Can Metformin Cause Iron Deficiency?

Metformin does lower several markers of iron status, including serum ferritin and transferrin saturation, and some large clinical trials have found a modestly higher rate of anemia among metformin users. But the story is more complicated than “metformin drains your iron.” People with type 2 diabetes often start with abnormally high iron stores, and much of the drug’s iron-lowering effect may be corrective rather than harmful. Whether that shift tips into true deficiency depends on dose, duration, diet, and the individual’s starting iron status.

What the Blood Work Actually Shows

The clearest snapshot comes from a 2025 study comparing people with type 2 diabetes who were taking metformin against those managed without it. Metformin users had lower serum iron (about 16 versus 17 µmol/L), substantially lower ferritin (roughly 66 versus 108 µg/L), and lower transferrin saturation (about 27% versus 31%). At the same time, their transferrin and unsaturated iron-binding capacity were higher, meaning the blood’s iron-carrying proteins had more room to spare.¹ In plain terms, metformin users had less iron floating around in their blood and less stored in their tissues.

A ferritin drop from over 100 down to 66 sounds dramatic, but both values are within normal range for most adults. The people on metformin were not, on average, iron-deficient. They simply looked less iron-loaded. That distinction matters because ferritin is not just a storage number; elevated ferritin in diabetes is itself tied to metabolic problems like insulin resistance and liver fat accumulation. Pulling it down can be a benefit rather than a side effect.

Why Diabetes Patients Often Start With Too Much Iron

Before worrying about metformin lowering iron, it helps to understand the baseline. People with diabetes tend to carry more iron than the general population. One study found that the prevalence of elevated transferrin saturation (above 35%) was three to four times higher in patients with diabetes compared with historical rates in the general population. Higher fasting glucose itself was independently associated with elevated transferrin saturation.²

Separate laboratory work has shown that sustained high blood sugar drives iron accumulation in the liver. Liver cells exposed to chronic high glucose ramp up their transferrin receptor expression and increase their iron-storage protein (ferritin heavy chain), essentially pulling in and hoarding more iron.³ So when metformin enters the picture and iron markers drop, the drug is often nudging an already-high iron load back toward normal, not depleting an adequate supply.

How Metformin Moves Iron Around

Researchers have pinpointed at least two molecular mechanisms by which metformin reshuffles iron handling. The first involves hepcidin, a hormone made in the liver that acts as a master switch for iron absorption. Hepcidin tells the gut to absorb less iron and tells cells to lock iron inside rather than releasing it. Metformin suppresses hepcidin production in liver cells in a dose-dependent way.⁴ Paradoxically, lower hepcidin should mean more iron absorption from the gut, not less. That is why the picture is so confusing: the hepcidin effect alone would predict higher iron levels, not lower ones.

The second mechanism works in the opposite direction. Metformin activates an energy-sensing enzyme called AMPK, which in turn boosts ferroportin, the protein that escorts iron out of cells and into the bloodstream for redistribution. In animal models of fatty liver disease, metformin treatment significantly reduced the amount of iron trapped inside liver tissue and protected cells from iron-driven oxidative damage.⁵ Cell culture experiments confirmed this: metformin decreased ferritin heavy chain levels and partially reversed the iron-hoarding phenotype triggered by high glucose.⁶

The net result is not a simple drain on iron supply but a redistribution. Iron that was stuck in liver cells gets liberated, hepcidin suppression loosens the grip on gut absorption, and stored ferritin drops. For someone with diabetic iron overload, that rebalancing can prevent a type of cell death called ferroptosis, where excess iron triggers oxidative damage that kills liver cells.⁷ For someone whose iron was already borderline low, though, the same redistribution could push them into deficiency territory.

Does This Actually Cause Anemia?

Lowering iron markers is one thing. Causing full-blown iron deficiency anemia, where hemoglobin drops enough to produce fatigue, pallor, and shortness of breath, is another. The evidence here pulls in opposite directions.

