Iron and aging

Iron is essential for life and impossible for the body to excrete — so in adults who aren't losing blood, it slowly accumulates and quietly fuels the oxidative damage that drives aging. The practical upshot for a healthy midlife adult is narrow and well-supported: don't supplement iron unless you're deficient, keep ferritin in the lower half of the "normal" range, and treat high-normal iron as a signal, not a comfort.

Iron sits at the centre of an evolutionary trade-off. Its ability to flip between two charge states (Fe²⁺ and Fe³⁺) is what makes it indispensable — for carrying oxygen in haemoglobin, for the electron-transport chain that produces cellular energy, for dozens of enzymes. The same chemistry makes loose iron dangerous: it catalyses the production of the most destructive free radical in biology. Humans evolved under iron scarcity and frequent blood loss, so the body has elaborate machinery to absorb and hoard iron but no regulated pathway to get rid of it. Once childhood growth and (in women) menstruation end, iron tends to creep upward for the rest of life — steadily in men, and in women after menopause, though women often never reach male levels — and the genetic evidence now links that creep to a measurably shorter lifespan.[1]

Why iron is a longevity problem

The strongest human evidence comes from genetics. A 2020 genetic study combining European-ancestry data on healthspan (about 300,000 people), the lifespans of about 1 million study participants' parents, and extreme longevity (about 11,000 long-lived people against 25,000 controls) flagged haem (blood-pigment) metabolism as a pathway linked to aging, and a follow-up step pointed to genetically higher serum iron as harmful to the combined aging measure.[2] Because inherited gene variants are randomly assigned at conception and fixed for life, comparing people by their genetically-predicted iron level (Mendelian randomization) sidesteps most of the confounding that plagues diet-and-disease studies. For the endpoint that matters most here — total lifespan — the signal is one-directional: each standard-deviation increase in genetically-predicted serum iron was associated with roughly 0.7 fewer years of lifespan (measured as parents' lifespan) and lower odds of surviving into the oldest age bands.[3] A second, independent genetic study in Norway's HUNT cohort found the same pattern: on average, genetically higher blood iron meant earlier death, with weak signs that more iron helps people whose levels sit below the normal range.[4] These estimates describe small lifelong differences, not what a supplement or a blood donation would do.

The disease-by-disease picture is not a single arrow, and it's worth stating plainly — Moderate genetic evidence. Higher genetically-predicted iron actually lowers the risk of coronary artery disease[5] and, counterintuitively given the iron that accumulates in Parkinsonian brains, of Parkinson's disease — about a 3% relative reduction per 10 µg/dL of serum iron, a direction the authors themselves flag as surprising and not yet clinically actionable.[6] It raises the odds of type 2 diabetes (by roughly 6–19%, depending on the iron marker)[7] and of stroke, mainly stroke caused by clots that form in the heart (cardioembolic stroke).[8] The honest one-line summary is that the net effect on lifespan is negative even though the disease-specific effects diverge — a reminder that "iron is bad" is too blunt, and that systemic iron and tissue iron can move in opposite directions.

This is causal-inference evidence, not a randomized trial, and the effect per unit is modest. Newer genetic studies add a supporting line: people genetically inclined to higher iron also tend to have "older" epigenetic clocks (blood-based estimates of biological age), roughly 0.3 to 0.5 years older per standard deviation of serum iron depending on the clock, while clock age does not predict iron.[9][10] These studies draw on the same few genetic markers for iron, so they support the lifespan finding rather than independently confirming it. But the genetic case converges with basic biochemistry and with the simple fact that the body cannot offload iron — which is why researchers increasingly treat iron as a hidden driver of aging.[11] It is not one of the 12 recognised hallmarks of aging,[12] but it feeds several of them: DNA damage, mitochondrial decline, and cellular senescence.

How excess iron damages cells

Inside a healthy cell, almost all iron is locked inside ferritin, a hollow protein shell that holds thousands of iron atoms in a chemically inert form, releasing them only on demand. Aging erodes that control, and free iron does its damage through one reaction.

The Fenton reaction. Normal respiration constantly produces small amounts of hydrogen peroxide. On its own it is mild. But in the presence of loose Fe²⁺ it is converted into the hydroxyl radical — the most reactive oxidant the body generates, with no enzyme to neutralise it. It indiscriminately shreds the nearest molecules: peroxidising membrane fats, misfolding proteins, and breaking DNA strands.[13] Mitochondria, which both consume the most iron and sit closest to the leak, take the worst of it — a self-reinforcing loop of damage, more leakage, and falling energy output.

