How Does Metformin Affect Cholesterol Levels?

Metformin modestly but consistently lowers total cholesterol, LDL cholesterol, and triglycerides, while leaving HDL cholesterol essentially unchanged. These effects have been confirmed across multiple meta-analyses of randomized trials in people with type 2 diabetes, and they appear in other populations too, including women with polycystic ovary syndrome and people without diabetes at all. The reductions are not as dramatic as what you would see with a statin, but the mechanisms behind them are surprisingly varied and, in some cases, still being untangled.

What the Pooled Trial Data Shows

The clearest picture comes from meta-analyses that combine results across many randomized controlled trials. A meta-analysis published in the Journal of Pharmacy & Bioallied Sciences found that metformin significantly reduced total cholesterol, LDL cholesterol, and triglycerides compared to placebo in patients with type 2 diabetes. The effect on HDL was essentially zero.¹ An earlier systematic review in the Journal of Internal Medicine reached a similar conclusion: metformin lowered total cholesterol, LDL, and triglycerides with statistical significance, while the changes in HDL were small and not meaningful.²

In elderly patients specifically, a meta-analysis in PLOS ONE confirmed that both total cholesterol and LDL dropped significantly with metformin treatment. Interestingly, in that older population, the triglyceride and HDL changes did not reach significance, suggesting the lipid response may shift somewhat with age or the conditions being treated alongside diabetes.³

To put the magnitude in perspective, these are reductions on the order of a few tenths of a mmol/L for total cholesterol and LDL. That is meaningful over time, especially when stacked with other interventions, but it is a fraction of what dedicated cholesterol-lowering drugs achieve. Metformin is not prescribed for its lipid effects; the lipid effects are a useful bonus.

How Metformin Pushes LDL Down

Metformin does not work like a statin. Statins block the liver enzyme responsible for manufacturing cholesterol. Metformin takes a different route, and researchers have identified at least two major pathways that contribute to its LDL-lowering effect.

The first involves an enzyme called AMPK, which acts as a metabolic sensor inside cells. When metformin activates AMPK in liver cells, it dials down the production of proteins that drive fat and cholesterol synthesis. In lab studies using human liver cells, this suppression of lipid-building machinery was shown to depend directly on AMPK activation.⁴

The second pathway involves a protein called PCSK9, which you may have heard of in connection with newer injectable cholesterol drugs. PCSK9 normally tags LDL receptors on liver cells for destruction. Fewer receptors on the surface means less LDL gets pulled out of the bloodstream. Metformin reduces PCSK9 levels in liver cells, which preserves more LDL receptors and enhances LDL uptake from the blood.⁵ This PCSK9-suppressing ability turns out to be especially relevant when metformin is used alongside statins, a point we will return to shortly.

Why HDL Does Not Really Change

Every major pooled analysis finds that metformin has no significant effect on HDL cholesterol levels. This can feel like a missed opportunity, since low HDL is one of the hallmarks of the metabolic profile in type 2 diabetes and PCOS. But the number on your lab report may not tell the whole story.

HDL particles do more than just sit in the blood at a certain concentration. Their main job is to pick up excess cholesterol from artery walls and ferry it back to the liver for disposal, a process called reverse cholesterol transport. In people with diabetes, high blood sugar can chemically modify HDL particles through a process called glycation, making them less effective at this cleanup work. Lab studies have shown that metformin can restore the cholesterol-removal capacity of glycated HDL, bringing it back to near-normal function.⁶ A systematic review corroborated this, finding that metformin attenuated the damage caused by glycation-related compounds on HDL’s ability to pull cholesterol out of cells.⁷

That said, this functional improvement does not always hold up in every context. A small randomized study in youth with type 1 diabetes found that while metformin altered the protein composition of HDL particles, it did not significantly improve cholesterol efflux capacity in that population.⁸ The upshot: metformin may help HDL work better in the specific setting of glycation-heavy environments like poorly controlled type 2 diabetes, but it is not a universal HDL booster.