On the side of concern, two major diabetes trials found that anemia risk was higher in metformin users than in people on other glucose-lowering drugs, with cases appearing as early as six months after starting treatment.⁸ A meta-analysis that pooled results from multiple studies confirmed that metformin-treated patients who developed iron deficiency anemia had significantly lower hemoglobin, averaging about 11.6 g/dL compared with 13.8 g/dL in controls without anemia. Those same patients also showed signs of elevated oxidative stress.⁹

On the reassuring side, a large study of urban Chinese patients with type 2 diabetes found the opposite pattern: the adjusted incidence of iron deficiency anemia was actually lower in the metformin group (about 26 per 1,000 person-years) compared with the non-metformin group (about 43 per 1,000 person-years). Even more striking, patients who took metformin covering 20% to 79% of their treatment days had the lowest anemia risk of all.¹⁰

How can these findings coexist? Part of the answer is likely population differences. The Chinese cohort may have had higher baseline iron stores, different dietary iron intake, or different rates of conditions like chronic kidney disease that independently affect anemia. Also, the comparison group matters: comparing metformin to no treatment is different from comparing metformin to a specific alternative drug. And metformin’s well-known effect on vitamin B12 absorption adds another layer, because B12 deficiency causes its own type of anemia that can be confused with or overlap iron deficiency anemia.

The B12 Wrinkle

Metformin impairs the absorption of vitamin B12 in the gut, a well-documented effect that shows up after roughly a year of continuous use. B12 deficiency causes a distinct form of anemia characterized by abnormally large red blood cells, which looks different from the small-cell anemia typical of iron deficiency. However, when both deficiencies coexist, the blood work can become misleading: the two effects can mask each other, making red blood cells appear normal-sized even though neither nutrient is adequate.⁸

This overlap is one reason why studies sometimes disagree about whether metformin causes iron deficiency specifically. If researchers measure only hemoglobin and mean cell volume without checking ferritin, serum iron, and B12 separately, the type of anemia gets misclassified. You could have a B12-driven anemia blamed on iron, or an iron-driven one attributed to B12, depending on which test was ordered.

Who Is Most Vulnerable

Not everyone on metformin faces the same risk. Several groups sit closer to the tipping point where the drug’s iron-lowering properties could become clinically significant:

Premenopausal women: Monthly menstrual blood loss already keeps iron stores lower than in men or postmenopausal women. Adding metformin’s ferritin-lowering effect on top of regular periods could push some women into deficiency, especially with heavy periods.
People on restricted diets: Vegetarian and vegan diets supply non-heme iron, which the body absorbs less efficiently. If metformin is simultaneously shifting iron handling, a marginally adequate dietary intake may no longer keep up.
Those with chronic kidney disease: Kidney problems independently impair red blood cell production and iron recycling. Diabetes itself is the leading cause of chronic kidney disease in many countries, so the overlap is common.
Long-term, high-dose users: The dose-response relationship is not straightforward. The Chinese study noted that people with moderate metformin coverage had lower anemia risk than both the lowest and highest coverage groups, suggesting a sweet spot rather than a simple “more drug, more risk” pattern.¹⁰

People who fall into none of these categories and who start metformin with healthy or elevated iron stores are unlikely to develop iron deficiency from the drug alone. Their ferritin will probably drop, their transferrin saturation will dip, and that shift will likely do them more good than harm by reducing oxidative stress in the liver.

Metformin, PCOS, and Ferritin

Metformin is widely prescribed for polycystic ovary syndrome, not just diabetes. Women with PCOS often have elevated ferritin, partly because of the insulin resistance that characterizes the condition. A study of women with PCOS found that metformin treatment significantly reduced serum ferritin, though the reduction was significant only in lean and overweight women, not in obese participants.¹¹

For many women with PCOS, this ferritin reduction is actually a therapeutic goal. Elevated iron stores in PCOS have been linked to worsened insulin resistance and inflammation. Bringing ferritin down into a healthy range may improve metabolic markers, not just iron status. But the same caveat applies here as in diabetes: a woman with PCOS who also has heavy periods and a plant-based diet may not have much ferritin to spare. Regular monitoring makes sense in that scenario.

What About Pregnancy

Metformin is sometimes used in gestational diabetes, and iron demands soar during pregnancy, raising an obvious question about whether the drug could worsen anemia in expectant mothers. The most useful data on this comes from a follow-up study of children born to mothers who took metformin during pregnancy. At ages seven to nine, the children of metformin-treated mothers actually had higher ferritin levels than children whose mothers were treated with insulin alone. No child in either group was iron deficient.¹²

This does not directly answer whether the mothers themselves experienced more iron deficiency, and it is only one study. Still, the finding that offspring iron status was unharmed, and if anything slightly better, is reassuring. Pregnant women on metformin are already monitored closely for anemia as part of routine prenatal care, so any developing deficiency would typically be caught early.

Practical Steps if You Take Metformin

Routine blood work for diabetes management usually includes a complete blood count, but that alone will not reliably detect early iron depletion. A complete blood count shows hemoglobin, which only drops once iron stores are substantially depleted. Asking for a ferritin level and transferrin saturation gives a much earlier signal. If your ferritin is comfortably in the normal range (say, above 50 µg/L for most adults), a modest decline on metformin is not a concern. If it is already on the low end, it is worth tracking more closely.

