3.3.2 Effect of metformin on CVD
Mounting data from observational studies have indicated that metformin exerts more CVD benefits than other hypoglycemic drugs do in T2DM patients 91, 100, 101. In a retrospective study, Roumie et al. compared the CVD events and mortality risk of diabetic veterans who initiated monotherapy with four different medications (metformin, sulfonylureas, rosiglitazone, and glibenclamide). They found that compared with metformin, the other three glucose-lowering drugs were all associated with significantly increased CVD events incidence or mortality 102. A large-scale prospective study also confirmed the protective CVD effect of metformin over diet. In the United Kingdom Prospective Diabetes Study (UKPDS), Holman and his colleagues compared the cardiovascular events between diabetic patients who underwent dietary therapy and obese diabetic patients who received metformin. Their results showed that metformin was associated with a significantly lower incidence of myocardial infarction (33%,P =0.005) compared with the diet-alone group103. Consistent with the UKPDS study, a small clinical trial by Hong et al. compared the effect of two glucose-lowering drugs, glipizide (30 mg daily) and metformin (1.5 g daily), in a 3-year treatment of 304 diabetes patients who had a history of coronary artery disease. The metformin group was found to have a significant reduction of major CVD events (HR 0.54 ,95% CI 0.30–0.90; P = 0.026) compared with glipizide group, after a median of 5.0 years of follow-up period 104.
IIn addition to diabetic patients, metformin has also been shown to prevent the development of CVD in pre-diabetic individuals. A clinical trial compared the coronary calcium score, a proxy for CVD, between the pre-diabetic patients who used metformin versus those who used placebo. Compared with the placebo group, the metformin group showed a significantly lower coronary calcium score in the male group but not the female group. This could be due to the interactions occurring between testosterone and metformin in male subjects 105.
Moreover, metformin treatment has been shown to reduce the risk of CVD in non-diabetic individuals. In a study, three hundred and eighty non-diabetics patients with ST-segment elevation myocardial infarction (STEMI) were assigned to receive metformin or placebo for 4 months. Lexis et al. found that CVD proxies, which included glycated hemoglobin, total cholesterol, low-density lipoprotein cholesterol, and body weight, were reduced in the metformin group relative to the placebo group (Lexis et al., 2015). Furthermore, multiple meta-analyses have also shown metformin use in non-diabetic subjects to beassociated with reduced systolic blood pressure, which is the one of most important risk factors for CVD. This effect was particularly prominent in those with impaired glucose tolerance or were obese 105.
Studies that investigate the CVD risk after treatment with metformin or other anti-diabetic drugs are also affected by bias in the experimental design. For example, both sulfonylureas and rosiglitazone are associated with CVD events 106, 107. Thus, using these hypoglycemic drugs as a comparison could not determine whether the reduced CVD events were due to the protective effect of metformin or the increased CVD risk associated with sulfonylurea and rosiglitazone.
In summary, compelling evidence from observational studies has suggested that the role of metformin in preventing CVD in people with and without diabetes. However, further elucidation of the role of metformin in retarding CVD is critically needed from well-designed large-scale clinical trials.