Therapeutic Use of Metformin in Prediabetes and Diabetes Prevention

Abstract

People with elevated, non-diabetic, levels of blood glucose are at risk of progressing to clinical type 2 diabetes and are commonly termed 'prediabetic'. The term prediabetes usually refers to high-normal fasting plasma glucose (impaired fasting glucose) and/or plasma glucose 2 h following a 75 g oral glucose tolerance test (impaired glucose tolerance). Current US guidelines consider high-normal HbA1c to also represent a prediabetic state. Individuals with prediabetic levels of dysglycaemia are already at elevated risk of damage to the microvasculature and macrovasculature, resembling the long-term complications of diabetes. Halting or reversing the progressive decline in insulin sensitivity and β-cell function holds the key to achieving prevention of type 2 diabetes in at-risk subjects. Lifestyle interventions aimed at inducing weight loss, pharmacologic treatments (metformin, thiazolidinediones, acarbose, basal insulin and drugs for weight loss) and bariatric surgery have all been shown to reduce the risk of progression to type 2 diabetes in prediabetic subjects. However, lifestyle interventions are difficult for patients to maintain and the weight loss achieved tends to be regained over time. Metformin enhances the action of insulin in liver and skeletal muscle, and its efficacy for delaying or preventing the onset of diabetes has been proven in large, well-designed, randomised trials, such as the Diabetes Prevention Program and other studies. Decades of clinical use have demonstrated that metformin is generally well-tolerated and safe. We have reviewed in detail the evidence base supporting the therapeutic use of metformin for diabetes prevention.

Figures

Fig. 1

Prevalence of prediabetes in three cohorts without a prior diagnosis of diabetes in the US, i.e. Screening for Impaired Glucose Tolerance (SIGT), the Third National Health and Nutrition Examination Survey (NHANES III), and the National Health and Nutrition Examination Survey 2005–2006 (NHANES 2005–2006). Adapted from data presented by Rhee et al. [16]. IFG impaired fasting glucose, IGT impaired glucose tolerance

Fig. 2

Mean changes in weight during the randomised phase of the Diabetes Prevention Program. Placebo and metformin were administered in combination with standard lifestyle advice. Adapted from data presented by the Diabetes Prevention Program Research Group [19]

Fig. 3

Effects of treatments in the Diabetes Prevention Program on the risk of diabetes following stratification of the population for age, FPG or BMI at baseline. a Age at baseline; b BMI at baseline; c FPG at baseline. Comparisons shown are for ILI vs. P, M vs. P, and ILI vs. M. In each case, a more strongly negative change in risk signifies greater efficacy of the first named agent. Adapted from data presented by the Diabetes Prevention Program Research Group [19]. FPG fasting plasma glucose, BMI body mass index, ILI intensive lifestyle intervention, M metformin, P placebo

Fig. 4

Main results of a meta-analysis of diabetes prevention studies with metformin. See text for a description of the ITT/worst-case scenario. Adapted from data presented by Lily and Godwin [127]. ITT intention-to-treat analysis

Fig. 5

Proportions of prediabetic subjects in the US meeting ADA criteria for treatment with metformin. Columns show estimated proportions meeting ADA criteria for use of metformin for IGT or IFG, pooled from data presented for three cohorts. Adapted from data presented by Rhee et al. [16]. ADA American Diabetes Association, IGT impaired glucose tolerance, IFG impaired fasting glucose