Nutrient sensing and autophagy. Beyond direct oxidation, iron acts as a growth signal that helps keep the growth-sensing mTOR pathway (mechanistic target of rapamycin) switched on. Sustained mTOR activity in adulthood suppresses autophagy — the cellular recycling that clears damaged mitochondria and protein junk — and is itself a recognised pro-aging driver (the same pathway that calorie restriction and rapamycin act on; see protein). In worms, restricting iron switches on stress-resistance and recycling programmes and extends lifespan, mirroring those interventions.[14]

Ferrosenescence — Weak / preliminary. A newer and still largely preclinical idea is "ferrosenescence": senescent (worn-out but un-dying) cells accumulate large amounts of iron, but trap it inside undegradable ferritin, which paradoxically makes them resist the iron-dependent cell death (ferroptosis) that might otherwise clear them. They persist instead, leaking inflammatory signals.[15] The mechanism is plausible and actively researched, but it is animal- and cell-level work — treat it as a hypothesis, not an established target. Ferroptosis itself — iron-driven cell death through runaway oxidation of membrane fats — is implicated across neurodegeneration, heart failure, and joint disease, but the human evidence is associative and no ferroptosis-targeting therapy is approved for any aging condition; it is a mechanistic frontier, not a lever.

The natural experiment: hereditary haemochromatosis — Strong. The clearest human demonstration that chronic iron excess causes cumulative damage comes from people who inherit two faulty copies of the HFE gene (about 1 in 150 of European ancestry), which lets iron load without limit.[16] In about 450,000 UK Biobank adults followed for over a decade, men with two copies of the main HFE variant were estimated by age 80 to have died more often (about 33% against 25%), and to have had more joint replacements (28% against 17%), more liver disease (20% against 8%), and excess dementia, delirium and Parkinson's disease. The excess held even in men never diagnosed with haemochromatosis.[17] Their risk of liver cancer was roughly ten times higher.[18] Women with the same genotype had more liver disease, joint replacements and delirium, but less excess than men,[19] and no clear excess in deaths.[20] Because the damage is largely preventable by regular blood-letting to keep iron down, it is a clean, hard-outcome anchor for the whole chronic-excess thesis.

The iron–heart hypothesis

In 1981 Jerome Sullivan proposed that premenopausal women have less heart disease than men because they store less iron: menstrual blood loss keeps their stores low, and the protection fades as iron builds up after menopause. He suggested that lowering iron by regular blood-letting could test the idea.[21] The mechanism is coherent — stored iron catalyses the oxidation of LDL cholesterol, and it is oxidised LDL that macrophages engulf to form the foam cells at the core of arterial plaque.

The epidemiology is supportive but the interventional evidence is not. The one large randomized test of the idea — the FeAST trial, which used repeated phlebotomy to lower iron in patients with peripheral artery disease — found no overall reduction in mortality or cardiovascular events: all-cause death and the combined death-heart-attack-stroke endpoint were both statistically null.[22] A pre-specified subgroup — the youngest quartile, roughly ages 43 to 61 — did show markedly fewer deaths and events, but this is a subgroup finding inside an overall-negative trial and should be read as hypothesis-generating, not as a green light to bleed toward benefit. So the honest reading is: avoiding iron overload is sound, but the claim that actively bleeding healthy people prevents heart attacks is not established by randomized data. A 2026 pooled analysis of small studies and a small trial, both from one research group, found that removing blood slightly improves insulin resistance on lab tests, in people with metabolic problems linked to high ferritin[23] and in overweight blood donors with prediabetes,[24] but nobody has shown that this prevents diabetes, heart attacks or deaths. That gap between surrogate markers and hard outcomes is the central calibration point of the whole topic — the genetic and mechanistic case is stronger than the treatment case.

Brain iron and neurodegeneration

The brain is unusually exposed. It burns oxygen heavily, is rich in the polyunsaturated fats that lipid peroxidation attacks, and accumulates iron with age in exactly the regions tied to movement and cognition — the substantia nigra and basal ganglia. This is not passive: in aging mice, the cortex ramps up production of hepcidin, the hormone that destroys the cell's only iron exporter, effectively trapping iron inside brain cells.[25]

Excess iron is a recurring feature across neurodegenerative disease: it concentrates in the substantia nigra in Parkinson's, and it accelerates the aggregation and toxicity of amyloid and tau in Alzheimer's.[26] Association is not causation, but genetic evidence now points to a causal role: in about 40,000 UK Biobank adults, genetically higher iron in specific deep brain regions raised the risk of Parkinson's and some non-Alzheimer dementias, though not Alzheimer's itself.[27] That sits alongside the finding above that genetically higher blood iron goes with slightly lower Parkinson's risk: brain iron and body iron are different things. Lowering body iron does not seem to clear brain iron either. In a small study of people with haemochromatosis, blood ferritin and how often they had been bled poorly predicted how much iron sat in their deep brain regions,[28] and whether blood donation changes brain iron in ordinary adults has not been tested.

