Herbs & Botanicals · Herbs · Daisy-family root and leaf, with low-certainty evidence for migraine and hay fever and a serious liver warning

Butterbur

Q: Does butterbur prevent migraine?

Possibly a little, but the evidence is low certainty. Two small trials of one special root extract found fewer attacks on the higher dose than on placebo (48% vs 26% fewer in the larger trial), while the lower dose did not beat placebo. Some authors are linked to the manufacturer. A 2006 review rated the evidence moderate for a higher dose; a 2020 review called it positive but limited with high risk of bias.

Q: Is butterbur safe for the liver?

Not reliably. Raw plant contains pyrrolizidine alkaloids that can poison the liver. Even products made to remove them have been linked to liver injury: a review of reports about one root extract found 48 spontaneous liver reports, nine severe and two needing a transplant, though the authors, who have ties to the maker, call the cases rare and often confounded by other drugs. NCCIH says PA-free products have seemed safe for up to 16 weeks, but rare liver injury has been reported even with those.

Q: What are pyrrolizidine alkaloids (PAs)?

PAs are natural plant chemicals. NCCIH says they can damage the liver and lungs and may cause cancer. One PA from a related butterbur species caused liver tumours in rats. In wild Swiss plants rhizomes held 5 to 90 parts per million and leaves 0.02 to 1.50. Regulated extracts try to remove them to trace amounts.

Q: Are PA-free butterbur products safe?

They are safer than raw plant, but not proven safe. NCCIH says PA-free products have seemed safe for up to 16 weeks, with rare liver injury reported even so. Products also vary: a lab test of 21 market supplements found PAs in seven, and a rat study of three PA-free products found one powder had no liver effect while two oily ones raised liver enzyme-gene activity.

Q: What did regulators say?

The UK MHRA advised consumers in 2012 not to take unlicensed butterbur products because of liver toxicity concerns; this is reported by the New Zealand regulator Medsafe, and the MHRA notice itself was not opened. A 2019 review says a German agency removed one product from the market in 2002 because of a change in its make-up. No Swiss action and no EU herbal monograph for Petasites hybridus were found.

Q: Does butterbur help hay fever?

Perhaps for one leaf extract. In manufacturer-linked trials it matched cetirizine and fexofenadine and beat placebo in the larger one. But a small independent placebo-controlled trial found no benefit, and a 2024 pooled review of single plants rates the placebo comparison as very low to low certainty.

Q: Can children take butterbur?

Small studies exist, but the evidence is poor and liver injury has been reported with the extract. One randomised trial in 58 children found a benefit over placebo only at the 6-month follow-up. Only use it under a paediatrician's advice.

Q: Is butterbur safe in pregnancy or breastfeeding?

Avoid it. NCCIH says products containing PAs should not be used in pregnancy or while breastfeeding, and that little is known about PA-free products in those situations.

Q: What are the common side effects?

Burping and a bitter taste were the most common. In a clinic series 28.3% had them for about 6 days, in a children's study 7.4% had undesired effects mostly burping, and in a hay fever surveillance study of 580 people 3.8% had mostly digestive complaints.

Petasites hybridus (L.) G.Gaertn., B.Mey. & Scherb. · also called Butterbur, common butterbur; the extract Petadolex (root) and Ze 339 (leaf) are brand-name products

Butterbur is a large-leaved plant of the daisy family that grows in wet ground in Europe and Asia. Extracts of its root (studied for migraine) and its leaf (studied for hay fever) have been tested in small trials. The plant naturally contains pyrrolizidine alkaloids (PAs), which can poison the liver and may cause cancer, so raw plant, tea and unprocessed products should be avoided. Even products made to remove PAs have been linked to rare but serious liver injury, and a US neurology guideline stopped recommending butterbur in 2015 for that reason, according to NCCIH. Talk to a doctor before any use.

General information, not medical advice. Last substantive update October 25, 2026 · Not medically reviewed. A named clinician has not signed off on this page. How this page is made

About Butterbur

Butterbur (Petasites hybridus) is a daisy-family plant whose root and leaf extracts have been tested for migraine prevention and hay fever. The evidence is small and rated low certainty at best. Two trials of one root extract found fewer migraine attacks on the higher dose than on placebo, and a leaf extract matched antihistamines in manufacturer-linked trials, though one independent trial found no benefit. The key safety message is liver injury: the plant contains pyrrolizidine alkaloids that can poison the liver, and even PA-reduced extracts have been linked to rare severe liver injury. The UK regulator warned against unlicensed products in 2012 and NCCIH says a US neurology guideline stopped recommending butterbur in 2015. Do not use raw plant or tea. Avoid it in pregnancy, breastfeeding and liver disease. Ask a doctor or pharmacist about amounts.

