Stinging Nettle for Joint Pain and Traditional Urtication

People deliberately sting themselves with nettles to relieve arthritis pain. Not metaphorically — they pick a fresh nettle, hold it against the painful joint, and rub. The practice is called urtication, from the plant’s Latin name, and it is documented in Europe going back to Roman writers who described nettle being applied to lame joints.

It would be easy to file this under harmless folk nonsense, except for one inconvenient detail: somebody actually ran the trial. A general practitioner in Plymouth named Colin Randall spent about a decade investigating it properly, including a randomized double-blind crossover study with a genuinely clever placebo — a non-stinging look-alike plant. The results are more interesting than either the enthusiasts or the sceptics would like. One trial was clearly positive. The next was completely flat. This article gives you both.

What Urtication Is
The Sting: What Is Actually in Those Hairs
The 1999 Study That Started It
The Thumb Trial: The Positive One
The Knee Trial: The Flat One
Why the Two Trials Disagree
How Counter-Irritation Might Work
How People Who Do This Actually Do It
Nettle Taken by Mouth for Joints
Creams, Extracts and Commercial Products
If You Are Going to Try It
Cautions
Key Research Papers
Connections

What Urtication Is

Urtication is the deliberate application of a stinging nettle to the skin over a painful area — usually an arthritic joint — to produce a controlled sting. The skin reddens, itches and burns within seconds, raises small weals that persist for anything from twenty minutes to several hours, and then subsides. The claim is that the joint feels better afterwards, sometimes for days.

It belongs to a family of practices that turn up independently across many cultures: making a painful area hurt in a different, more superficial way in order to relieve the deeper pain. Cupping, moxibustion, mustard plasters, capsaicin creams, wintergreen rubs and menthol gels are all variations on the same idea. Capsaicin cream is the most respectable modern member of that family — it is licensed, sold in pharmacies, and works by burning.

What makes nettle unusual is that a plant you can pick from a hedgerow performs the same trick for free, and that in Britain the practice never quite died out. It survived in rural memory well enough that when Randall advertised for people who used nettle sting for joint pain, he had no difficulty finding them.

The Sting: What Is Actually in Those Hairs

The mechanism of the sting is well understood. Nettle leaves and stems are covered in trichomes — hollow hairs with a brittle silica tip sitting on a bulb of fluid. Brush against one and the tip snaps off at a pre-formed break point, leaving a bevelled, needle-sharp opening that penetrates the skin. Pressure on the bulb injects the contents. It is a genuine hypodermic syringe, evolved independently of anything human beings invented.

What comes out of it is where the story gets more interesting than most herbals admit.

The classic account

Emmelin and Feldberg, working at Cambridge in 1947, identified histamine and acetylcholine in nettle-hair fluid and showed these accounted for much of the immediate sting. Later work added serotonin (5-HT). Formic acid — the compound in ant venom — entered the story early and has been repeated ever since. Oliver and colleagues (1991) studied nettle contact urticaria in humans with histology, electron microscopy and pharmacology, confirming a histamine-driven immediate weal, and noting a persistent component that antihistamines did not fully explain.

The revision

That last point matters, because the classic four do not account for the part of the sting that lasts. Fu and colleagues, publishing in Annals of Botany in 2006, took the stinging hairs of Urtica thunbergiana, extracted the fluid and analysed it by HPLC. They found the major constituents were histamine, oxalic acid and tartaric acid — and that serotonin and formic acid were present only at low concentrations. Behavioural testing in rats found that 2% oxalic acid and 10% tartaric acid produced persistent pain, while 10% formic acid and 2% serotonin produced only moderate pain in the first ten minutes, and acetylcholine and histamine at those concentrations produced no significant pain response at all.

Their conclusion, in their own words, is that the general view that formic acid, histamine and serotonin are the pain-inducing agents in Urtica dioica stinging hairs “may require updating”.

Two honest caveats. That study was on Urtica thunbergiana, a related Asian species, not on Urtica dioica itself — species differ, and the finding has not been fully replicated in the European nettle. And histamine clearly does something in humans: the immediate weal and flare are exactly what histamine produces in skin, as Oliver’s work showed. The reasonable synthesis is that histamine drives the immediate itch and weal, while organic acids drive the burn that outlasts it — and that formic acid, the compound every popular article names first, is probably the least important thing in the list.

