Return to Normoglycemia from Prediabetes Associated with Reduced Risk of Developing Diabetes

In the Diabetes Prevention Program (DPP) trial, both lifestyle modification and metformin treatment were associated with reduced progression to diabetes in patients with prediabetes (N Engl J Med 2002 Feb 7;346(6):393), and several other trials have shown similar results. A large cohort of patients from the DPP trial had continued follow-up for diabetes outcomes in the DPP Outcome Study. A newly published analysis investigated the association between regression to normoglycemia and risk of progression to diabetes in 1,990 patients from this cohort who had not progressed by the end of the original DPP trial. Of these patients, 37% had been randomized to lifestyle modification, 32.5% to metformin, and 30.5% to placebo. Median follow-up was 3.2 years during the DPP trial and 7.5 years during the cohort study. All patients were assessed yearly for glycemic control by oral glucose tolerance tests.

Normoglycemia was defined as fasting plasma glucose < 5.6 mmol/L (100 mg/dL) and 2-hour plasma glucose < 7.8 mmol/L (140 mg/dL). Prediabetes was defined as fasting plasma glucose 5.6-6.9 mmol/L (100-124 mg/dL) and/or 2-hour plasma glucose of 7.8-11 mmol/L (140-198 mg/dL). Higher levels indicated progression to diabetes.

During the DPP trial, 894 patients (45%) had at least 1 glucose tolerance test on which they regressed to the normoglycemic range (including 53.5% of the lifestyle modification group, 42.3% of metformin group, and 37.2% of placebo group). The remaining 1,096 patients had consistent tests indicating prediabetes during the trial.

Regression to normoglycemia on at least 1 test during the DPP trial was associated with reduced risk of progression to diabetes compared to consistent prediabetes (hazard ratio 0.44, 95% CI 0.37-0.55). There were no significant differences among the 3 randomized groups in the relationship between regression to normoglycemia and progression to diabetes. Diabetes risk was reduced regardless of the manner in which normoglycemia was achieved, even with placebo.

Subgroup analyses were performed in patients who failed to achieve normoglycemia during the trial to compare the effects of the randomized treatments on later glycemic control during observational follow-up. Of note, failure to normalize glucose with lifestyle modification was associated with decreased likelihood of later regression to normoglycemia (odds ratio 0.59, 95% CI 0.42-0.82) and increased risk of progression to diabetes compared to placebo (hazard ratio 1.31, 95% CI 1.03-1.68). There were no significant differences in regression to normoglycemia or progression to diabetes comparing metformin vs. placebo (Lancet 2012 Jun 8 early online). These data suggest that patients who are diagnosed with prediabetes can be stratified regarding their risk for progression to diabetes by whether or not they subsequently have normal values on glucose tolerance testing.

For more information, see the Prediabetes and Risk factors for diabetes mellitus type 2 topics in DynaMed.