Home / Resources / Articles / Metformin for Diabetes Prevention Is Safe, Well-Tolerated

Metformin for Diabetes Prevention Is Safe, Well-Tolerated

Mar 28, 2012

Long-term treatment with metformin is safe for preventing or delaying the development of type 2 diabetes….

New data from the open-label Diabetes Prevention Program Outcomes Study (DPPOS) demonstrate that metformin is linked to “modest but durable weight loss” of 2% body weight over the course of 10 years, and appears to be safe and well tolerated. The findings support the conclusions of the original 3-year DPP double-blind study showing that use of metformin (850 mg twice daily) encouraged stable weight loss. In both the metformin and placebo groups, weight loss was a strong predictor of diabetes prevention.


According to the Diabetes Prevention Program Research Group, the pattern of metformin-associated weight loss appears to differ from that observed with caloric restriction, in that adipose tissue is affected more than lean tissue mass. Metformin may also mimic the effects of exercise.

Vivian Fonseca, MD, president, medicine and science, American Diabetes Association, in a news release stated that, “We now know how to prevent type 2 diabetes and have the data to show that doing so is not only safe, it is cost-effective.” “We should be taking much greater steps on a broad scale to reduce this serious health epidemic in our country…. I would encourage every American to estimate their own risk using simple tools (you can take the test by following this link) and then, if at risk, take simple measures to prevent the disease.”

In the DPP, overweight and obese participants with impaired glucose tolerance were randomly assigned to lifestyle intervention, metformin, or placebo. Results reported in 2002 indicated that metformin therapy resulted in a 31% reduction in the development of diabetes during an average 2.8 years of follow-up.

During the study, participants randomly assigned to receive metformin experienced a decrease in body weight and waist circumference relative to placebo (mean, 2.06% ± 5.65% vs 0.02% ± 5.52%, and 2.13 cm ± 7.06 cm vs 0.79 cm ± 6.54 cm, respectively; P < .001 for both).

Throughout the unblinded follow-up study that covered 10 years from the start, weight loss remained significantly greater in the metformin group relative to placebo (2.0% vs 0.2%; P < .001), and was directly associated with the degree of therapeutic compliance (P < .001).

Participants highly adherent to metformin therapy achieved a weight loss of 3.5% (3.1 kg, 6.8 pounds), whereas those in the low-adherence group experienced an initial weight loss followed by weight changes similar to placebo until 5 years, at which point weight increased.

The authors note that waist circumference increased after 2 years, with the exception of highly adherent participants for whom the increase occurred after the 5-year point. Because body weight did not increase, the authors deduced that central adiposity was increased by a redistribution of body fat.

“Metformin-induced weight loss is almost exclusively confined to reductions in adipose mass with little change in lean tissue,” the authors write, emphasizing that the pattern is different than that observed with caloric restriction, which tends to induce loss of lean, as well as adipose, tissue.

The authors also note that metformin also has several effects on energy metabolism that parallel physical exercise, such as phosphorylation of AMP-activated protein kinase, which is an important regulator of mitochondrial biogenesis, hepatic and muscle fatty acid oxidation, glucose transport, insulin secretion, and lipogenesis.

“Collectively, the clinical data suggesting ‘durable weight loss’ combined with the proposed cellular effects on energy metabolism continue to support metformin as a viable strategy for widespread translational efforts in prevention,” comments William T. Cefalu, MD, from the Pennington Biomedical Research Center at Louisiana State University in Baton Rouge, in an accompanying editorial.

Diabetes Care. 2012;35:663-665, 731-737.