July 18, 2007 (Düsseldorf, Germany) - A new Cochrane review of rosiglitazone (Avandia, GlaxoSmithKline [GSK]) has found no evidence of any benefits of the drug over other diabetes medications and, because of side effects such as edema, fractures, and possible increased risk of MI, the review advises a "very cautious approach to rosiglitazone use" and recommends that if possible, other antidiabetic medications should be employed [1].

The review, which is published online on July 18, 2007, in the Cochrane Database of Systematic Reviews, was led by Dr Bernd Richter (Heinrich-Heine University, Düsseldorf, Germany). It examined only published studies of at least 24 weeks in duration of rosiglitazone in type 2 diabetics.

Richter commented to heart wire : "We are not adding any more information on cardiovascular risk. The recent New England Journal of Medicine meta-analysis by Nissen et al [2] has more information on cardiovascular risk than our study, as we included only published studies in diabetic patients, and the only one large enough to look at cardiovascular risk was the ADOPT trial. Our review is looking more at the bigger picture for rosiglitazone. We wanted to look at the whole risk/benefit ratio. We didn't find any benefits of rosiglitazone over other drugs for type 2 diabetes, but the drug is clearly associated with increased risks of edema, fractures, and weight gain. These adverse effects, together with the suggestion of an increased cardiovascular risk, lead us to conclude that doctors should think twice about using rosiglitazone, as we have good alternatives."

The review included 18 trials, which randomized 3888 patients with type 2 diabetes to rosiglitazone treatment. The median duration of treatment was 26 weeks. The longest duration was four years (the ADOPT trial). Richer et al report that in these studies rosiglitazone did not positively influence patient-oriented outcomes such as mortality, morbidity, adverse effects, costs, and health-related quality of life. Metabolic control as measured by glycosylated hemoglobin A1c (HbA1c) did not demonstrate clinically relevant differences from other oral antidiabetic drugs. Occurrence of edema was more than doubled; the single large randomized trial (ADOPT) indicated increased cardiovascular risk, and new data on raised fracture rates in women reveal extensive action of rosiglitazone in various body tissues, they add.

More details of adverse effects

While the percentage of overall adverse events was comparable between rosiglitazone and control groups in the studies reviewed, serious adverse events appeared more often after rosiglitazone treatment (median of 6% vs 4% in the control groups). Discontinuation rate following rosiglitazone administration was also higher than after control therapy (median of 7% vs 4%). Three studies evaluated and reported a more pronounced (apparently dose-related) decrease of hemoglobin after rosiglitazone intake in comparison with other active compounds or placebo, with hemoglobin reductions ranging between 0.5 and 1.0 g/dL. There were 11 studies that evaluated body weight and observed an increase of as much as 5.0 kg after rosiglitazone treatment; four studies described changes in body-mass index as increasing as much as 1.5 kg/m2.

Seven of the 18 included studies showed data on hypoglycemic episodes: compared with active monotherapy control, rosiglitazone treatment resulted in somewhat lower rates of hypoglycemia, especially when compared with sulfonylureas. Severe hypoglycemic events were rarely reported.

Data on edema were available in nine of 18 studies. Overall, 4739 participants provided information on the occurrence of edema. The total number of events was 287 in the rosiglitazone and 134 in the control groups, giving an odds ratio of 2.27 (95% CI 1.83-2.81).

The review also notes that the ADOPT trial showed higher fracture rates in women with rosiglitazone than with metformin or glyburide. "If you are a woman--and especially if you are thin--you probably should avoid this medication due to the risk of bone fractures," Richter said.

Cardiovascular risk

Noting that the Nissen meta-analysis included more studies than they did (Nissen included unpublished studies and studies in nondiabetes patients in addition to the studies in the Cochrane review) and found a significant increase in the risk of MI (OR 1.43, p=0.03) and a borderline significant increase in cardiovascular death (OR 1.64, p=0.06) with rosiglitazone, the Cochrane authors performed another meta-analysis using Nissen's data for the MI rates for type 2 diabetes patients only. They report that this analysis could not confirm significant differences in odds ratios for rosiglitazone vs controls, but all odds ratios (with the exception of the comparator glyburide--three studies only) indicated trends toward increased risk with rosiglitazone treatment.

They describe the meta-analysis performed by GlaxoSmithKline of 42 studies suggesting a 30% increased risk of cardiovascular events with rosiglitazone as "disturbing," adding: "Questions of timing of this information . . . arise. Ongoing trials using rosiglitazone (eg, RECORD) may provide additional data, but for a drug that was approved in 1999, the delay in obtaining information about the benefit/risk ratio is considerable."

Changes needed in drug-approval requirements

The Cochrane authors say they agree with the views expressed by others that current drug-approval requirements for antidiabetic medications and possibly all new drugs need to be changed. "The use of proxy indicators like metabolic control is not sufficient to approve drugs that many patients have to take for the rest of their lives," they write.

They conclude that the benefit/risk ratio of rosiglitazone therapy in type 2 diabetes mellitus needs urgent clarification and that patient-oriented end-point studies are urgently needed for the management of type 2 diabetes mellitus. But they add that it appears questionable whether new studies with rosiglitazone would be ethical, given the fact that less dangerous therapeutic alternatives exist.

GSK's statement in response

In a statement issued in response to the Cochrane report, GlaxoSmithKline says the review is another analysis of existing data that have previously been reported. GSK believes that the limited number of studies evaluated generate misleading conclusions and provide no new evidence about the use of rosiglitazone in clinical practice and research. The company points out that the studies included in the review primarily measured differences in blood glucose control over the short term, not mortality or morbidity outcomes, and it is therefore not surprising that positive conclusions were not drawn on this outcome.

It further notes that clinical guidelines have highlighted the importance of maintaining long-term glycemic control and that the ADOPT study did show a benefit in this regard of rosiglitazone over metformin and glyburide. The company also strongly disagrees with the Cochrane authors' questioning of the ethics of starting new studies with rosiglitazone. "We are committed to patient safety and carry out our clinical trials with the highest level of ethical conduct. The totality of the data--including long-term studies such as ADOPT and RECORD and an epidemiological analysis of more than 33 000 patients in a US managed care database--show that rosiglitazone has a comparable ischemic cardiovascular profile to other oral antidiabetic medicines. GSK stands firmly behind the safety profile of rosiglitazone when used appropriately," it states.

The Cochrane review in part contributes to the ongoing critical appraisal of randomized trials investigating the risk/benefit ratio of thiazolidinedione use by the German Institute for Quality and Efficiency in Health Care. Richer said he had no conflict of interest or association with any drug company.

  1. Richter B, Bandeira-Echtler E, Bergerhoff K et al. Rosiglitazone for type 2 diabetes mellitus. Cochrane Database Syst Rev 2007; Issue 3. Art No: CD006063. DOI: 10.1002/14651858.CD006063.pub2. Available at: https://www.cochrane.org .

  2. Nissen SE and Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med 2007; 356:2457-2471.

The complete contents of Heart wire , a professional news service of WebMD, can be found at www.theheart.org, a Web site for cardiovascular healthcare professionals.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.