Horsetail, Silica and Bone Health

Horsetail has been recommended for bones for a very long time — for setting fractures, for “weak bones,” and in the modern era for osteoporosis. The modern rationale is specific and, unusually for a folk remedy, correct as far as it goes: silicon appears to be involved in the earliest stages of building bone, and horsetail is one of the most silicon-dense plants on the planet. That is a better mechanistic story than most herbs have.

What follows is the part that is usually left out. Between “silicon matters for bone” and “taking horsetail strengthens your bones” there are three separate gaps, and each one is wide enough to matter: the form of the silica in the plant, how much silicon actually reaches your blood, and whether raising silicon intake changes bone density or fracture risk in a person who takes it. This page walks those gaps honestly, then sets out what the evidence says about the other measures for thin bones.


Table of Contents

  1. The Claim and the Three Gaps
  2. What Silicon Does in Bone
  3. The Population Data and What They Can Show
  4. Silica in the Plant: How Much, and in What Form
  5. Absorbed Silicon: The Step Nobody Measured for Horsetail
  6. Human Trials of Horsetail for Bone
  7. What Actually Moves Bone Density
  8. The Diuretic Paradox: A Possible Mineral Cost
  9. Cartilage, Tendon and the Wider Scaffold
  10. Getting Silicon from Food
  11. Practical Facts About Its Use
  12. Bottom Line
  13. Key Research Papers
  14. Connections
  15. Featured Videos

The Claim and the Three Gaps

The bone claim for horsetail is a chain with four links. Only the first and the third have solid support.

  1. Horsetail is rich in silica. True, and easily measured. Tier: established analytical chemistry.
  2. That silica becomes absorbed silicon in the bloodstream. Unmeasured for horsetail specifically, and made doubtful by the fact that absorption depends almost entirely on chemical form. Tier: gap.
  3. Silicon participates in bone formation, and higher silicon intake tracks with higher bone density. Supported by laboratory work and by large observational cohorts. Tier: preliminary mechanism plus human observational.
  4. Therefore taking horsetail improves bone density or reduces fractures. Never demonstrated. Tier: gap.

A herb page that quotes links one and three and then asserts link four is doing something specific: it is using true statements to imply an untested conclusion. Once you can see the shape of the argument, you can evaluate almost any silica supplement claim in about ten seconds.

What Silicon Does in Bone

Bone is not simply a lump of calcium. It is a composite: a woven organic scaffold of type I collagen and associated proteins, into which mineral crystals of calcium phosphate are deposited. The scaffold gives bone its toughness and resistance to cracking; the mineral gives it stiffness. A bone with plenty of mineral and a poor scaffold is brittle, which is part of why bone mineral density alone is an imperfect predictor of fracture.

Silicon's proposed role is on the scaffold side, at the beginning of the process. The foundational report is Edith Carlisle's 1970 paper in Science, “Silicon: A Possible Factor in Bone Calcification,” which found silicon concentrated at the active growth fronts of forming bone and cartilage, with concentrations falling as the tissue matured and mineralised. That pattern — present where building is happening, absent where it has finished — is what pointed to silicon being involved in the process rather than merely being deposited in the product. Tier: preliminary (animal and tissue), but historically decisive. PubMed search: silicon bone calcification

Subsequent work in cells and animals has explored silicon's effects on osteoblast activity, collagen output, and the enzymes that cross-link the collagen matrix; silicon-deprived animals in classic experiments showed abnormal bone and cartilage development. In cultured bone cells, orthosilicic acid has been reported to increase collagen type I synthesis and markers of osteoblast differentiation. Tier: preliminary (cell and animal). PubMed search: orthosilicic acid osteoblast collagen type I

This is a genuinely respectable mechanism. It is also worth noticing what it implies: silicon is described as permissive — needed for the building process to run properly — not as an anabolic stimulus that drives bone growth beyond normal. A permissive nutrient helps most in someone who lacks it, and does little in someone who does not. Since human silicon deficiency has never been defined as a clinical condition, that limits how much benefit to expect from adding more.