The trials that lowered brain iron backfired — Moderate. This is the crucial calibration for anyone tempted to treat brain iron as a lever. Two well-run randomized trials used deferiprone, a drug that crosses into the brain and does lower brain iron, and both made outcomes worse, not better. In newly-diagnosed, untreated Parkinson's disease, deferiprone cut brain iron on MRI but produced worse movement scores than placebo over 36 weeks — probably because iron is needed to make dopamine, so chelating it in untreated patients starves dopamine synthesis.[29] In amyloid-confirmed early Alzheimer's, deferiprone likewise lowered brain iron but accelerated cognitive decline over a year, with greater frontal brain-volume loss.[30] The one partial exception is pantothenate-kinase-associated neurodegeneration, a rare inherited brain-iron-overload disease, where deferiprone lowered brain iron and showed a trend toward slower progression that missed statistical significance in its randomized trial.[31] Either way, that is a genetic overload disorder, not ordinary aging. So iron is one thread in dementia risk, not a proven lever, and in the brain the trial evidence argues actively against lowering it, even if brain iron does contribute to the disease — the established levers live under dementia prevention.

Dietary iron: heme vs. non-heme

Where iron comes from matters more than the headline milligram total.

  • Heme iron — from red meat, poultry, and fish — is absorbed at 15–35% through a dedicated pathway that is less tightly governed by iron status than non-heme absorption, so the body only partly throttles it when stores are already full. It is also a pro-oxidant in the gut. In three large US cohorts (about 205,000 adults followed for up to 36 years), people in the top fifth for heme iron intake had about a quarter higher risk of type 2 diabetes than those in the bottom fifth, and heme iron accounted for roughly two-thirds of the diabetes risk linked to unprocessed red meat.[32] The colorectal-cancer angle is covered under red and processed meat.
  • Non-heme iron — from plants and fortified foods — is absorbed less efficiently (2–20%) and, crucially, is regulated: the gut takes up less when stores are adequate. In the same US cohorts non-heme iron was unrelated to diabetes,[33] and in a Chinese cohort of about 6,000 adults, people eating the most non-heme iron had about 30% lower risk of developing two or more cardiometabolic diseases over six years.[34]

Because non-heme absorption is regulated and modifiable, it's also controllable at the table. Polyphenols and tannins (in tea, coffee, cocoa) and phytates (in legumes and whole grains) bind non-heme iron and can cut its absorption from a meal by half or more: in single-meal studies, a cup of tea cut it by about two-thirds and a cup of coffee by about 40%.[35] Coffee drunk an hour before the meal had no effect, but coffee drunk an hour after it inhibited absorption as much as coffee with the meal. See tea and coffee. Vitamin C does the opposite, sharply increasing absorption. So the iron-deficient should pair plant iron with citrus and away from tea; the iron-replete adult can do the reverse without effort.

Supplements are the sharp end — Caution. Iron supplements deliver a large, unregulated bolus. No benefit has been shown in people who are not deficient. In US national survey data, adults whose total iron intake ran above the 45 mg/day upper limit, mostly from supplements, had blood-biomarker ages about 1.8 years older, partly through inflammation,[36] while iron from food went with a younger biological age.[37] These are snapshot studies that cannot prove cause. Don't take iron without a documented deficiency.

Biomarkers: what to actually measure

Standard lab reference ranges are built to catch overt anaemia or full iron-overload disease, not to optimise aging — they tolerate a lot of accumulation before flagging anything. Two cheap tests carry most of the signal, and both have a U-shape: too little is as bad as too much.