Triglycerides and the Fat-Clearing Pathway

The triglyceride reduction with metformin is well documented in pooled trial data, though the size of the effect varies across populations. In people with type 2 diabetes, the drop is statistically clear. In older adults, it tends to be smaller and sometimes fails to reach significance.

The mechanism is interesting because, at least in animal models, metformin does not seem to reduce triglyceride production by the liver. Instead, a mouse study found that metformin selectively boosted the clearance of triglyceride-rich particles by brown fat tissue, without changing how much triglyceride the liver was packaging into VLDL particles.⁹ In humans, a detailed metabolic study confirmed that fasting VLDL-triglyceride levels fell with metformin treatment, though the picture was more complex: metformin actually increased the relative contribution of newly made fat to VLDL particles, even as overall circulating VLDL-triglyceride went down.¹⁰ Researchers are still working out how these counterbalancing forces play out over the long term in different populations.

It Changes the Type of LDL Particles Too

Standard cholesterol panels report the total amount of LDL in your blood, but not all LDL particles are the same. Small, dense LDL particles are considered more harmful than large, buoyant ones because they penetrate artery walls more easily and are more prone to oxidation. A trial in non-diabetic patients who had just had a heart attack found that metformin treatment reduced large LDL particles and overall LDL particle size, while slightly lowering total LDL levels. Apolipoprotein B, a marker of total particle count, was unchanged.¹¹

Meanwhile, metformin appears to protect against the damage oxidized LDL does once it enters artery walls. In cell studies, metformin blocked the chain of events that oxidized LDL triggers in immune cells called macrophages, preventing both cell death and the uptake of more fatty material.¹² This matters because macrophages gorging on oxidized LDL is a key step in the formation of arterial plaques.

Effects in PCOS and Non-Diabetic Populations

Metformin’s lipid benefits are not limited to type 2 diabetes. Women with polycystic ovary syndrome often have elevated insulin levels that distort their lipid profile, pushing triglycerides up and HDL down. A study comparing hyperinsulinemic women with PCOS to their less insulin-resistant peers found that six months of metformin dropped total cholesterol by about 11%, LDL by about 12%, and triglycerides by roughly a third. After treatment, the lipid levels of the hyperinsulinemic group were comparable to those of women with normal insulin.¹³

A large systematic review and meta-analysis of randomized trials in women with PCOS confirmed these directions. Metformin led to reductions in total cholesterol, LDL, and triglycerides compared to placebo. The effect on total cholesterol and LDL was more pronounced in women with a BMI of 25 or higher.¹⁴ The researchers rated the certainty of evidence as low to moderate, so the effect sizes should be taken as approximate rather than nailed-down, but the direction is consistent.

In people with type 1 diabetes, the three-year REMOVAL trial found that metformin reduced LDL cholesterol and body weight, though it did not meet its primary endpoint of slowing artery wall thickening.¹⁵ Even in a population where the core metabolic problem is completely different from type 2 diabetes, the LDL-lowering signal persisted.

What Happens When You Combine Metformin with a Statin

Since metformin and statins lower LDL through different mechanisms, combining them can produce additive benefits. A double-blinded randomized trial (the ATOMIC trial) in patients with both type 2 diabetes and dyslipidemia concluded that the combination of metformin and atorvastatin was effective at reducing both LDL and blood glucose, suggesting a complementary role for the two drugs in managing cardiovascular risk.¹⁶

There is a subtler interaction at play here. Statins, while highly effective at lowering LDL, also trigger the liver to produce more PCSK9. That extra PCSK9 partially counteracts the statin’s benefit by breaking down LDL receptors. Metformin suppresses PCSK9, which means it can blunt this rebound effect. A study in patients with coronary artery disease who did not have diabetes found that adding metformin to statin therapy reduced LDL by about 21% within a month, with roughly three-quarters of patients reaching guideline-recommended LDL goals. PCSK9 levels dropped by about 15%, and the number of small, dense LDL particles fell significantly compared to statin therapy alone.¹⁷ That is a provocative result because it opens the door to using metformin as a lipid-management tool even in people without diabetes, though this is still off-label and not yet reflected in treatment guidelines.