The Population Data and What They Can Show

The most-quoted human evidence is an analysis from the Framingham Offspring cohort, published by Jugdaohsingh R and colleagues in the Journal of Bone and Mineral Research in 2004, which estimated dietary silicon intake from food-frequency questionnaires and compared it with hip bone mineral density. Higher silicon intake was associated with higher hip bone density in men and in premenopausal women, with a weaker or absent association in postmenopausal women. The study covered 2,847 people (1,251 men and 1,596 women, aged 30 to 87). Between the fifth of people eating the most silicon (more than 40 mg a day) and the fifth eating the least (less than 14 mg a day), hip bone density differed by up to 10%. In the main analysis there was no significant association at the lumbar spine (lower back) in any group, although a grouped analysis did find one in men. The authors also suggested that part of the link between moderate drinking of alcoholic beverages and bone density might be due to their silicon (PubMed PMID: 14969400). Tier: human observational. PubMed search: dietary silicon intake bone mineral density Framingham

Similar positive associations have been reported in other cohorts, and the pattern has been reviewed by Price CT and colleagues in the International Journal of Endocrinology in 2013 in the specific context of postmenopausal osteoporosis. Tier: narrative review. PubMed search: silicon and postmenopausal osteoporosis review

Three limits deserve stating clearly.

Association is not causation, and dietary silicon has an awkward companion. In Western diets, the largest single contributors to silicon intake are beer, drinking water, and whole grains. People whose questionnaires show high silicon intake therefore differ from low-silicon people in ways that have nothing to do with silicon — alcohol intake, whole-grain intake, overall diet quality, body weight and physical activity all travel with it. Statistical adjustment helps, but it cannot fully separate a mineral from the dietary pattern that carries it.

The exposure was food, not a herb. Silicon from beer and water is soluble and well absorbed. Nothing in the Framingham analysis speaks to silica from plant particulates.

The endpoint was density, not fracture. Bone mineral density is a surrogate. It correlates with fracture risk but imperfectly, and interventions that raise density do not automatically prevent fractures.

Taken together, the population data justify saying that silicon is probably worth having in the diet. They do not justify a bottle.

Silica in the Plant: How Much, and in What Form

Equisetum is a silicon accumulator in the strong sense. The plant takes up dissolved silicic acid through its roots, transports it, and precipitates it in the cell walls and epidermis of the stems as hydrated amorphous silica — biogenic opal, chemically similar to the silica in grass phytoliths or in diatom shells. Reported silica figures vary a great deal with species, growing site, harvest time and which fraction is measured, which is one reason a single percentage quoted on a product label is only approximate.

The important word is amorphous and polymerised. Silica in this form is not the small soluble molecule the gut absorbs. It is a mineral solid, mechanically hard enough that horsetail was used to sand wood and burnish metal. For it to become absorbable silicon it has to dissolve back into silicic acid — which happens slowly, depends on pH, temperature, surface area and time, and is exactly why a traditional horsetail preparation is a long simmered decoction rather than a five-minute steep.

This gives a genuinely useful practical inference. If silicon is the reason you are taking horsetail, then a prolonged decoction of finely cut herb plausibly liberates more soluble silicon than a capsule of dry powder swallowed whole or a brief infusion, and an alcohol tincture — which extracts flavonoids well and minerals poorly — is probably the worst choice of all. That reasoning is chemistry, not clinical evidence, and it should be labelled as such. Tier: mechanistic inference. PubMed search: Equisetum biogenic silica deposition

Absorbed Silicon: The Step Nobody Measured for Horsetail

Human silicon absorption has been studied properly, just not for this herb. The reference work is Sripanyakorn S and colleagues, “The comparative absorption of silicon from different foods and food supplements,” British Journal of Nutrition, 2009, which fed a range of sources to volunteers and measured urinary silicon excretion as a marker of absorption. The finding that matters here is the spread. Measured by urinary silicon, absorption was 64% of the dose for alcohol-free beer and for a monomethyl silanetriol supplement, 44% for green beans, 43% for a plain orthosilicic acid solution, 17% for choline-stabilised orthosilicic acid, 4% for bananas and for a magnesium trisilicate antacid, and 1% for colloidal silica (PubMed PMID: 19356271). Single-molecule (monomeric) silicon was absorbed readily, and particulate silicon was absorbed less and less well the more polymerised it was. An ordinary food, green beans, did better than two of the supplements. Tier: human absorption study. PubMed search: comparative absorption of silicon

