Breay W. Paty, MD, FRCPC (biography and disclosures)
Update September 23, 2010:
The FDA has announced severe restrictions on the access to Avandia in the United States. Current users of Avandia who are benefiting from the drug will be able to continue using the medication if they choose to do so. However, Avandia will be available to new patients only if they are unable to achieve glucose control on other medications and are unable to take Actos (pioglitazone), the only other drug in this class.
Furthermore, on the same date as the FDA announcement (September 23, 2010), the European Medicines Agency recommended that Avandia be withdrawn from the European market.
In addition, the US agency halted the TIDE (Thiazolidinedione Intervention with Vitamin D Evaluation) trial — the head-to-head cardiovascular safety trial of rosiglitazone versus pioglitazone. It was felt that, given the current state of evidence, the study’s outcome would not realistically yield a significant benefit to public health.
Although the evidence of CV risk of Avandia is still not considered conclusive, this action will have the practical effect of severely curtailing the prescribing of Avandia in the United States. These restrictions do not apply to Canada, but given that practice patterns are similar between the two countries, it will likely also have a chilling effect on Avandia prescriptions in Canada.
What I Did Before
Thiazoladinediones, such as rosiglitazone (Avandia®) and pioglitazone (Actos®), have been prescribed commonly in the past decade for the treatment of type 2 diabetes. These agents improve peripheral insulin sensitivity and delay the onset of diabetes in patients at risk. Since their introduction, certain contraindications and side effects have been well known, including worsening heart failure, edema (including macular edema) and weight gain. In the past, I avoided TZDs in anyone with NYHA Class III or IV heart failure, or peripheral or macular edema. Also, if a patient gained more than 10% of their pre-treatment body weight after starting a TZD, I would discontinue the drug. I still adhere to these guidelines, but I have become much more selective in how I prescribe TZDs.
What Changed My Practice
In 2007, a controversial meta-analysis showed a statistically significant increase in myocardial infarction, in patients taking rosiglitazone. Subsequent post-hoc analyses of other large studies, including ACCORD and VADT, failed to confirm this effect. Furthermore, a prospective analysis of the RECORD study showed no increased risk of death or hospitalization for cardiovascular disease (CVD) in patients taking rosiglitazone. However, because most of these studies were not originally designed to examine CV risk in patients taking rosiglitazone, the data is considered inconclusive and the question has not entirely been answered. During this period, another side effect emerged; increased risk of distal bone fractures in women on TZDs. This finding was first observed in the ADOPT trial in 2006 and subsequently seen in other TZD trials. These fractures are uncommon and do not appear to be related to osteoporosis and are not observed in men taking TZDs.
What I Do Now
To summarize: 1) the evidence for CV risk with TZDs is limited and the data is not conclusive; 2) TZDs improve glycemic control; but 3) there are other options for treating diabetes (including insulin). Consequently, my approach to this problem might be characterized as “proceed with caution”. I continue to prescribe TZDs for certain patients, such as younger (<50 years) men, who are failing metformin monotherapy, with A1C levels > 8.5% and no history of heart disease (either CVD or CHF). I avoid TZDs in patients with established CVD or who are at high risk of CVD (≥ 2 non-diabetes risk factors). I no longer prescribe TZDs for patients with any degree of heart failure (NYHA Class I through IV). Furthermore, I do not prescribe TZDs in post-menopausal women, or any woman at risk of fracture (e.g. smoking, family history of fractures, hyperparathyroidism, etc). If I am considering a TZD (usually as a second or third line agent), I discuss the treatment options with my patient, weighing the risks/benefits, including the risks of poor glucose control on microvascular complications. Until more definitive data emerges, I tend to err on the side of caution when it comes to CV risk and diabetes.
References: (Note: Direct article downloads and article requests require a login ID with the BC College of Physicians website)
- The Action to Control Cardiovascular Risk in Diabetes Study Group. Effects of Intensive Glucose Lowering in Type 2 Diabetes. N Engl J Med 2008; 358: 2545-59. (Article Request Form)
- Duckworth W. Abraira C, Moritz T, et al. Glucose Control and Vascular Complications in Veterans with Type 2 Diabetes. N Engl J Med 2009; 360: 1-11. (Article Request Form)
- Home PD, Pocock, SJ, Beck-Neilsen H, et al. Rosiglitazone evaluated for cardiovascular outcomes in oral agent combination therapy for type 2 diabetes (RECORD): a multicentre, randomized, open-label trial. Lancet 2009; 373: 2125–35. (Full Article)
Loading ...
I think this gives us guidance in the use of TZDs with all the differing opinions floating out there. It is a balanced, cautious approach until more confirmatory evidence comes to light which is quite helpful.
Good summary of the evidence and good decision support for practice. Are you aware of any existing (well designed, specific to this question) study to clarify the cardiovascular risk?
A major trial is underway, sponsored by GlaxoSmithKline (manufacturers of Avandia), called the Thiazolidinedione Intervention With Vitamin D Evaluation (TIDE) trial.
