In patients at increased risk of developing type 2 diabetes mellitus, is metformin effective for the prevention or delay of diabetes onset and its associated complications?
In patients at increased risk of developing type 2 diabetes, metformin reduces the risk (number needed to treat [NNT] = 7; 95% CI, 6 to 10) compared with counseling on standard diet and exercise. Data are limited regarding adverse effects and long-term outcomes.1 (Strength of Recommendation: B, based on inconsistent or limited-quality patient-oriented evidence.)
Patients with prediabetes, defined as impaired fasting glucose, impaired glucose tolerance, and/or an A1C of 5.7% to 6.4%, are at increased risk of developing type 2 diabetes. Type 2 diabetes has been associated with many complications, including neuropathy, retinopathy, cardiovascular disease, and kidney dysfunction.1 According to the Centers for Disease Control and Prevention, 13% of the U.S. population 18 years and older were diagnosed with type 2 diabetes in 2018.2 From 2013 to 2016, 34.5% of the U.S. population 18 years and older were diagnosed with prediabetes2; 15% to 30% of these patients progress to type 2 diabetes within five years.3
The authors of this Cochrane review investigated whether metformin reduces the risk of type 2 diabetes in patients with prediabetes.1 The review included 20 randomized controlled trials and 6,774 patients. The duration of intervention ranged from one to five years. The authors included trials that compared metformin with any pharmacologic intervention, behavioral intervention, placebo, or standard of care in populations with prediabetes. The primary outcomes of interest were all-cause mortality, incidence of type 2 diabetes, serious adverse effects, cardiovascular mortality, nonfatal myocardial infarction or stroke, health-related quality of life, and socioeconomic effects.
Patients who used metformin for one to five years were less likely to develop type 2 diabetes compared with patients counseled on standard (not intensive) diet and exercise (NNT = 7; 95% CI, 6 to 10). However, there was no significant reduction in the development of type 2 diabetes in patients taking metformin compared with those who were counseled on intensive diet and exercise. The Cochrane review did not define intensive vs. standard diet and exercise counseling. Additionally, there was no significant reduction in the development of type 2 diabetes in patients who were given metformin plus intensive diet and exercise counseling compared with patients who received only intensive diet and exercise counseling. Metformin ($220 to $1,177) was more expensive than the standard diet and exercise intervention ($61 to $184). However, the cost of the intensive diet and exercise intervention ($225 to $3,628) was higher than that of metformin.
This review had some limitations. Notably, 48% of the patients were from one study, and 60% of the reviews were conducted in a single country (China). The results were consistent across studies, although the results of some studies were not statistically significant. Importantly, there were limited data on mortality between treatment groups (complications were rare), as well as long-term outcomes.
A previous meta-analysis also found that metformin reduced the risk of developing diabetes (NNT = 14; 95% CI, 10 to 22).4 Guidelines from the National Institute for Health and Care Excellence recommend metformin as an adjunct to intensive lifestyle modification or for patients incapable of lifestyle modification.5 The American Diabetes Association recommends the use of metformin to reduce the risk of developing diabetes in high-risk individuals but suggests that it is inferior to lifestyle modifications.6 It should be noted that within the United States, the cost of metformin is typically low. It is often included on the low-cost medication options at pharmacies for patients who need to pay out of pocket for their medications.
The practice recommendations in this activity are available at http://www.cochrane.org/CD008558.
Editor's Note: The numbers needed to treat and confidence intervals reported in this Cochrane for Clinicians were calculated by the authors based on raw data provided in the original Cochrane review.