Type 2 diabetes mellitus (T2DM) is considered as an independent risk factor for the development of CAD [
10]. Therefore, tight blood glucose control is critical to limit the mortality and morbidity from CVD in T2DM patients. Metformin, a first line anti-diabetic drug, has been reported to reduce major cardiovascular events associated with atherosclerotic cardiovascular disease (ASCVD) in T2DM patients or improve the surrogate endpoints of ASCVD such as carotid intima-media thickness (CIMT). A landmark study in this area was the UK Protective Diabetes Study (UKPDS) [
11], which randomized 1704 overweight (> 120% ideal body weight) patients with newly diagnosed T2DM to receive conventional treatment with diet alone in one trial arm or either metformin, sulphonylurea, or insulin in the other. After a median follow up of 10.7 years, the metformin group had a 36% lower all-cause mortality (
P = 0.011) and a concurrent a 39% lower risk (
P = 0.011) in the incidence of myocardial infarction than the conventional treatment group but did not differ significantly from the other intensive glucose control treatment group. However, in a combined analysis of a supplementary of the same clinical trial, where 537 non-overweight and overweight patients with uncontrolled plasma blood glucose (6.1–15.0 mmol/L) were treated with sulfonylureas with and without metformin, the effect of metformin on cardiovascular outcomes was not statistically significant. A possible explanation of this phenomenon could be related to the beneficial effect of tight glycemic control from metformin that prevented future cardiovascular consequences. Subsequent follow-up study of 10 years after UKPDS, however, reported continuous risk reduction of myocardial infarction (33%,
P = 0.005) in patients treated with metformin despite no changes in the glycated hemoglobin (HbA1c) levels [
12]. Since no new glucose-lowering therapy was introduced in the study cohort during this period, the results highlight the beneficial cardiovascular effects of metformin. This effect of metformin was particularly pronounced among overweight patients after a long duration of follow-up. In a continued effort to study the add-on effects of metformin on macrovascular or microvascular disease in insulin-treated T2DM patients, Kooy et al. [
13] randomized 390 T2DM patients, with a mean age of 53 years, to receive metformin (850 mg/day) or placebo and followed them for 4.3 years. The results show that metformin treatment significantly improved the macrovascular end point compared with placebo (HR 0.61, 95% CI 0.40–0.94,
P = 0.02), which could not be explained solely by the difference in weight and metformin-associated changes in metabolic or hemodynamic variables, such as HbA1C level [
13]. Further study by Katakami et al. [
14] in a cohort of 118 patients with T2DM who were randomized to receive glibenclamide (n = 59), gliclazide (n = 30), glibenclamide plus metformin (n = 29), with a median follow-up duration of 3 years, also showed that the CIMT in glibenclamide plus metformin group was significantly smaller than that in the glibenclamide and gliclazide groups in univariate and multivariate regression analysis (P < 0.05). A meta-analysis of 35 clinical RCTs confirmed the cardiovascular benefits of metformin in comparison to placebo in younger population followed over a long duration [
15]. Taken together, these observations strengthen the rationale behind the use of metformin in ‘at-risk’ population (T2DM patients) from a younger age to decrease cardiovascular events.
A seminal study in the field, The study on the prognosis and effect of antidiabetic drugs on type 2 diabetes mellitus with coronary artery disease (SPREAD-DIMCAD) was conducted to evaluate the major cardiovascular events and mortality among type 2 diabetic patients with CAD after their treatment with glipizide or metformin [
16]. Among the 304 T2DM patients enrolled for the RCT who were followed up for 5 years, the metformin group showed a significantly lower cardiovascular endpoint (recurrent cardiovascular events, including nonfatal myocardial infarction, nonfatal stroke or arterial revascularization by percutaneous transluminal coronary angioplasty (PTCA) or by coronary artery bypass graft, death from a cardiovascular cause) than the glipizide group (HR 0.54, 95% CI 0.30–0.90,
P = 0.026). However, the glycated hemoglobin values in the two groups were similar (7.0% vs 7.1%, P > 0.05) [
16]. The results were ‘proof of concept’, suggesting the pleiotropic effects of metformin in the heart and blood vessels, independent of its glucose-lowering activity.
