A Comparison of Lipid and Glycemic Effects of Pioglitazone and Rosiglitazone in Patients With Type 2 Diabetes and Dyslipidemia: Response to Goldberg et al.

Goldberg et al. (1) report a randomized comparison of the effects of two thiazolidinediones (TZDs), pioglitazone (Actos) and rosiglitazone (Avandia), on lipid and lipoprotein levels in individuals with diabetes treated with diet and/or oral monotherapy. Many changes in these lipids and lipoproteins seemed to be in favor of pioglitazone. Two important pieces of information were not included in the published study.

First, ∼80% of subjects were termed completers. Completers were defined as those who had at least one blood sample drawn on a TZD. If they quit the study, this last observation was carried forward for final analysis. It is possible that a number of subjects dropped out while on the initial dose of the respective TZD and would be a completer. By study design, this would enrich the population with subjects on two-thirds of the final dose of pioglitazone (30 of 45 mg/day) compared with one-half of the final dose (4 of 8 mg) of rosiglitazone. The authors should provide data about subjects remaining in the study at each visit to evaluate the possibility of outcome differences in lipids due to differences in relative drug dose. Perhaps it would be of interest to report the relevant data of those on each drug who actually completed all planned visits in the study.

Second, what were HDL subfraction responses as determined by nuclear magnetic resonance? Was there a similar response of big, intermediate, and small HDL particles? Some of the lipoprotein and lipid changes seen with pioglitazone appear to be those that would be expected as a peroxisome proliferator–activated receptor-α effect. What was the response of big and intermediate size HDL, those that are antiathergenic (2), and the small HDL that are less antiatherogenic?