NPC Archive Item: Cardiovascular safety of glitazones in older people

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review

NPC Logo

Lipscombe L, Gomes T, Lévesque LE, et al. Thiazolidinediones and cardiovascular outcomes in older patients with diabetes. JAMA. 2007;298:2634-43.

A large population-based study from Canada increases concerns over the cardiovascular risk of glitazones, and in particular rosiglitazone, in older people with diabetes, regardless of their baseline cardiovascular (CV) risk.

What is the background to this?

Patients with diabetes are at an increased risk of CV disorders, including heart failure (HF) and ischaemic heart disease, due to the underlying condition. The increased risk of HF with glitazones is well recognised. In the December Safety Update from the Medicines and Healthcare Products Regulatory Agency (MHRA) advice is given for their use in people at risk from HF, and for the close monitoring of patients during treatment for signs and symptoms of fluid retention, including weight gain and oedema. Furthermore, recognising that rosiglitazone might be associated with a small increased risk of cardiac ischaemia, particularly in combination with insulin, the MHRA advise that rosiglitazone should only be used in patients with previous or current ischaemic heart disease after careful evaluation of individual risk.

What does this study claim?

The authors claim that this is first study to evaluate glitazone-related outcomes in an entire population of older people with diabetes. It demonstrates an increase risk of HF, MI and all-cause mortality with the use of glitazones (mainly rosiglitazone) compared with non-glitazone oral hypoglycaemic therapy. They estimate the numbers needed to harm (NNH) over four years to be as low as 34 for HF, 26 for MI, and 22 for death in the population studied.

How does this relate to other studies?

Older people form a large proportion of the population of patients with diabetes, and are often underrepresented in many RCTs. This study provides an insight into the effects of glitazones in “real-life” clinical practice where treatment of older people with diabetes is often complicated by comorbidities that put them at higher risk. The data provides support for the view (see MeReC Extras 29 and 30 and NPC blog) that there is a greater risk of cardiovascular events with rosiglitazone than with the older more established oral hypoglycaemic agents (i.e. metformin and sulphonylureas). GlaxoSmithKline point out in a press release in response to this article that not all epidemiological studies have identified an increased risk of cardiovascular events or death. However, it could equally be pointed out that no studies have demonstrated a significant reduction in these outcomes —something, that might reasonably be expected of an effective anti-diabetic medicine.

So what?

Older people with diabetes are at high baseline cardiovascular risk. Any increased risk of CV events or death from use glitazones (or rosiglitazone at least) may outweigh any benefits obtained from improved glycaemic control.


Prescribers should only prescribe glitazones (and rosiglitazone in particular) in people with diabetes only after carefully consideration of the benefits and their potential harms (including the risk if cardiovascular events) on a case-by-case basis.

Our view is that the time has come for prescribers to begin proactively reviewing patients taking rosiglitazone and discussing these data and alternative strategies. It seems less clear whether there are similar issues with pioglitazone.

Clinicians and patients should bear in mind that controlling cardiovascular risk in type 2 diabetes by smoking cessation, controlling blood pressure, and improving the TC:HDL ratio with a statin reduces macrovascular events, whereas intensively controlling blood glucose – at least with insulin and sulphonylureas – does not (see the Type 2 diabetes floor on NPCi).

Study details:

This was a large Canadian, retrospective, nested case-control study of 159,026 people with diabetes aged 66 years or older (median follow-up 3.8 years). Current treatment with glitazone monotherapy was associated with a significantly increased risk of HF, MI or all cause mortality when results were adjusted for a wide range of confounding factors including insulin therapy, income, disability, previous coronary heart disease or renal disease, and use of other drugs — rate ratio [RR] 1.60; 95% confidence interval [CI] 1.21 to 2.10; P<001), acute MI (RR 1.40; 95%CI 1.05 to 1.86; P=0.02), and death (RR 1.29; 95%CI, 1.02-1.62; P=0.03) compared with other oral hypoglycaemic agent combination therapies. Although a statistically significant difference was evident in those patients receiving rosiglitazone for all of these outcomes (P<0.001, P<0.001, and P=0.005, respectively). Significant increases in the risk of these outcomes were also found for these outcomes when current combination therapy or past treatment with rosiglitazone (or glitazones as a whole) was compared with other oral hypoglycaemic agent combination therapies. No significant increased risk for pioglitazone could be demonstrated for any of the outcomes. However, the possibility of an increased risk with pioglitazone can not be ruled out, as the number of events in these patients was low and the confidence intervals wide.