Metformin Cardiovascular Impact Long-Term: What the Evidence Actually Shows

What UKPDS 34 actually randomized

Of 4,075 patients recruited to the UK Prospective Diabetes Study, 1,704 overweight patients with raised fasting glucose after three months of diet alone were eligible, and 753 entered the randomized comparison: 342 assigned metformin and 411 assigned a conventional policy, primarily diet alone, followed for a median of 10.7 years [1].

Against conventional therapy, metformin reduced any diabetes-related endpoint by 32% (95% CI 13 to 47), diabetes-related death by 42% (9 to 63) and all-cause mortality by 36% (9 to 55). Median HbA1c was 7.4% on metformin and 8.0% on conventional therapy [1].

The trial also set metformin against the other intensive arms. Metformin showed a greater effect than chlorpropamide, glibenclamide or insulin for any diabetes-related endpoint (p=0.0034), all-cause mortality (p=0.021) and stroke (p=0.032) [1].

Did the benefit survive after the trial stopped

Ten-year post-trial monitoring (UKPDS 80) followed the original participants without trying to maintain their assigned treatments. Between-group HbA1c differences were lost after the first year [2]. In the metformin group, risk reductions persisted for any diabetes-related endpoint (21%, P=0.01), myocardial infarction (33%, P=0.005) and death from any cause (27%, P=0.002) [2].

For context, the sulfonylurea-or-insulin intensive group showed a 15% reduction in myocardial infarction (P=0.01) and a 13% reduction in death from any cause (P=0.007) at the same follow-up [2]. The metformin findings apply to the overweight group; the sulfonylurea-or-insulin group was analyzed separately [2].

What mechanisms have actually been measured

The metformin hydrochloride tablet label describes the drug in glycemic terms: metformin decreases hepatic glucose production, decreases intestinal glucose absorption, and improves insulin sensitivity by increasing peripheral glucose uptake and utilization, while insulin secretion stays unchanged [3].

Inflammation markers were studied in a large diabetes cohort and in a trial involving people with heart failure but without diabetes [6]. A separate 26-person trial measured oxidative-stress and platelet-activation markers [7].

In the inflammation work, a cohort of treatment-naive patients with diabetes started on metformin had a lower neutrophil-to-lymphocyte ratio after 8 to 16 months than those started on a sulfonylurea, a mean difference of 0.09 log units (95% CI 0.02 to 0.17; P=0.013), and in a double-blind, placebo-controlled trial in people with heart failure but without diabetes, metformin suppressed plasma cytokines including CCL11 [6]. In laboratory experiments, AMPK was not required for one inflammatory pathway measured by the researchers [6].

Platelet and oxidative-stress markers come from a 12-week trial that randomized 26 newly diagnosed patients to metformin (n=13) or gliclazide (n=13). Urinary 8-iso-PGF2alpha fell from 708 to 589 pg/mg creatinine on metformin (p<0.001) and was unchanged on gliclazide; urinary 11-dehydro-thromboxane B2, a thromboxane-dependent marker of platelet activation, fell from 2,190 to 1,753 pg/mg creatinine (p<0.05) and did not change on gliclazide, despite comparable glycemic improvement in both groups [7].

These are surrogate markers, not cardiovascular events, and they do not show that any of these pathways produces the event reduction seen in UKPDS.

What happens in people who already have coronary disease

One randomized cardiovascular comparison in this group is SPREAD-DIMCAD. It enrolled 304 adults with type 2 diabetes and a history of coronary artery disease, mean age 63.3 years, and assigned them to glipizide 30 mg daily or metformin 1.5 g daily for 3 years, with a median follow-up of 5.0 years [4].

The primary composite was cardiovascular death, death from any cause, nonfatal myocardial infarction, nonfatal stroke or arterial revascularization. By the end of follow-up 91 participants had had 103 primary endpoints, and the adjusted hazard ratio for metformin versus glipizide was 0.54 (95% CI 0.30 to 0.90; P=0.026). Secondary endpoints and adverse events did not differ significantly between the groups [4].

Both arms reached similar glycemic control, with HbA1c of 7.0% on metformin and 7.1% on glipizide. The comparator was glipizide rather than placebo [4].

Heart failure: what the label says and what the cohorts show

The immediate-release label does not list stable heart failure among its contraindications. Those are severe renal impairment with eGFR below 30 mL/min/1.73 m2, hypersensitivity to metformin, and acute or chronic metabolic acidosis including diabetic ketoacidosis [3]. Heart failure appears instead as a risk factor: acute congestive heart failure, particularly with hypoperfusion and hypoxemia, is one of the hypoxic states in which lactic acidosis has been reported, and the label says to discontinue metformin when such an event occurs [3].

The comparative data in heart failure are observational. A systematic review of nine cohort studies involving about 34,000 patients with diabetes and heart failure found metformin associated with lower mortality than other glucose-lowering drugs, mostly sulfonylureas: 23% versus 37%, pooled adjusted risk estimate 0.80 (95% CI 0.74 to 0.87) [5]. The same review found no increased mortality risk in patients with reduced left ventricular ejection fraction (pooled adjusted risk estimate 0.91, 95% CI 0.72 to 1.14), a small reduction in all-cause hospitalization (0.93, 95% CI 0.89 to 0.98) and no increased risk of lactic acidosis. It identified no randomized trials in this population [5].

GLP-1 receptor agonist outcome trials, for context

These trials did not compare GLP-1 drugs with metformin; each compared a GLP-1 drug with placebo.

LEADER randomized 9,340 patients with type 2 diabetes and high cardiovascular risk to liraglutide or placebo and followed them for a median of 3.8 years. The composite of cardiovascular death, nonfatal myocardial infarction or nonfatal stroke occurred in 13.0% versus 14.9% (hazard ratio 0.87, 95% CI 0.78 to 0.97) [8].

