This article is more than 5 years old. Information may no longer be current.
Heart disease and diabetes: a continuing controversy
Click Here to Manage Email Alerts
After the presentations at the annual meeting of the American Diabetes Association in June, it is now as if nothing new was presented there. July saw the presentations at an FDA Advisory Committee meeting return again and again to the same old refrain of possible lack of cardiovascular safety for some antidiabetic agents. More astonishing, however, were the expressions in support of a new requirement for cardiovascular outcome trials for new antidiabetic medications, despite the trials presented at the ADA. How have we come to such an end?
In the beginning there was much epidemiology showing a clear relationship between hyperglycemia, diabetes and excess cardiovascular risk. Numerous studies in the United States and in Europe showed such a nexus. But, of course, this was insufficient in itself to mandate good glycemic control. Therefore, a randomized multicenter intensive treatment study in patients with type 1 diabetes was organized and named the Diabetes Control and Complications Trial. This study of 1,441 patients eventually showed clear interventional benefits with good glycemic control and the complications of retinopathy, nephropathy and neuropathy, but no cardiovascular benefit. The long-term observational extension, EDIC, has seemed to indicate some residual coronary heart disease benefit in the intensively treated cohort after years of nonintensive follow-up.
|
|
Two studies in patients with type 2 diabetes were later undertaken to define the relationship between glucose control and diabetic complications, including macrovascular complications in type 2 disease. Both the Kumamoto Study and the much larger United Kingdom Prospective Diabetes Study returned essentially similar results: Namely that there were equivalent reductions in microvascular endpoints, particularly hard endpoints in the UKPDS, but no CHD benefits to intensive glycemic control in patients with type 2 diabetes, similar to what was seen in the DCCT in patients with type 1 diabetes.
So, at the end of 1998, there seemed to be no use to controlling glucose with regard to CHD, but much to be gained with regard to microvascular complications, which are the leading cause of diabetic disability in the United States. This dichotomy of benefit seemed to fly in the face of the natural order of things and clearly upset numerous diabetes specialists who firmly believed that if the glucoses could be normalized, all would be well in their patients with diabetes. Accordingly, three additional intensive treatment multicenter cardiovascular trials were organized around the globe to try to set things right and properly define the relationship, if any, between CHD and glycemic control.
More recent findings
The findings of all three studies were presented at the ADA meeting in June. The first of these, the ACCORD study, was organized and funded by the National Heart, Lung, and Blood Institute. It was composed of more than 10,000 patients arranged in a complex 3x2 factorial treatment design to investigate glycemic control, hypertension management and dyslipidemia interventions. Unfortunately, the glycemic control arm did not prespecify a glucose control algorithm but rather a control target HbA1c of <6% compared with a conventional target of 7.5%. Both arms had much lower than expected rates of CHD based upon historical data. Although total CHD events were similar in both intensive and conventional arms, there were 54 excess deaths in the intensive arm, with correspondingly fewer nonfatal events.
This glucose control arm of the study was stopped by the Data Safety Monitoring Board in order to protect patient welfare. No single treatment was identified by the investigators as causal for the excess fatal events, including insulin and rosiglitazone. However, the data suggested that the policy of intensive control, which was associated with a higher frequency of hypoglycemia and weight gain, did account for this excess risk.
On the other hand, ADVANCE was an intensive treatment protocol prespecifying initial therapy with long-acting gliclazide. It was performed in Australasia with a similarly sized population somewhat less involved with pre-existing heart disease. Of interest was the finding in ADVANCE that reconfirmed the findings of microvascular benefit but not CHD benefit with good glycemic control. There was, however, no adverse experience with CHD and good control. Unlike ACCORD, ADVANCE showed, if anything, a trend toward benefit, not harm, with regard to CHD.
Similarly, the findings of the VA Diabetes Trial were also presented in June at the ADA Annual Meeting. Here again, in an 1,800-patient population with a high incidence of pre-existing CHD, there appeared to be no conclusive benefit to glucose control and CHD risk reduction, although a microvascular benefit seemed evident. In both of these studies, specific agents were not associated with excess risk, not insulin nor rosiglitazone, nor was there specific benefit either.
Now, a few months later, what is it that we can conclude about diabetes control and CHD risk with regard to specific anti-diabetic agents?
Despite the epidemiologic evidence, there are no interventional data to suggest that any strategy for good glycemic control can yield a cardiovascular benefit. This is supported by all six trials thus far reported. On the other hand, one trial does suggest harm for glycemic control efforts, but not the other five and not the epidemiologic database. All trials to date show clear benefit with respect to microvascular complications of diabetes. No trial shows long-term harm, although there may be some short-term disadvantage to over-rapid control with respect to retinopathy and neuropathy. No single anti-diabetic agent uniquely benefits nor harms patients with diabetes.
Thus, it is clear that we at long last must separate microvascular from macrovascular targets and identify separate treatment strategies for each. Multiple therapies will be necessary to achieve the glucose control necessary to minimize microvascular disease. This is a highly useful goal. Other unrelated therapies already exist to lower CHD risk. Many are in widespread use, including ACE inhibitors and angiotensin receptor blockers, aspirin and statins in particular. It is perhaps for this reason that present rates of CHD are so low as to make a CHD interventional trial all but impractical at present.
Now, however, members of the FDA Advisory Board have voted for CHD trials to evaluate new drug safety in patients with diabetes. The necessity of this action seems to spring from the findings of the somewhat peculiar situation with rosiglitazone. Here we saw a meta-analysis show some increased CHD risk, but no support for this finding in any of the long-term randomized trials then or now with the newer data from the three trials presented at ADA. These more traditional means to evaluate drug efficacy and safety are highly reassuring. Thus, there clearly seems to be a diminishing need for CHD safety trials as newer trial data become available, and it also appears that such trials will be fundamentally flawed by their likely inability to show benefit in any comparator arm.
Further, because of the ever-decreasing rates of CHD in patients with diabetes, it seems unlikely that sufficient numbers of patients can be recruited for such a study, or that the data can be adequately analyzed given that diabetes is a progressive disease and that ever greater amounts of medicines are required over time to control patients’ blood sugars.
It is therefore necessary to conclude that the nexus between glycemia and CHD is unproven, but that between glycemia and microvascular disease is clear. We should treat for microvascular benefit; we should treat along different algorithms for CHD. Both treatment algorithms may be unrelated but neither should be overlooked.
Alan J. Garber, MD, PHD, is a Professor in the Departments of Medicine, Biochemistry and Molecular Biology, and Cellular & Molecular Biology at Baylor College of Medicine, Houston; he is the Chief Medical Editor of Endocrine Today.