The pharmacological literature on orthosilicic acid exists precisely because of this problem. Choline-stabilized orthosilicic acid was developed to keep silicon in its monomeric, absorbable state rather than letting it polymerise; the review by Jurkić LM and colleagues in Nutrition & Metabolism in 2013 lays out the chemistry and the therapeutic rationale. Tier: review. PubMed search: ortho-silicic acid bioavailability review

So the honest statement about horsetail is not “the silicon is not absorbed.” It is: some soluble silicon is certainly extracted into a decoction, nobody has published a proper bioavailability study of horsetail in humans, and the chemistry predicts the yield is well below what the total silica content would suggest. An unmeasured quantity is not zero, but it is not a basis for a confident dose either. PubMed search: Equisetum arvense silicon bioavailability

Human Trials of Horsetail for Bone

Searching the literature for a randomized controlled trial of horsetail with a bone endpoint returns essentially nothing usable. There is no adequately sized, well-controlled trial of Equisetum arvense measuring bone mineral density, bone turnover markers, or fracture. PubMed search: Equisetum arvense bone mineral density osteoporosis

What exists at a lower tier:

This is the crux of the page. The evidence tier for “horsetail builds bone in humans” is traditional use plus mechanistic plausibility — the same tier occupied by hundreds of herbs, and several tiers below what anyone treating osteoporosis needs.

What Actually Moves Bone Density

If bone is the real concern, it is worth being blunt about the size of the effects available, because the contrast is instructive.

Against that list, horsetail's plausible contribution is a small amount of a trace mineral whose deficiency in humans has never been shown to exist. It is not harmful to include; it is harmful to substitute.

The Diuretic Paradox: A Possible Mineral Cost

Here is a consideration that is almost never raised, and it cuts against the herb rather than for it. Horsetail is a genuine diuretic — that is its best-evidenced action. Diuresis increases urine flow, and some diuretics increase urinary calcium and magnesium loss. Loop diuretics notably raise calcium excretion, and chronic use is associated with bone loss; thiazides, interestingly, do the opposite and reduce calcium excretion.

So which way would horsetail go? The honest answer is that we have one relevant data point. In the 2014 randomized crossover trial of Equisetum arvense extract in 36 healthy male volunteers, the herb increased urine output about as much as the diuretic hydrochlorothiazide, without significant changes in electrolyte excretion. That is reassuring for sodium and potassium over 24 hours. It was an acute study (four days per treatment), in healthy men, and it does not tell you what months of daily use does to calcium balance. Tier: randomized clinical trial, acute, not designed for this question. PubMed search: Equisetum arvense diuretic electrolyte excretion

A longer trial followed in 2022. In a randomised double-blind study by the same Brazilian group, 58 adults aged 25 to 65 with stage I high blood pressure took 900 mg a day of a standardised dried horsetail extract or 25 mg a day of the diuretic hydrochlorothiazide for 3 months. Blood pressure fell by an average of 12.6 mmHg (systolic) and 8.1 mmHg (diastolic) on horsetail, with no significant difference from the drug, and before-and-after blood tests showed no significant kidney, liver or blood-count changes. The abstract reports no calcium-balance or bone measurement, so the question above remains open (PubMed PMID: 35168030). Tier: randomized clinical trial, 3 months, not designed for this question.

The upshot: the theoretical risk of mineral loss from prolonged herbal diuresis is unquantified, and it sits poorly with indefinite daily use for a skeletal goal that would take years to assess. It is genuinely odd to take a diuretic for years to build bone.

Cartilage, Tendon and the Wider Scaffold

Silicon's mechanistic story is about connective tissue in general, not bone alone, and cartilage was part of Carlisle's original observation. Reasonable people extend the argument to joint cartilage, tendon and blood-vessel wall, and there is a coherent basis for doing so — all are collagen-and-glycosaminoglycan tissues.

