EVIDENCE-BASED ANSWER
LIFESTYLE CHANGES AND SOME DRUGS CAN HELP. Lifestyle interventions aimed at weight loss of 5% to 10% of body weight along with moderate aerobic exercise such as brisk walking for 150 minutes a week are the most effective means to prevent impaired fasting glucose (IFG) or impaired glucose tolerance (IGT) from progressing to diabetes (strength of recommendation [SOR]: A, several meta-analyses, including a recent Cochrane review).
Effective pharmacologic interventions include metformin (SOR: A, meta-analysis), acarbose (SOR: A, meta-analysis), and orlistat (SOR: B, meta-analysis).
Although thiazolidinediones, such as rosiglitazone, can decrease the rate of progression to diabetes (SOR: B, randomized controlled trial [RCT]), they pose a significant risk of fluid overload and heart failure. Angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers aren’t recommended for the express purpose of preventing diabetes in patients with IGT or IFG (SOR: B, RCTs with inconsistent results).
Evidence summary
Patients with either IFG or IGT have a significant risk of progressing to diabetes. A recent Cochrane review evaluated studies that randomized patients to intensive exercise and diet counseling or standard advice with follow-up for 1 to 6 years.1 Although treatment arms varied from study to study, exercise recommendations averaged at least 150 minutes a week and diet recommendations included counseling by dieticians. These lifestyle interventions reduced progression to diabetes by 37%. Another meta-analysis showed a relative risk reduction of 49% for lifestyle changes.2
Metformin is the most effective drug; acarbose and orlistat also help
This meta-analysis also encompassed 10 RCTs that evaluated the effectiveness of pharmacologic interventions, including oral diabetes drugs (metformin, phenformin, acarbose, glipizide, flumamine) and an antiobesity drug (orlistat).2 Metformin was the most effective drug, with a 31% to 35% relative risk reduction (number needed to treat [NNT]=7-14), although acarbose and orlistat also produced significant reductions. In a small study, glipizide didn’t prevent development of diabetes.
A 2006 Cochrane review of acarbose encompassed 5 RCTs, including the Study to Prevent Non-Insulin-Dependent Diabetes (STOP-NIDDM), which showed a relative risk reduction of 22%.3
Rosiglitazone offers benefit, but also hazard
The Diabetes Reduction Assessment with ramipril and rosiglitazone Medication (DREAM) Trial of 5269 participants with IGT or IFG compared treatment with rosiglitazone and placebo.4 Rosiglitazone reduced development of diabetes for participants (hazard ratio=0.40; 95% confidence interval [CI], 0.35-0.46; P<.0001; NNT=5).
However, the risk of heart failure was higher in the rosiglitazone arm, with a risk ratio of 7.00 (95% CI, 1.59-30.76) and a number needed to harm of 250. No statistically significant difference was noted in cardiovascular-related deaths. This trial is consistent with recent evidence showing an increased risk of fluid overload and heart failure in patients with diabetes who take thiazolidinediones.5
ACE inhibitors aren’t routinely recommended
Early evidence suggested that blockade of the renin-angiotensin system might delay the onset of diabetes.6 The DREAM trial included a ramipril arm to evaluate for this effect in IGT and IFG.7 After 3 years of follow-up, no significant decrease in the incidence of diabetes was found among participants taking ramipril compared with placebo.