Skip to main content

07-07-2010 | Cardiology | Article

BARI-2D results suggest rosiglitazone may not cause CV events


Meeting website

MedWire News: New results from a post-hoc analysis of the BARI-2D trial show no association between rosiglitazone and cardiovascular (CV) events.

However, the results confirmed previously reported links between rosiglitazone and fracture.

Richard Bach (Washington University School of Medicine, St Louis, Missouri, USA) presented the results of a post-hoc study of 2368 patients enrolled in the BARI-2D (Bypass Angioplasty Revascularization Investigation 2 Diabetes) study at the American Diabetes Association 2010 Scientific Sessions held in Orlando, Florida.

The patients all had Type 2 diabetes and angiographically-confirmed coronary artery disease (CAD) plus at least one significant lesion suitable for revascularization. The trial aimed to assess the long-term outcomes of rosiglitazone therapy.

The main results of BARI-2D have already been published. Patients were randomized to treatment with prompt revascularization and medical therapy versus medical therapy and delayed or no revascularization, and to insulin sensitization versus insulin provision. The findings showed that the rate of overall survival and freedom from CV events did not vary significantly between the different treatment groups, as previously reported by MedWire News.

For the purposes of this study, Bach and team compared 992 patients treated with rosiglitazone with 1199 patients who were not treated with a thiazolidinedione (TZD).

At baseline, the patients were aged 62 years on average, had a mean glycated hemoglobin level of between 7.5% and 7.8%, and had had diabetes for approximately 10 years. In addition, a quarter of patients in both groups had undergone prior revascularization.

Bach reported that, at 4.5 years, patients taking rosiglitazone had a significantly lower rate of stroke and the combination of death, myocardial infarction (MI), and stroke per 100 patient years compared with patients not treated with TZDs. In addition, the rate of death and MI were non-significantly lower, and chronic heart failure was non-significantly higher in rosiglitazone-treated patients.

Co-administration with insulin, metformin, nitrates, and angiotensin-converting enzyme inhibitors, did not significantly influence the rate of CV outcomes in rosiglitazone treated patients.

Notably, as found in previous studies, the relative risk for fracture was a significant 62% higher in rosiglitazone-treated patients than in those who took no TZDs. When stratified by gender, the increase in relative risk was statistically significant in women only, at 82%.

Bach acknowledged that the study was limited by its non-randomized nature and added that the fact that many patients on the study were taking more than one anti-diabetic drug might have limited the ability of the researchers to discriminate the effects of any one drug from another.

"I think these data are important because they suggest there is no significant cardiovascular harm posed by taking rosiglitazone for patients with Type 2 diabetes and coronary heart disease," said Bach.

These new data "contribute to the scientific inquiry regarding associated events with rosiglitazone."

He added: " There is an increase in fractures, but when one considers the dramatic morbidity and mortality associated with ischemic cardiovascular events in patients with diabetes, these data are reassuring."

MedWire ( is an independent clinical news service provided by Current Medicine Group, a trading division of Springer Healthcare Limited. © Springer Healthcare Ltd; 2010

By Helen Albert


Related topics