MarkerStandard "normal"Longevity-oriented targetWhy it matters
Ferritin (stored iron)~10–300+ µg/LAbove 30 µg/L; below 150 µg/L (women who menstruate) or 200 µg/L (men, other women) — the lowest mortality among about 162,000 Copenhagen adults whose doctors had ordered blood tests sat near 60 µg/L in women and 125 µg/L in men without marked inflammation, an association rather than a tested target [38]Under 30 µg/L usually means iron deficiency [39]; 200 µg/L or more went with about 10% higher mortality in a Danish cohort [40]; very low impairs energy, endurance, cognition
Transferrin saturation (iron in transit)20–50%24–40%Above about 55% went with roughly 60% higher all-cause mortality in US adults followed from the early 1970s [41]; a later US analysis of about 29,000 adults saw risk rise above about 50%, and people above the lowest quarter (under about 17%) had 6–8% lower mortality than those in it [42]; below 20% signals iron restriction [43]

A practical caveat: ferritin is also an inflammation marker, so a high reading must be interpreted alongside an inflammation marker (hs-CRP, high-sensitivity C-reactive protein) and transferrin saturation before being called iron overload. If ferritin is high with normal inflammation and saturation above ~45%, a hereditary haemochromatosis work-up is appropriate. The full panel and how to read it sit in midlife labs.

Supporting evidence for the U-shape: a large Catalan records study compared about 7,400 people who reached 100 with people over 92 who died before 100. After adjustment, those with ferritin in the middle of the range were about 20% more likely to reach 100 than those with the lowest levels, while the highest levels showed no clear advantage.[44] This is observational, and the authors warn of possible residual confounding, so it supports "too little is bad and more is not better" rather than a "lower is better" rule.

What to do

Ranked by evidence-to-effort for a healthy midlife adult:

  1. Don't supplement iron unless a blood test shows deficiency. This is the single highest-yield rule. Multivitamins-with-iron are unnecessary for most non-menstruating adults.
  2. Measure ferritin and transferrin saturation yearly, with routine bloods, and aim for the range in the table — above 30 µg/L and below the WHO overload thresholds. Interpret a high ferritin against hs-CRP.
  3. Moderate heme iron. Keep unprocessed red meat occasional rather than daily; the protein and B12 are available from fish, poultry, and plants with less pro-oxidant load.
  4. Use ordinary dietary levers. Tea or coffee with iron-rich meals lowers non-heme absorption; save vitamin-C pairing for when you actually want more iron.
  5. Blood donation is reasonable and pro-social, within limits. It reliably lowers iron stores. The mortality benefits claimed in donor studies are heavily confounded by the "healthy donor effect," so donate because it helps others and modestly offloads iron — not as a proven longevity treatment. Cap it at roughly 3–4 times a year for men and 1–2 for premenopausal women, leaving at least 12 weeks between donations for men and 16 weeks for women (the WHO minimums).[45] In a trial of about 45,000 English donors, donating more often than the UK standard lowered haemoglobin and ferritin and brought more tiredness, dizziness and deferrals for low haemoglobin.[46] Routine haemoglobin checks at donation don't catch falling iron stores: when Dutch donation centres began measuring ferritin and lengthening the gap to 6 months below 30 µg/L (12 months below 15 µg/L), the odds of iron deficiency among donors fell by about three-quarters in men and postmenopausal women and by about half in premenopausal women.[47] If you donate regularly, get ferritin checked.

What's overrated

  • Aggressive phlebotomy as anti-aging therapy. The randomized data (FeAST) didn't show the mortality benefit the hypothesis predicts. Offloading overload is sensible; bleeding toward deficiency is not.
  • Iron-chelator supplements (EGCG, quercetin) for "brain detox." Green-tea EGCG (epigallocatechin gallate) does chelate iron and crosses into the brain in lab models,[48] but the human evidence for slowing neurodegeneration by this route is preliminary. Drinking tea is fine; high-dose EGCG extracts carry liver-toxicity risk and aren't a proven neuroprotectant.
  • Precise dietary "breakpoints." Two US survey snapshots tied faster biological aging to supplement iron, not food iron: one put the turning point for total iron intake at about 18.4 mg/day,[49] the other saw harm only from supplements above 18 mg/day.[50] Neither measured heme iron nor could show cause, so the real signal is the supplement source and the body's stored level, not a decimal-point intake target.

