| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
COMMENTARY |
Tulane University Health Sciences Center (V.A.F.), New Orleans, Louisiana 70112; and VA Puget Sound Health Care System and University of Washington (S.E.K.), Seattle, Washington 98108
Address all correspondence and requests for reprints to: Vivian A. Fonseca, M.D., Professor of Medicine, Director, Diabetes Program, Tulane University Medical Center, 1430 Tulane Avenue, SL 53, New Orleans, Louisiana 70112-2699. E-mail: vfonseca{at}tulane.edu.
The old adage that "prevention is better than cure" holds particularly well for a chronic disease like type 2 diabetes, with its epidemic scale and multiple crippling complications. Clearly, any intervention that can halt this epidemic safely and effectively will have huge public health implications.
Several clinical trials have therefore attempted to address the question of preventing diabetes. Among these, the Diabetes Prevention Program clearly demonstrated that intensive lifestyle change was effective not only in decreasing the progression of impaired glucose tolerance to diabetes, but was also associated with significant improvement in other cardiovascular risk factors (1). The Finnish Diabetes Prevention Study reported similar findings (2). One would think that therein lies the simple answer to diabetes prevention. Unfortunately, translation of these results to the general public has been difficult. Large, randomized trials have also assessed the potential of the medications metformin (Diabetes Prevention Program) and acarbose (STOP-NIDDM) to slow the progression to diabetes, and these have both demonstrated some benefit (1, 3). When one compares the results in the Diabetes Prevention Program, metformin was only about half as effective as was intensive lifestyle change. Acarbose, although not compared head-to-head with intensive lifestyle, appears to be no better than metformin. At this time, neither of these drugs is currently approved for diabetes prevention. Whether they ever will be approved is another question. With the availability of other classes of agents with different mechanisms of action, many pharmaceutical companies and clinical trialists continue to investigate the effect of pharmacological therapy to prevent diabetes.
The most recent large clinical trial to address the prevention of diabetes has been the Diabetes Reduction Assessment with Ramipril and Rosiglitazone (DREAM) trial (4, 5), for which there was a good rationale. First, post hoc analyses of a variety of studies have shown that blockade of the renin-angiotensin system decreases incident diabetes. In particular, the HOPE study demonstrated a significant reduction in new-onset diabetes in people with increased cardiovascular risks (6). Second, rosiglitazone is a thiazolidinedione insulin-sensitizer, and a previous drug in this class, which is no longer available, had been shown to prevent diabetes in a small single-center study in Hispanic women with a history of gestational diabetes (7); the same was suggested in a discontinued arm in the Diabetes Prevention Program (8).
The DREAM trial was an international effort performed in 5269 individuals (59% women) of whom 44% had a history of hypertension, 44% were current or former smokers, and 36% had dyslipidemia. The results of the study were mixed (4, 5), which for the most part was not unexpected. The effect of ramipril (titrated to 15 mg/d) on glucose metabolism was small. Although the study demonstrated a modest glucose-lowering effect, the drug did not slow the rate of development of diabetes. In contrast, rosiglitazone (titrated to 8 mg/d) very significantly reduced the development of diabetes by some 60% in this relatively healthy population with impaired fasting glucose or impaired glucose tolerance. Technically, therefore, the study outcome was positive in meeting its primary endpoint and clearly demonstrated efficacy of rosiglitazone for prevention. However, enthusiasm for clinical application of the results must be balanced by the occurrence of known adverse effects, particularly weight gain, which was 2.2 kg greater with rosiglitazone, and an increase in congestive heart failure events. Although the number of congestive heart failure events was small (14 with rosiglitazone vs. two with placebo, producing a hazard ratio of 7.0 with very broad confidence intervals of 1.60–30.9), it needs to be viewed in the context of the relatively healthy study cohort. For every 1000 rosiglitazone-treated people over 3 yr, about 144 cases of diabetes will be prevented with an excess of four to five cases of congestive heart failure. Expressed in another way, the "number needed to harm" would be 238 patients treated for 3 yr to subject one patient to heart failure. Clearly, this is a daunting adverse effect for those affected, and further information on the severity, reversibility, and predispositions to this side effect are needed, as the investigators have said will be forthcoming. A related observation was the apparently higher prevalence of congestive heart failure in participants on a combination of rosiglitazone and ramipril. Although the 2 x 2 factorial design provided an opportunity in a single study to efficiently examine the effect of each drug vs. placebo, it did not allow for a direct comparison between the two active agents (9). Although there was no interaction between these drugs in terms of efficacy, the study design leaves us less well informed about potential side effect interactions between these medications with overlapping actions (ramipril may have insulin sensitizer effects and rosiglitazone may have vasodilating effects).
What conclusions can we draw from this well-designed and well-executed study? First, it once again demonstrates the importance of large clinical trials with major outcomes as endpoints. If it were not for this trial, many would still think ramipril can decrease progression to diabetes. Even negative studies are useful in formulating guidelines and treatment policies.
Second, thiazolidinediones clearly lower glucose, even at very early stages of disordered glucose metabolism, and technically prevent diabetes; but does this effect occur only while individuals are taking the medication, or is the drug altering a fundamental abnormality in the long term and providing some sustained protective effect? A washout period was included in the study design and the results will soon be announced. Because altering the progression of hyperglycemia prevents microvascular complications and possibly even macrovascular disease, a safe approach that delays microvascular complications is worthwhile, whether we actually prevent diabetes or not.
Finally, the increased congestive heart failure in this study is clearly of concern, particularly because it occurred in a relatively healthy population. At the same time, it is important to remember that, in the PROACTIVE trial, pioglitazone was also associated with increased congestive heart failure, but its onset did not lead to increased mortality (10). Clinicians will ask what can be done to minimize the fluid retention and weight gain that may have precipitated heart failure in these patients. This area of research has been largely neglected despite the fact that this side effect has been evident since these compounds were first developed. Indeed, it has been shown that weight loss is possible with calorie restriction in patients treated with a thiazolidinedione (11). One is further reminded that the DREAM study did not examine effectiveness of the two agents vs. lifestyle intervention, or what may be the most important: combination treatment, intensive lifestyle change plus an insulin sensitizer.
Thus, the DREAM trial provides some enlightenment about what is possible today in diabetes prevention. Although increased risk of a serious side effect makes it unlikely at this time that it will lead to major changes in clinical guidelines for drug treatment to prevent type 2 diabetes, in individuals who cannot or will not change their lifestyles, clinicians should discuss the study results, including risks of adverse events, and help them make an informed decision about their choice of therapies to prevent diabetes.
 |
Footnotes |
|---|
First Published Online October 24, 2006
Received October 3, 2006.
Accepted October 18, 2006.
|
References |
|---|
 Top References |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