It is a small but useful lesson in how herbal facts propagate. “Nettles sting because of formic acid” has been copied from book to book for over a century, and the one team that measured it found barely any.

The 1999 Study That Started It

Randall, Meethan, Randall and Dobbs published an exploratory study in Complementary Therapies in Medicine in 1999. It is important to be precise about what it was: eighteen self-selected people who already used nettle sting for joint pain were interviewed about how they applied it, at what dose, and with what effect. All but one were sure nettles had helped a great deal, and several considered themselves cured. No side effects were reported beyond a transient urticarial rash.

This is qualitative research, not a trial. Eighteen people who chose to keep doing something and then volunteered to talk about it are the most favourably biased sample imaginable — anyone for whom it did not work simply stopped and never answered the advertisement. The authors understood this perfectly, which is why the paper ends by announcing that a randomized controlled trial was planned with a rheumatology specialist.

That is the correct way to handle a striking folk practice: take it seriously enough to describe it properly, then test it.

The Thumb Trial: The Positive One

Randall and colleagues published that trial in the Journal of the Royal Society of Medicine in 2000, and its design is the reason it is worth discussing at all.

Design. A randomized, double-blind, crossover study in 27 patients with osteoarthritic pain at the base of the thumb or index finger — carpometacarpal or metacarpophalangeal joint osteoarthritis, a common and stubbornly painful condition. Patients applied stinging nettle leaf (Urtica dioica) daily for one week to the painful area. After a five-week washout, they crossed over to the placebo: white dead-nettle (Lamium album), a plant that looks much like a nettle and does not sting. Pain and disability were recorded across twelve weeks.

Result. After one week of nettle sting, reductions on both the visual analogue pain scale and the health assessment questionnaire for disability were significantly greater than with placebo — P = 0.026 for pain and P = 0.0027 for disability.

The placebo problem, and how they handled it. Blinding a treatment whose entire point is that it hurts is close to impossible. Everyone knows whether they have been stung. Using a look-alike plant is the best available answer — it controls for the ritual, the leaf, the rubbing, and the expectation attached to applying a plant to your hand — but it cannot control for the sensation itself. That is an acknowledged and unavoidable limitation, not a flaw the authors hid. The disability result, incidentally, has the stronger P value, and disability is harder to talk yourself into than pain.

Twenty-seven patients is small. But this is a real, published, randomized, double-blind, placebo-controlled crossover trial of stinging yourself with a weed, and it was positive.

The Knee Trial: The Flat One

Eight years later the same group published in Complementary Therapies in Medicine (2008), and the result did not repeat.

Design. A patient-blinded pilot randomized controlled trial in two inner-city primary care practices in Plymouth. 42 patients aged 55–80 with knee pain and a presumptive clinical diagnosis of knee osteoarthritis, with a baseline WOMAC pain subscale score above 4. Randomized to Urtica dioica or to placebo — this time Urtica galeopsifolia, a nettle variant that does not sting — daily for one week.

Result. Mean reduction in WOMAC pain score was 1.7 (95% CI 0.6 to 2.9) in the nettle group and 1.6 (95% CI 0.5 to 2.7) in the controls.

Those two numbers are indistinguishable. Both groups improved by about the same amount, which is what you expect from regression to the mean, attention, and the natural fluctuation of osteoarthritis pain. The trial found no treatment effect.

What it did establish — and what it was actually designed to establish, since it was a feasibility study — is that this research is practical: 42 of 45 eligible patients agreed to take part, every GP practice approached was willing, and patients found the sting to be no more than a minor irritation. Several liked the treatment specifically because it was “natural”. The authors concluded that larger rigorous studies are justified.

As far as the published record goes, those larger studies have not been done. Twenty-five years after the thumb trial, the total human evidence for urtication remains 27 patients and 42 patients.

Why the Two Trials Disagree

Several explanations are compatible with the data, and it is worth being clear that nobody knows which is right.