SUSTAIN-6 randomized 3,297 patients to once-weekly semaglutide 0.5 mg or 1.0 mg, or placebo, for 104 weeks. The same composite occurred in 6.6% versus 8.9% (hazard ratio 0.74, 95% CI 0.58 to 0.95) [9].

Both trials enrolled many more patients than UKPDS 34 or SPREAD-DIMCAD. A 2020 review in the Journal of the American College of Cardiology says SGLT2 inhibitors and GLP-1 receptor agonists have consistently shown safety and reduction in cardiovascular events in patients with established cardiovascular disease and are becoming essential tools for cardioprotection in that group, while describing the effects of biguanides and several older classes on cardiovascular events as variable [10].

Does metformin prevent events before diabetes develops

The Diabetes Prevention Program randomized 3,234 adults with elevated fasting and post-load glucose to placebo, metformin 850 mg twice daily, or an intensive lifestyle program. Over an average of 2.8 years, diabetes incidence was 11.0, 7.8 and 4.8 cases per 100 person-years respectively; metformin reduced incidence by 31% (95% CI 17 to 43) and lifestyle by 58% (48 to 66), and lifestyle was significantly more effective than metformin [11].

The follow-up study (DPPOS) tracked 2,776 of the surviving cohort for a mean of 15 years. Diabetes incidence was reduced by 18% with metformin (HR 0.82, 95% CI 0.72 to 0.93) and 27% with lifestyle (HR 0.73, 95% CI 0.65 to 0.83), with cumulative incidence at 15 years of 56% for metformin, 55% for lifestyle and 62% for placebo. There was no overall difference in the aggregate microvascular outcome: 13.0% with metformin, 11.3% with lifestyle and 12.4% with placebo [12]. The primary outcomes of this 15-year report were diabetes development and microvascular disease, not cardiovascular events [12].

What the label says to watch over years of use

A boxed warning covers metformin-associated lactic acidosis. Reported cases have resulted in death, hypothermia, hypotension and resistant bradyarrhythmias; onset is often subtle, with malaise, myalgias, respiratory distress, somnolence or abdominal pain. Laboratory findings include blood lactate above 5 mmol/L, anion gap acidosis, an increased lactate-to-pyruvate ratio, and metformin plasma levels generally above 5 mcg/mL [3].

Named risk factors are renal impairment, interacting drugs such as carbonic anhydrase inhibitors, age 65 or older, radiological studies with contrast, surgery and other procedures, hypoxic states such as acute congestive heart failure, excessive alcohol intake, hepatic impairment and mitochondrial disease. If lactic acidosis is suspected, the label says metformin should be discontinued, general supportive measures instituted in a hospital setting, and prompt hemodialysis is recommended [3].

How often this happens is a separate question. A Cochrane review pooled 347 comparative trials and cohort studies and found no cases of fatal or nonfatal lactic acidosis in 70,490 patient-years of metformin use or in 55,451 patient-years of non-metformin treatment; the calculated upper limit for the true incidence was 4.3 cases per 100,000 patient-years with metformin and 5.4 without it [14].

The label says the postmarketing lactic acidosis cases primarily occurred in patients with significant renal impairment. It says to obtain an eGFR before starting and at least annually, contraindicates metformin below an eGFR of 30 mL/min/1.73 m2, and does not recommend starting it at 30 to 45 mL/min/1.73 m2 [3]. For iodinated contrast, the label says to stop metformin at the time of or before the procedure in patients with eGFR 30 to 60, a history of hepatic impairment, alcoholism or heart failure, or when intra-arterial contrast is used; eGFR is re-evaluated 48 hours later and metformin restarted if renal function is stable [3].

Vitamin B12 is the other long-term issue. In 29-week trials, subnormal serum B12 levels developed in approximately 7% of patients on metformin hydrochloride tablets, and the label says to measure hematologic parameters annually and vitamin B12 at 2 to 3 year intervals [3]. In DPPOS, low B12 of 203 pg/mL or less occurred in 4.3% of metformin users versus 2.3% of placebo at 5 years (P=0.02) and 7.4% versus 5.4% at 13 years, which was not statistically significant; combined low and borderline-low B12 was more common with metformin at both points, and the odds of deficiency rose 13% per year of metformin use (OR 1.13, 95% CI 1.06 to 1.20) [15].

Label-listed interactions: carbonic anhydrase inhibitors such as topiramate, zonisamide, acetazolamide and dichlorphenamide can lower serum bicarbonate and add to lactic acidosis risk; ranolazine, vandetanib, dolutegravir and cimetidine interfere with the renal transporters that clear metformin, raising exposure; alcohol potentiates metformin's effect on lactate metabolism; and insulin or an insulin secretagogue raises hypoglycemia risk [3]. Thiazides, corticosteroids and other drugs listed in the label can raise glucose and cause loss of glycemic control [3].

The label lists immediate-release starting doses of 500 mg twice daily or 850 mg once daily with meals, with a maximum of 2,550 mg per day; extended-release tablets have separate dosing instructions [3][16]. The extended-release label says the tablets should be swallowed whole and never crushed, cut or chewed [16].

What is still unresolved

A 2016 review described the TAME trial as a planned test of metformin against aging, noting that metformin has demonstrated protective effects against several age-related diseases in humans [17].

For cardiovascular events, the randomized evidence cited here is the 753-patient UKPDS 34 comparison with its 10-year extension and the 304-patient comparison against glipizide, supplemented by observational cohort data. That is a much smaller randomized base than the GLP-1 outcome trials above.