Timing matters for iron absorption, too. Metformin taken with meals can compete with iron supplements for gut absorption. If you need supplemental iron, spacing it at least two hours away from your metformin dose helps both medications work as intended. Taking iron with a source of vitamin C (a glass of orange juice, bell peppers, strawberries) enhances absorption, while taking it with dairy, coffee, or tea reduces it.

Because metformin also impairs B12 absorption, many clinicians now recommend periodic B12 screening alongside iron markers. Catching both deficiencies early is straightforward and inexpensive, and treatment for each is simple: oral iron for iron deficiency, intramuscular or high-dose oral B12 for B12 deficiency. Neither requires stopping metformin.

The Hepcidin Puzzle Is Not Fully Solved

One of the unresolved pieces of this story is the conflicting evidence about how metformin suppresses hepcidin. Two studies in liver cells agree that metformin powerfully inhibits hepcidin production stimulated by a signaling molecule called BMP6, but they disagree on the mechanism. One group found that the effect depends on AMPK activation through a regulatory protein called SHP.⁴ Another group, using liver cells from mice genetically lacking AMPK, found that metformin suppressed hepcidin even more powerfully in the absence of AMPK, meaning the effect was clearly not going through that pathway at all.¹³

This disagreement matters because understanding the mechanism would help predict who is most affected and whether the hepcidin suppression can be managed without stopping the drug. If AMPK is the pathway, then the iron effect is tightly linked to the same metabolic signaling that makes metformin useful for blood sugar. If AMPK is not involved, the iron effect is an off-target action that might be separable from the drug’s therapeutic benefit. For now, both findings are considered valid, and the field treats metformin’s iron effects as coming through multiple overlapping routes.

When Metformin’s Iron Effect Is Actually Protective

For a sizable share of metformin users, the drug’s iron-lowering property is not a side effect but an unintended benefit. Non-alcoholic fatty liver disease affects a large fraction of people with type 2 diabetes, and hepatic iron overload worsens the condition by fueling oxidative damage and cell death. In animal models of fatty liver, metformin significantly reduced iron concentrations in liver tissue and protected hepatocytes from ferroptosis.⁵ Given that elevated ferritin and transferrin saturation are common findings in diabetes patients,² metformin’s ability to pull those numbers down likely contributes to its overall metabolic benefits in ways that are only now being appreciated.

This dual nature, where the same pharmacological action helps some patients and harms others, is why the answer to “can metformin cause iron deficiency” has to be conditional. For someone with already-elevated iron and fatty liver, the iron reduction is welcome. For someone with marginal iron stores, the same reduction can tip them into deficiency. The drug does not care which scenario it is acting in; it simply pushes iron markers in the same direction. Your starting point determines whether that push is a correction or a problem.

1
BMJ Open Diabetes Research & Care. Association of metformin treatment with changes in metal dynamics in individuals with type 2 diabetes
2
PubMed Central. Elevated iron indices in patients with diabetes
3
iScience. Hyperglycemia induces hepatic iron overload and redistributes iron via transferrin-dependent pathways
4
Scientific Reports. Orphan nuclear receptor SHP regulates iron metabolism through inhibition of BMP6-mediated hepcidin expression
5
PubMed Central. Metformin alleviates hepatic iron overload and ferroptosis through AMPK-ferroportin pathway in HFD-induced NAFLD
6
iScience. High glucose diet induces hepatic iron overload contributing to metabolic dysfunction
7
iScience. Metformin alleviates hepatic iron overload and ferroptosis in non-alcoholic fatty liver disease by upregulating ferroportin via the AMPK/FPN pathway
8
PubMed Central. Anemia risk and mitigation strategies in type 2 diabetic patients: The role of novel antidiabetic agents
9
PubMed Central. Iron Deficiency Anemia and Oxidative Stress in Type 2 Diabetic Patients on Metformin: A Meta-Analysis
10
PubMed Central. Association of Metformin Use with Iron Deficiency Anemia in Urban Chinese Patients with Type 2 Diabetes
11
PubMed Central. Effect of Metformin on Serum Ferritin Level in Women with Polycystic Ovary Syndrome
12
BMJ Open Diabetes Research & Care. Metformin in gestational diabetes: the offspring follow-up (MiG TOFU): body composition and metabolic outcomes at 7–9 years of age
13
Scientific Reports. AMPK is not required for the effect of metformin on the inhibition of BMP6-induced hepcidin gene expression in hepatocytes