Chance. The simplest explanation. A positive result in 27 patients and a null result in 42 is entirely consistent with a treatment that does nothing, and with a small trial that got lucky. Small trials produce false positives routinely.
Anatomy. The thumb base is a small joint sitting a few millimetres beneath thin skin. The knee is a large joint under a thick soft-tissue envelope. If the effect depends on stimulating nerves near the joint, the thumb is a far more promising target — and this is the explanation the researchers themselves favoured.
Outcome measures. The thumb trial used a visual analogue scale and a disability questionnaire; the knee trial used the WOMAC pain subscale. Different instruments have different sensitivity to change.
The placebo. Lamium album is visibly a different plant to anyone who looks closely. Urtica galeopsifolia is a nettle that does not sting — a tighter control, and arguably a harder test.
Duration. Both trials ran one week. Osteoarthritis is measured in years.

The honest position: urtication is an old practice with one small positive trial in a small joint, one small null trial in a large joint, and no adequately powered study of either. It is not established. It is also not refuted, and it has not been shown to be harmful.

How Counter-Irritation Might Work

If there is an effect, the most likely explanation is counter-irritation, and there is real neurophysiology behind that idea rather than hand-waving.

Gate control. The classical account: intense input along fast-conducting sensory fibres inhibits transmission of slower pain signals at the spinal cord. This is why rubbing a banged elbow helps, and it operates on a timescale of seconds to minutes.
Diffuse noxious inhibitory control (DNIC). A better fit for a longer effect. A painful stimulus applied anywhere on the body activates descending pathways from the brainstem that suppress pain signalling generally — pain inhibiting pain. This is a measurable phenomenon in humans, and it is notably weakened in people with chronic pain conditions.
Neuropeptide depletion. The mechanism of capsaicin. Repeated stimulation of TRPV1-expressing sensory neurons depletes substance P and eventually causes reversible defunctionalisation of those nerve endings. This takes days to weeks of repeated application, which fits the way regular nettle users describe it — several sessions before the benefit builds.
Local vasodilation and inflammatory mediators. Histamine causes a local flare and increased blood flow. Whether that does anything useful for an arthritic joint is unknown.
Expectation. A treatment that produces a dramatic, unmistakable sensation is a powerful generator of expectation. This is not an insult to anyone who finds it helps — expectation-mediated analgesia involves real endogenous opioid release — but it is a mechanism that would produce the same reported benefit whether or not the nettle contributed anything specific.

How People Who Do This Actually Do It

There is an unusual document in this literature: White, Randall, Harding and Paterson published a patient consensus in 2011, developed by an expert-panel method with people who actually use nettle sting for musculoskeletal pain. Sixteen users responded to media appeals, nine attended a meeting, they answered questions on method, discussed, and answered again; consensus was set at 70% agreement.

Three points reached consensus immediately:

It is important to feel a good strong sting — a weak sting is considered not to work.
Immediately sting the area again after the first application.
Sting once a day.

Users did not agree on method and site of application, or on how a course of treatment should be structured, so the published consensus document incorporates a range of practice rather than a single protocol.

The reason to cite this is not that patient consensus establishes efficacy — it does not. It is that if anyone ever runs the adequately powered trial, this is the dosing schedule it should test, arrived at by the people with the most practical experience.

Nettle Taken by Mouth for Joints

Separately from the sting, nettle leaf is taken orally as a mild anti-inflammatory. The evidence here is also thin, and it points in a specific direction: not that nettle relieves arthritis, but that it might let people use less anti-inflammatory drug.

Chrubasik 1997. Forty people with acute arthritis, open-label randomized, comparing 50 mg diclofenac plus stewed nettle herb against 200 mg diclofenac alone. Thirty-seven completed. Both groups improved by about 70% on CRP and clinical scores. The authors concluded that stewed nettle may enhance the antirheumatic effect of an NSAID.

Read carefully, that is an interesting design and a weak study. Interesting, because getting the same result on a quarter of the drug dose would be genuinely valuable — NSAID side effects (gastrointestinal bleeding, kidney injury, cardiovascular risk) are dose-related and are what limit their use in older people. Weak, because it was open-label with no placebo, no nettle-only arm, and 40 participants. It cannot distinguish a real drug-sparing effect from the natural course of acute arthritis in both arms. It has not been replicated in the twenty-eight years since.