This study will test the cardiovascular effects of long-term treatment with rosiglitazone or pioglitazone compared to standard of care in patients with type 2 diabetes with a history or risk of cardiovascular disease.
The second question will compare the effects of long-term supplementation of vitamin D on death and cancer.
The enrollment will be 16,000 subjects (large trial), but the expected completion date is not until 2015. So, the question will remain open until that time.
I have definitely seen Avandia associated with worsened macular edema in my patients. Most of my patients are no longer on this drug for diabetes.
A clear summary of an approach. Everyone agrees to start with metformin. Are sulfonylureas not the most commonly used second agents? I have tended to use these two agents and then go to insulin.
A clear summary of an approach. Everyone agrees to start with metformin. Are sulfonylureas not the most commonly used second agents? I have tended to use these two agents and then go to insulin.
There are lots of benefits of using TZDs in patients with T2DM. What I do?
Although not fully conclusive, but once there are reports of increased CV events and controversies regarding rosiglitazone since 2007, I have deleted rosiglitazone from my prescription list but continuing pioglitazone.
Pio acts as a PPAR-gamma agonist and in addition to this it has also PPAR-alpha stimulating activity to a lesser extent.
Pio leads to weight gain but this weight gain mainly due to increase in subcutaneous fat (not visceral) having high adiponectin activity.
Of course I categorically avoid pio to patients with heart failure having II, III, IV stages according to NYHA classification.
Once mono-therapy is required, if patient can not tolerate metformin (it must be the first line) why not to go for pio before going to SU to preserve residual beta cells?
It’s a real shame there is only one agent with benefits on hard clinical endpoints – metformin. It’s one thing to lower the a1c, but unless there’s benefits on CV, renal, eye and neuropathy, we are fooling ourselves we are making a meaningful difference. Sulfonylureas, TZDs, insulin – all have very shaky evidence for benefit and some evidence of harm (wt gain, increased insulin resistance, CHF), despite the improvements in a1c.
The benefits of improved glycemic control on microvascular complications of diabetes, such as retinopathy, nephropathy and neuropathy, have been confirmed in several large studies (DCCT, UKPDS, ADVANCE). The cardiovascular benefits are less well established, but appear to be evident if good glycemic control is started at a younger age, before the duration of diabetes is > 15 years. Trials, such as VADT and UKPDS 10-year follow-up suggest that patients with better glycemic control early in the course of their disease have lower rates of CV events in the future. However, once CVD is estastablished, the clear benefits of tight glycemic control become less evident.
I recommend using any and all appropriate agents (including TZDs) in younger patients, to reduce the long-term risk of CV complications. However, once CV disease or CV risk factors are established, more caution is needed in selecting agents and establishing glucose targets.
In the poll section at the end of the article, you should maybe think of putting tick box for “I already do this”, because in this case, well, I already do this! :)
In July 2010, the United States FDA held Avisory Committee hearings on the cardiovascular safety of Rosiglitazone (Avandia). The conclusions were mixed and no firm consensus was reached about the overall CV risk of Avandia or of TZDs in general.
On July 21st, the TIDE trial was placed on partial clinical hold, so that no new patients could be enrolled, but patients already enrolled would be allowed to continue in the trial. This would allow time for the participating IRBs to review the FDA findings and update consent forms as needed. It is not yet clear whether the trial will be halted entirely.
Overall, no fundamentally new data arose from the FDA hearings. The evidence was not sufficient to draw a firm conclusion about the CV risks of Avandia or TZDs in general.
It is up to each individual practitioner to make his/her own decision regarding the safety of TZDs. One could argue that, since there are other ways to treat diabetes, we can avoid TZDs altogether. However, although I am cautious about the use of TZDs, I continue to feel that these agents have a role to play in certain diabetic patients (mostly young (<50), obese men). Also, there is growing evidence for benefit of TZDs in non-diabetes settings, such as non-alcoholic steato-hepatitis (NASH).
Unfortunately, until we have more evidence, practitioners will have to manage with uncertainty.
Hmmn- maybe ‘normalizing’ blood sugar in this fashion doesn’t correct the underlying issues- we need to understand what they are. Maybe this makes cells ‘sicker’.
I had great response with Actos in a number of very ill type 2 diabetic patients with CHF and renal failure. I saw them almost weekly and worked very hard together to manage BP/ glucose/ diet/ weight and all the metabolic abnormalities that arise in this group PLUS the CHF component.
I wonder – many things. Were these patients adequately managed for BP. cholesterol!!! etc/
And again the underlying deficit in type 2 diabetes is obesity/ glucose intolerance ( I have met skinny elderly type 2 diabetics) That is what really needs correction- bariatric surgery is more effective than anything. —
Be simpler if people could just stop eating-eh?
The benefits of TZDs are modest as far as I can tell. As some of the previous commentators have suggested, using metformin, salphonylureas and insulin seems a tried and trusted approach.