The REversing with MetfOrmin Vascular Adverse Lesions (REMOVAL) Trial is the largest and longest double-blind placebo-controlled RCT to evaluate cardiovascular effect of metformin in adults with type 1 DM (T1DM) with a median follow up duration of 5 years in patients with high cardiovascular risk (have ≥ 3 of 10 specified cardiovascular risk factors) [
17]. In REMOVAL trial, 428 insulin-treated patients were randomly assigned to metformin and placebo. At a follow-up visit, atherosclerosis progression, measured by the averaged maximal CIMT, was significantly reduced with metformin (− 0.013 mm/year, − 0.024 to − 0.003; P = 0.0093) [
18], indicating possible cardiovascular benefits, but warrants further investigation. However, the significant reduction in HbA1c (− 0.13%, 95% CI − 0.22 to − 0.037; P = 0.0060) over 3 years was more robust over the initial 3-month after the commencement of treatment (− 0.24%, − 0.34 to − 0.13; P < 0.0001) [
18]. In the SPREAD-DIMCAD trial, the metformin group showed a significantly lower cardiovascular endpoint with similar HbA1c level, suggesting the metformin effect is independent of blood sugar level. However, in REMOVAL trial, the reduction of CIMT was associated with HbA1c reduction with metformin. Taken together, the cardioprotective properties of metformin appears to be a consequence of its atheroprotective effect of metformin. However, the conundrum, concerning the association of the atheroprotective effect of metformin with its anti-hyperglycemic effects in T1DM patients, still remains unanswered.
Recent evidence demonstrates the synergistic effects of co-administration of metformin with other drugs. Treatment with metformin alongside empagliflozin, a new sodium glucose cotransporter-2 (SGLT2) inhibitor, has significantly improved arterial stiffness compared to metformin alone in T1DM patients [
19]. The effect was higher than either combination of glitazones or alpha-glucosidase inhibitors with metformin and was also associated with lower major adverse cardiovascular events (MACE) risk in comparison to sulphonylureas and metformin combinatorial treatment in T2DM patients [
20]. Metformin with Saxagliptin has been shown to improve the endothelial dysfunction in early diabetics [
21] and its combination with vildagliptin is poised as a viable alternative in the treatment of T2DM and CAD due to the lower rate of recurrent cardiovascular events, in part due to its anti-inflammatory property [
22]. Even, metformin with ascorbic acid has been shown to be effective in reducing risks for diabetes-related long-term complications (including albumin/creatinine ratio) [
23]. However, increased BMI in metformin-exposed children during intrauterine development might confer them a higher risk of developing cardiometabolic diseases later in their adulthood [
24]. Recent evidence supports the use of liraglutide as a viable alternative to metformin for recent-onset T2DM in women during their child-bearing age to circumvent such clinical scenarios [
25].
Although the cardiovascular benefits with metformin are well-established in diabetic patients, their role in non-diabetic patients remains elusive. In a small randomized double-blind placebo-controlled study consisting of 33 non-diabetic women, it was shown that metformin can reduce myocardial ischemia in female patients with angina, compared to placebo [
26]. However, a study by Hao et al. [
9] consisting of 130 patients with dyslipidemia and obesity who were randomized to atorvastatin or atorvastatin plus metformin, showed atorvastatin combined with metformin was more effective than atorvastatin monotherapy group in improving the rate of obesity and subclinical inflammation. A subsequent study by Eduardo et al. [
27] confirmed that metformin decreased the CIMT (− 0.1 mm, P = 0.04, vs − 0.02 mm, P = not significant) in comparison to control group in patients with metabolic syndrome, thereby indicating its role in cardioprotection.
However, recent studies have questioned the validity of the conclusion from prior studies. The Carotid Atherosclerosis: MEtformin for insulin ResistAnce (CAMERA) [
28] study involving 173 non-diabetic patients with CAD who were on statin therapy, were assigned to either metformin or matching placebo. No improvement in CIMT was reported in the metformin group (slope difference 0.007 mm/year, P = 0.29) when compared with a placebo group, although HbA1c, insulin, and insulin resistance index decreased significantly in the metformin group (P < 0.05) [
28]. The possible explanations for the conflicting outcomes in these studies can be attributed to the difference in baseline characteristics of the patients (a type of disease, age, taking other hypoglycemic and lipid-lowering drugs), study endpoints and follow-up duration of the individual clinical study. Whether metformin has a cardiovascular benefit in pre-diabetic patients is not clear. An area of active research, ongoing multicenter RCT Glucose Lowering in Non-diabetic Hyperglycemia Trial (GLINT, ISRCTN34875079) is currently enrolling non-diabetic patients for treatment with metformin to evaluate the incidence of cardiovascular death and non-fatal myocardial infarction events. The findings from this study will provide more insight in the prophylactic use of metformin in a similar cohorts.