Creams, Extracts and Commercial Products

Nettle creams. Topical nettle preparations are widely sold for arthritis. Rayburn (2009) discussed nettle cream for osteoarthritis in Alternative Therapies in Health and Medicine. A cream is not urtication — it contains no intact stinging hairs and delivers no sting — so whatever the trials of the fresh plant showed, they do not transfer to it. Any effect would have to come from absorbed constituents, which is a different and untested claim.
Hox alpha. A standardised 2-propanolic dry extract of nettle leaf sold in Germany for osteoarthritis. A 2026 analysis of the German Pain e-Registry (Überall and colleagues) compared 1,073 propensity-matched osteoarthritis patients on this extract with 1,073 on over-the-counter NSAIDs, and reported better outcomes and far fewer adverse drug reactions with the nettle extract — 13.1% versus 46.8% of patients affected. Read that with the design in mind: it is a retrospective registry analysis, not a randomized trial. Propensity matching cannot remove the fundamental difference between people who choose a herbal product and people who take NSAIDs, and the authors state plainly that the findings are exploratory and hypothesis-generating and need confirmation in randomized trials. The adverse-event comparison is the more believable half of the result — NSAIDs genuinely do cause a great deal of gastrointestinal trouble.
Phytalgic. A combination food supplement containing nettle alongside fish oil and vitamin E, tested against placebo in knee and hip osteoarthritis (Jacquet 2009) with positive results. It attracted a published critique the following year (Christensen and Bliddal 2010) questioning the trial and its reporting. Whatever the truth of it, a three-ingredient product cannot tell you what the nettle did — the fish oil alone has its own literature in joint pain.

If You Are Going to Try It

Urtication is unproven. It is also close to free, does not interact with anything you swallow, and in two clinical trials caused nothing worse than a transient rash. That is an unusual risk-benefit profile, and for a small, superficial joint it is a defensible thing to experiment with.

Small superficial joints only. Thumb base, finger joints, wrist. The one positive trial was in the thumb; the null trial was in the knee.
Fresh plant, picked cleanly. Only the living plant stings. Away from roadsides, sprayed ground and dog-walking routes.
Gloves to pick, bare hand to apply. Hold the leaf by the stem and stroke the underside of the leaf against the skin over the joint.
Once daily, per the users’ consensus, with a second immediate application to the same area.
Give it a week or two and write it down. Score pain 0–10 daily before and during. Both trials ran one week; that is a fair test.
Stop if the skin reaction is more than transient — if the weals last beyond a few hours, spread, or blister.
Do not do this instead of treatment. Inflammatory arthritis — rheumatoid, psoriatic, gout — is not osteoarthritis and needs proper drug treatment to prevent joint destruction. Get the diagnosis right first.
Consider capsaicin. If counter-irritation is the mechanism you are after, capsaicin cream does the same job in a standardised dose, from a pharmacy, with a real evidence base and no need to identify a plant.

Cautions

Never near the eyes, on broken skin, on mucous membranes or on genitals. Obvious, and it needs saying.
Not on numb skin. Anyone with peripheral neuropathy — diabetic or otherwise — cannot judge how strong a sting they are getting. The same applies to anyone with impaired sensation from any cause.
Not on poor-healing or fragile skin. Long-term oral steroids, poorly controlled diabetes, peripheral vascular disease, lymphoedema in the limb, or very thin elderly skin all make deliberate skin trauma a bad idea.
Nettle allergy exists. Uncommon but real. Systemic reactions to nettle sting have been described. If you get anything beyond local weals — widespread hives, wheeze, swelling of lips or throat — stop and seek help immediately.
Not in children. Deliberately hurting a child with a plant is not an acceptable treatment plan.
Persistent skin reactions. Most stings settle within a few hours. Oliver’s work documented that some people get a longer-lasting reaction. If yours lasts more than 12 hours or recurs, stop.
Oral nettle leaf carries the leaf cautions — principally the vitamin K and warfarin interaction, plus possible additive effects with antihypertensives, diuretics and diabetes medication. See the nutrition article.
Do not skip the diagnosis. A hot, swollen, exquisitely tender single joint can be septic arthritis or gout, both of which are urgent. Neither is a nettle problem.

Key Research Papers

Every identifier was verified live against NCBI E-utilities before being written. The null result is listed here with the same prominence as the positive one, because that is the honest shape of this evidence base.