Plaque Stability Beyond Cholesterol Numbers

Cholesterol levels matter because of what high cholesterol does to arteries over time: it drives plaque buildup. But not all plaques are equally dangerous. Unstable plaques with thin caps and high inflammatory activity are the ones that rupture and cause heart attacks. Metformin appears to improve plaque stability through a mechanism that has nothing to do with lipid levels. Researchers found that metformin directly binds to an enzyme called MMP-9, which normally degrades the collagen that holds plaques together. By binding to MMP-9 and promoting its breakdown, metformin helped preserve the collagen content in plaques and reduced their vulnerability index in an animal model of carotid artery disease.¹⁸

This finding, still in the animal-model stage, helps explain why metformin’s cardiovascular benefits in clinical trials have sometimes seemed larger than what you would predict from its modest lipid improvements alone. The landmark UKPDS trial, which followed patients with newly diagnosed type 2 diabetes over roughly two decades, found that metformin significantly reduced the risk of heart attacks, diabetes-related death, and all-cause death. Those benefits persisted even ten years after the randomized treatment period ended.¹⁹ A small LDL reduction alone probably cannot account for that magnitude of cardiovascular protection. The plaque-stabilizing, anti-inflammatory, and anti-oxidative properties of metformin likely contribute as well.

Liver Fat and the Lipid Connection

Fatty liver disease, now commonly called metabolic dysfunction-associated steatotic liver disease (MASLD), is tightly linked to the same insulin resistance that metformin treats. When excess fat accumulates in liver cells, it alters how the liver produces and exports lipoproteins, often worsening cholesterol and triglyceride levels. In theory, if metformin reduces liver fat, you would expect downstream lipid improvements.

Animal studies have been encouraging. In mouse models of fatty liver, metformin has been shown to reduce liver fat by ramping up the cellular cleanup process known as autophagy and by inhibiting lipid production.²⁰ The human data, however, is less clear-cut. A review in Gut and Liver noted that while preclinical evidence supports metformin’s role in reducing liver fat, clinical studies in adults with fatty liver have produced mixed results for both liver enzymes and actual fat content.²¹ A separate commentary put it bluntly: metformin looks like a powerful weapon against liver fat in rodents, but the evidence in humans is less convincing.²² Current guidelines do not recommend metformin specifically for fatty liver disease, despite the theoretical appeal.

Why the Same Dose Does Not Work the Same for Everyone

If you have been on metformin and noticed your cholesterol numbers barely moved while a friend on the same dose saw clear improvement, genetics could be a factor. Metformin enters liver cells primarily through a transporter protein called OCT-1. Genetic variations in the gene encoding OCT-1 can alter how efficiently metformin gets into the cells where it needs to work. A study that compared metformin responders to non-responders found specific gene variants in OCT-1 that were present only in the non-responder group. Lab experiments confirmed that these variants changed OCT-1’s structure and reduced metformin transport into cells, weakening activation of the downstream metabolic pathway.²³

This is not something most patients are tested for, and routine genetic screening before starting metformin is not standard practice. But it does mean that a lackluster cholesterol response to metformin is not necessarily a sign that something else is wrong, or that you are not taking it correctly. Some people are simply wired to get less lipid benefit from the drug. For those individuals, the glucose-lowering effects may also be blunted, which is something worth discussing with a doctor if metformin seems to be underperforming on multiple fronts.

Aside from genetics, other factors shape the lipid response. Baseline insulin resistance matters: the more insulin-resistant you are, the more room there is for metformin to correct the metabolic dysfunction that drives lipid abnormalities. Diet, weight loss (which metformin can modestly promote), and the specific combination of other medications all contribute. In the PCOS literature, the biggest lipid improvements showed up in women who were both overweight and hyperinsulinemic, reinforcing the idea that metformin’s lipid effects are most visible when insulin resistance is driving the problem.¹³