Cautions

  • Deficiency is the more common everyday problem, especially in premenopausal women, endurance athletes, frequent donors, and long-term plant-based eaters. Low iron causes fatigue, hair loss, poor exercise recovery, restless legs, and impaired cognition — sometimes before ferritin drops below the usual cutoffs. WHO defines deficiency as ferritin below 15 µg/L in healthy adults, or below 70 µg/L during infection or inflammation; UK gastroenterology guidance uses 30 µg/L, the floor in the table above.[51] Some experts argue for a cutoff of about 45–50 µg/L,[52] but WHO has not adopted it, and critics note it would label nearly half of women deficient. Specialists already use higher cutoffs in some conditions: the US sleep-medicine guideline advises iron for restless legs when ferritin is 75 µg/L or below, an expert-consensus figure.[53] The goal is low-normal, not empty.
  • In older adults, the deficiency side of the U becomes a serious clinical problem — Strong for prevalence. Anemia affects roughly one in ten community-dwelling people over 65 and more than one in five past 85, with far higher rates in hospital and nursing homes.[54] Even mild anemia independently tracks with higher mortality in this age group, and US death-certificate data for people over 65 show deaths with iron-deficiency anemia listed as a cause holding steady from 1999 to 2013, then rising by about 10% a year to 2019.[55] Iron deficiency is only part of the picture — it accounts for roughly one-sixth of anemia cases in the elderly, with inflammation, kidney disease, and an unexplained group making up the rest — so don't equate "anemia in an older adult" with "needs iron." Two distinct problems hide behind the same low serum iron: true iron-deficiency anemia (real depletion, which needs replacement) versus anemia of chronic inflammation (functional iron restriction driven by the hormone hepcidin, in which supplementing does nothing useful and may cause harm). Because a low ferritin can also flag occult bleeding or malignancy, low iron in an older adult is a reason for a clinical work-up, not a reason to self-treat. Even when the deficiency is real, correcting it is not a longevity lever: in adults with heart failure and genuine iron deficiency, three large trials of intravenous iron found at most a small, borderline drop in heart-failure hospital stays and no clear gain in survival.[56][57][58] Fixing a real deficiency is still worth doing.
  • Don't act on a single ferritin value during illness or hard training — both inflate it.
  • This is general information, not a prescription. Hereditary haemochromatosis, anaemia, and kidney disease all change the calculus and need clinical management.

Further reading

  • Mangan D. Iron: an underrated factor in aging. Aging (Albany NY) 2021.[59]
  • Daghlas I, Gill D. Genetically predicted iron status and life expectancy. Clinical Nutrition 2021.[60]
  • Timmers PRHJ et al. Multivariate genomic scan implicates novel loci and haem metabolism in human ageing. Nat Commun 2020.[61]
  • Tian Y et al. Iron metabolism in aging and age-related diseases. Int J Mol Sci 2022.[62]
  • Sato T et al. Aging is associated with increased brain iron through cortex-derived hepcidin expression. eLife 2022.[63]
  • Ward RJ et al. The role of iron in brain ageing and neurodegenerative disorders. Lancet Neurology 2014.[64]
  • Chen et al. Dietary iron intake and the incidence of cardiometabolic multimorbidity: the mediating role of biological age. BMC Public Health 2025.[65]
  • Moreno MA et al. Blood-based biomarkers in centenarians and non-centenarians: a matched, population-based retrospective cohort study. Biogerontology 2025.[66]
  • Cançado RD et al. Defining Global Thresholds for Serum Ferritin: A Challenging Mission in Establishing the Iron Deficiency Diagnosis in This Era of Striving for Health Equity. Diagnostics (Basel) 2025.[67]
  • Devos D et al. Trial of deferiprone in Parkinson's disease (FAIRPARK-II). New England Journal of Medicine 2022.[68]
  • Ayton S et al. Deferiprone in Alzheimer disease: a randomized clinical trial. JAMA Neurology 2025.[69]
  • Pilling LC et al. Common conditions associated with hereditary haemochromatosis genetic variants: cohort study in UK Biobank. BMJ 2019.[70]
  • Lucas MR et al. HFE genotypes, haemochromatosis diagnosis and clinical outcomes at age 80 years: a prospective cohort study in the UK Biobank. BMJ Open 2024.[71]
  • Atkins JL et al. Association of Hemochromatosis HFE p.C282Y Homozygosity With Hepatic Malignancy. JAMA 2020.[72]
  • Wang F et al. Integration of epidemiological and blood biomarker analysis links haem iron intake to increased type 2 diabetes risk. Nat Metab 2024.[73]
  • Guralnik JM et al. Prevalence of anemia in persons 65 years and older in the United States. Blood 2004.[74]
  • Zacharski LR et al. Reduction of iron stores and cardiovascular outcomes in peripheral arterial disease (FeAST). JAMA 2007.[75]

— § —