Fractures, heart failure and fears of myocardial ischaemia: has the RECORD stuck for rosiglitazone and the thiazolidinediones?

Thiazolidinediones (TZDs) are extensively prescribed worldwide to lower blood glucose levels in type 2 diabetes mellitus (DM). Safety concerns regarding this class of drugs have arisen from a recently published meta‐analysis as well as an underpowered, interim report from the RECORD study group. The meta‐analysis compared treatment with rosiglitazone with other therapies for type 2 DM or placebo, and found that rosiglitazone was associated with a significant increase in the risk of myocardial infarction (MI) and a borderline‐significant finding from death from cardiovascular causes. Though the actual numbers of events were small and the methodology open to criticism, the subsequent early report of the RECORD study also suggests an increased cardiovascular risk. Moreover, the latter study corroborated with data from the previously reported DREAM, ADOPT and PROactive studies in highlighting an increased risk of fluid retention and cardiac failure. Recent data from both animal and human studies have described the detrimental effect of TZD treatment upon bone mineral density in men and women. The rosiglitazone cardiovascular safety data, however, must be considered alongside the benefits ascribed to pioglitazone treatment from the PROactive study in terms of risk reduction of the trial's secondary endpoint of all‐cause mortality, non‐fatal MI and non‐fatal stroke, as well as subsequent data from this trial group in terms of stroke and MI in high‐risk patients with type 2 DM. Further data regarding the safety profile of rosiglitazone are required prior to a definitive conclusion and it may be appropriate, for patients stable on rosiglitazone treatment, to remain on it, but the data should be considered by both the treating physician and the patient. Copyright © 2007 John Wiley & Sons.