Randall C, Meethan K, Randall H, Dobbs F. Nettle sting of Urtica dioica for joint pain — an exploratory study of this complementary therapy. Complementary Therapies in Medicine. 1999;7(3):126–131. Qualitative, 18 self-selected users interviewed. Not a trial.
Randall C, Randall H, Dobbs F, Hutton C, Sanders H. Randomized controlled trial of nettle sting for treatment of base-of-thumb pain. Journal of the Royal Society of Medicine. 2000;93(6):305–309. Human RCT, n = 27, crossover, placebo = Lamium album. Positive: pain P = 0.026, disability P = 0.0027.
Randall C, Dickens A, White A, Sanders H, Fox M, Campbell J. Nettle sting for chronic knee pain: a randomised controlled pilot study. Complementary Therapies in Medicine. 2008;16(2):66–72. Human RCT, n = 42, placebo = Urtica galeopsifolia. Null: WOMAC pain fell 1.7 with nettle, 1.6 with control.
White AR, Randall C, Harding G, Paterson C. Patient consensus on mode of use of nettle sting for musculoskeletal pain. Complementary Therapies in Medicine. 2011;19(4):179–186. Consensus study. Strong sting, sting twice, once daily.
Fu HY, Chen SJ, Chen RF, Ding WH, Kuo-Huang LL, Huang RN. Identification of oxalic acid and tartaric acid as major persistent pain-inducing toxins in the stinging hairs of the nettle, Urtica thunbergiana. Annals of Botany. 2006;98(1):57–65. Chemistry plus animal behaviour. Different Urtica species.
Oliver F, Amon EU, Breathnach A, et al. Contact urticaria due to the common stinging nettle (Urtica dioica) — histological, ultrastructural and pharmacological studies. Clinical and Experimental Dermatology. 1991;16(1):1–7. Human, mechanistic.
Emmelin N, Feldberg W. The mechanism of the sting of the common nettle (Urtica urens). The Journal of Physiology. 1947;106(4):440–455. The original identification of histamine and acetylcholine in nettle-hair fluid.
Chrubasik S, Enderlein W, Bauer R, Grabner W. Evidence for antirheumatic effectiveness of Herba Urticae dioicae in acute arthritis: a pilot study. Phytomedicine. 1997;4(2):105–108. Human, open-label, n = 40, no placebo arm. Diclofenac 50 mg plus stewed nettle versus diclofenac 200 mg.
Johnson TA, Sohn J, Inman WD, Bjeldanes LF, Rayburn K. Lipophilic stinging nettle extracts possess potent anti-inflammatory activity, are not cytotoxic and may be superior to traditional tinctures for treating inflammatory disorders. Phytomedicine. 2013;20(2):143–147. Cell culture.
Überall MA, Küster MA, Müller-Schwefe PCG, Müller-Schwefe GHH. Safety, tolerability, and efficacy of Hox alpha, a dry extract from stinging nettle leaves versus OTC NSAIDs in osteoarthritis: a retrospective, propensity-matched 12-week analysis from the German Pain e-Registry (SIPHARO study). Journal of Pain Research. 2026;19:546747. Retrospective registry, 1,073 per arm. Not randomized; the authors call it exploratory and hypothesis-generating.
Rayburn K. Stinging nettle cream for osteoarthritis. Alternative Therapies in Health and Medicine. 2009;15(4):60–61. Short report.
Jacquet A, Girodet PO, Pariente A, Forest K, Mallet L, Moore N. Phytalgic, a food supplement, vs placebo in patients with osteoarthritis of the knee or hip: a randomised double-blind placebo-controlled clinical trial. Arthritis Research & Therapy. 2009;11(6):R192. Human RCT, combination product.
Christensen R, Bliddal H. Is Phytalgic a goldmine for osteoarthritis patients or is there something fishy about this nutraceutical? A summary of findings and risk-of-bias assessment. Arthritis Research & Therapy. 2010;12(1):105. Critical commentary on the trial above.
Chrubasik JE, Roufogalis BD, Wagner H, Chrubasik S. A comprehensive review on nettle effect and efficacy profiles, Part I: herba urticae. Phytomedicine. 2007;14(6):423–435. Review of the aerial parts, including the joint literature.

Live PubMed Searches

Connections

Back to Table of Contents