The evidence, however, is thinner still than for bone: mechanistic and animal work, essentially no clinical trials of silicon for osteoarthritis or tendon healing, and none at all for horsetail with those endpoints. Anyone taking horsetail for joints is operating on plausibility. Tier: preliminary. PubMed search: silicon cartilage tendon connective tissue See silicon and connective tissue and osteoarthritis.

Getting Silicon from Food

Because absorption is the bottleneck, the cheapest route for anyone persuaded by the silicon argument is food sources with actual absorption data. Silicon in ordinary diets comes mainly from:

Nothing here needs a supplement. The supplement form with a bone trial behind it is choline-stabilised orthosilicic acid, and its results were narrow: in a 2008 randomised placebo-controlled trial, 136 of 184 women with low bone density completed 12 months of calcium and vitamin D3 plus 3, 6 or 12 mg of silicon a day or placebo. Only a marker of collagen formation (PINP) rose significantly, at the 6 and 12 mg doses; lumbar spine bone density did not change significantly, and a femoral-neck benefit appeared only in an after-the-fact subgroup analysis at the 6 mg dose (PubMed PMID: 18547426). In the 2009 absorption study that same form was absorbed less (17%) than green beans (44%). See silicon and silicon and bone density.

Practical Facts About Its Use

Horsetail is an educational and traditional herb rather than a bone treatment, and there is no established dose for a skeletal indication. These are the facts that bear on its use:

  1. Species. The species studied and traded as horsetail is Equisetum arvense. Equisetum palustre is a toxic look-alike and a documented contaminant of commercial horsetail.
  2. Preparation. On the chemical reasoning above, a decoction of the cut dried herb — simmered, not briefly steeped — plausibly releases more silicon than other forms, and tinctures extract the mineral fraction poorly.
  3. Course length. Horsetail contains thiaminase, an enzyme that breaks down vitamin B1 (thiamine), which is why its use is limited to weeks rather than open-ended months, whatever the goal.
  4. Medicines. Its diuretic action adds to that of prescription diuretics, and a diuretic effect can shift lithium levels and the potassium balance that digoxin depends on; these are the interactions of concern.
  5. The measures with evidence — loading the skeleton, adequate calcium, vitamin D and protein, and a proper assessment after a fragility fracture — are separate from the herb, as set out above.
  6. Measuring bone density. A DEXA scan (dual-energy X-ray absorptiometry) and, where relevant, bone turnover markers give a real baseline. Taking a herb for years without measuring anything is how a decade gets lost.

Bottom Line

Horsetail's bone story is the best-reasoned weak claim in herbal medicine. Silicon genuinely appears to matter for the collagen scaffold that bone is built on, higher dietary silicon genuinely tracks with higher hip bone density in large cohorts, and horsetail genuinely is loaded with silica. Every one of those statements is true, and the conclusion still does not follow, because the silica in the plant is a poorly absorbed mineral solid, nobody has measured how much usable silicon a horsetail preparation delivers, and no trial has tested the herb against a bone outcome in people.

What the evidence supports is a short-course traditional herb with a plausible mechanism — correctly identified, decocted rather than tinctured, and not used for years on end. Nothing in it would let it stand in for load-bearing exercise, adequate calcium and vitamin D, enough protein, or prescribed treatment for osteoporosis. The tier label on “horsetail builds bone” is traditional use plus plausible mechanism, and pretending otherwise is how people spend a decade of bone loss on the wrong intervention.

Back to Table of Contents

Key Research Papers

Named papers link to their PubMed records; the remaining entries are PubMed topic searches that surface the wider literature.

