LDL cholesterol) [1, 2]. All these side effects were observed in the present study, despite its short duration. In this respect, rosiglitazone appears to share the same properties as troglitazone [1, 2].
Since there is no currently available study demonstrating the positive effects of glitazones on clinical (reduction of complications) rather than biological endpoints (better blood glucose control, lower circulating insulin levels), it is difficult at present to hypothesize what might be the overall effect of glitazone administration on the long-term prognosis of diabetic patients. This is the reason why, despite the demonstrated efficacy of glitazones as monotherapy in Type 2 diabetic patients (as shown with rosiglitazone in the present study), thiazolidinediones are not recognized as first-choice drug therapy by the European Medicines Evaluation Agency, but rather should be used in combination and reserved for patients insufficiently controlled by treatment with sulphonylureas [3] and/or metformin [4]. It is indeed well recognized that Type 2 diabetes is a heterogeneous disease with multiple metabolic defects requiring, in most cases, a combined strategy in order to sufficiently improve metabolic control [5, 6]. Further long-term studies are requested, and expected with great interest, which should demonstrate both the safety and clinical efficacy of compounds of the glitazone family (rosiglitazone and pioglitazone) in the overall management of patients with Type 2 diabetes, particularly as first-line monotherapy.

References
1. Scheen AJ, Lefèbvre PJ. Troglitazone: antihyperglycemic activity and potential role in the treatment of type 2 diabetes. Diabetes Care 1999; 22: 1568–77.
2. Saleh YM, Mudaliar SR, Henry RR. Metabolic and vascular effects of the thiazolidinedione troglitazone. 
Diabetes Rev 1999; 7: 55–76.
3. Wolffenbuttel BH, Gomis R, Squatrito S et al. Addition of low-dose rosiglitazone to sulphonylurea therapy improves glycaemic control in type 2 diabetic patients. Diabetic Med 2000; 17: 40–7.
4. Fonseca V, Rosenstock J, Patwardhan R, Salzman A. Effect of metformin and rosiglitazone combination therapy in patients with type 2 diabetes mellitus: a randomized controlled trial. J Am Med Assoc 2000; 283: 1695–702.
5. Scheen AJ, Lefèbvre PJ. Oral antidiabetic agents: a guide to selection. Drugs 1998; 55: 225–36.
6. Chehade JM, Mooradian AD. A rational approach to drug therapy of type 2 diabetes mellitus. Drugs 2000; 60: 96–113.

Summary and Comment:
A.J. Scheen, Liège, Belgium