  1. Carlisle EM. Silicon: A Possible Factor in Bone Calcification. Science, 1970. Silicon found concentrated at active bone and cartilage growth fronts and falling as tissue mineralised — the origin of the entire silicon-and-bone field. Tier: preliminary (animal/tissue). PubMed PMID: 5410261
  2. Jugdaohsingh R and colleagues. Dietary silicon intake is positively associated with bone mineral density in men and premenopausal women of the Framingham Offspring cohort. Journal of Bone and Mineral Research, 2004. The main human observational support — association, from food, with a density endpoint. Tier: human observational. PubMed PMID: 14969400
  3. Price CT and colleagues. Silicon: A Review of Its Potential Role in the Prevention and Treatment of Postmenopausal Osteoporosis. International Journal of Endocrinology, 2013. Collects the case and is candid about the missing trials. Tier: narrative review. PubMed PMID: 23762049
  4. Sripanyakorn S and colleagues. The comparative absorption of silicon from different foods and food supplements. British Journal of Nutrition, 2009. Silicon absorption ranged from 64% of the dose (alcohol-free beer, monomethyl silanetriol) to 1% (colloidal silica), depending on chemical form. Tier: human absorption study. PubMed PMID: 19356271
  5. Jurkić LM and colleagues. Biological and therapeutic effects of ortho-silicic acid and some ortho-silicic acid-releasing compounds: New perspectives for therapy. Nutrition & Metabolism, 2013. Why the monomeric soluble form is the one that matters. Tier: review. PubMed PMID: 23298332
  6. Spector TD and colleagues. Choline-stabilized orthosilicic acid supplementation as an adjunct to calcium/vitamin D3 stimulates markers of bone formation in osteopenic females: a randomized, placebo-controlled trial. BMC Musculoskeletal Disorders, 2008. Only the collagen-formation marker PINP changed significantly; lumbar spine bone density did not. Tier: randomized clinical trial. PubMed PMID: 18547426
  7. Orthosilicic acid in bone cells. Cell-culture reports of increased type I collagen synthesis and osteoblast differentiation markers with soluble silicon. Tier: preliminary (in vitro). PubMed search
  8. Biogenic silica deposition in Equisetum. Plant-science literature on how horsetail takes up silicic acid and precipitates amorphous silica in its stems — the reason the mineral is hard and poorly soluble. Tier: plant chemistry. PubMed search
  9. Equisetum extracts and bone cells. The small preliminary literature testing horsetail extracts on osteoblast cultures and animal bone models. Tier: preliminary. PubMed search
  10. Carneiro DM and colleagues. Randomized, Double-Blind Clinical Trial to Assess the Acute Diuretic Effect of Equisetum arvense (Field Horsetail) in Healthy Volunteers. Evidence-Based Complementary and Alternative Medicine, 2014. Relevant here because it addresses electrolyte handling. Tier: randomized clinical trial, acute. PubMed PMID: 24723963
  11. Carneiro DM and colleagues. Antihypertensive effect of Equisetum arvense L.: a double-blind, randomized efficacy and safety clinical trial. Phytomedicine, 2022. 58 adults with stage I high blood pressure, 900 mg/day standardised extract versus hydrochlorothiazide for 3 months; no significant kidney, liver or blood-count changes. Tier: randomized clinical trial, 3 months. PubMed PMID: 35168030
  12. Exercise and bone mineral density. The randomized literature on resistance and impact loading — the comparison any herbal bone claim has to be measured against. Tier: randomized clinical trials. PubMed search
  13. Diuretics and bone. The literature on how loop and thiazide diuretics differ in urinary calcium handling and bone outcomes — context for the unquantified question about a long-term herbal diuretic. Tier: human observational and trials. PubMed search
  14. Horsetail thiaminase. Why courses are limited in duration, which conflicts directly with the years-long horizon a bone benefit would require. Tier: established biochemistry. PubMed search

Back to Table of Contents

Connections


Safety and disclaimer. This page is educational and is not medical advice. Horsetail is not a treatment for osteoporosis and has no evidence that would let it replace assessment, exercise, adequate calcium and vitamin D, or prescribed therapy after a fragility fracture. The herb contains thiaminase, which degrades vitamin B1, plus trace nicotine, and it has a toxic look-alike species (Equisetum palustre); the material studied is correctly identified Equisetum arvense used in short courses. There is no adequate safety data for pregnancy, breastfeeding or children. The thiaminase matters most for people with low thiamine status, the fluid loss matters in kidney disease and heart failure, and the diuretic action adds to prescription diuretics and can shift lithium levels and the potassium balance digoxin depends on.

Back to Table of Contents