What it is

Butterbur (Petasites hybridus) is a large-leaved perennial of the daisy family (Asteraceae). It is an accepted species in the global plant database GBIF.

Medicines are made from the root or rhizome (used in the migraine trials) or from the leaf (used in the hay fever trials). These extracts are different products and should not be treated as the same.

^{15 33 23}

What makes it distinctive

The active ingredients are thought to be petasin and related compounds called petasins. In lab work they quiet pain nerve endings and reduce release of CGRP, a molecule linked to migraine. This has not been shown in people.

The plant also contains pyrrolizidine alkaloids (PAs), toxic to the liver. Rhizomes held far more PAs than leaves in wild Swiss plants. Manufacturers remove PAs to trace amounts, and a lab test of 21 market supplements found PAs in seven, so quality varies.

^{5 22 45 4}

Historical and current uses

A 2022 review lists traditional uses of butterbur species for migraine, asthma, hay fever, bronchitis, spastic cough, high blood pressure, stomach ulcers and period pain. Tradition is not proof.

In recent decades butterbur was tested for migraine prevention (root extract) and hay fever (leaf extract). No EU herbal monograph for these uses was found.

^{23 13}

What the research shows

For migraine, a 2006 review found only two randomised trials (both of one root extract), and a 2020 review called the evidence positive but limited with high risk of bias. For hay fever, reviews are encouraging for one leaf extract but note manufacturer funding, and a small independent trial found no effect.

Studies in children, asthma and skin allergy are small and mostly uncontrolled or null.

^{1 30 17 16 28}

At a glance

Traditional uses history, not proof

Traditional medicine in Europe and Asia since ancient times for migraine, asthma, hay fever, bronchitis, spastic cough, high blood pressure, stomach ulcers and period pain (a 2022 review)
Rhizomes used for inflammatory conditions (Fiebich 2005)

Studied for research

Preventing migraine in adults (root extract): low certainty
Migraine in children and teenagers: very low certainty
Hay fever (leaf extract): low certainty, with one null independent trial
Asthma, allergic skin reactions and year-round rhinitis: very low certainty
Inflammation and traditional uses beyond these: insufficient evidence

Main compounds

Petasin, isopetasin and related petasins: the compounds thought responsible for the anti-migraine and hay fever effects
Pyrrolizidine alkaloids such as senecionine and senkirkine: liver-toxic and carcinogenic in animals, removed (to trace amounts) from certified products
Other sesquiterpenes and phenolic compounds: over 200 sesquiterpenes are reported across the genus

Common forms

Special standardised root extract (Petadolex), used in migraine trials
Leaf CO2 extract (Ze 339), used in hay fever trials
Raw plant, tea and unprocessed root products: not recommended because of the PA content

Research links: NCCIH: Butterbur, usefulness and safety · Medsafe: Butterbur and liver toxicity (2012) · AAN/AHS 2012 guideline abstract (PubMed) · Hepatobiliary events with Petadolex, Anderson and Borlak 2019 · Research Library · Genes & Biomarkers Library

The plant and its history

Accepted scientific name: Petasites hybridus (L.) G.Gaertn., B.Mey. & Scherb.
Naming authority: (L.) G.Gaertn., B.Mey. & Scherb. per GBIF
Synonyms: Petasites major Cariot & St.-Lag.; Petasites ovalis Hill; Petasites reuterianus Jord.
Common names: Butterbur (English); Common butterbur (English)
Botanical family: Asteraceae (the daisy family, also called Compositae).
Classification: Plantae > Tracheophyta > Magnoliopsida > Asterales > Asteraceae > Petasites > Petasites hybridus
Native range: Not covered by the sources read beyond that the genus is used in Europe and Asia traditionally (Kulinowski 2022).
Habitat: Not covered by the sources read; Swiss wild populations were sampled (Wildi 1998).
What the plant looks like: Roots: Underground rhizomes and roots hold the highest alkaloid content (Wildi 1998).; Stems: Not described in the sources read.; Leaves: Leaves are a source of petasins with lower PA content than rhizomes (Wildi 1998).; Flowers: Not described in the sources read.; Fruit: Not described in the sources read.
Cultivation: Not covered by the sources read.
Harvest and processing: Extracts are made from root or leaf. A manufacturer-linked review says the extraction solvent for the root extract was changed in 1988 to better remove PAs; one leaf extract is a CO2 extract (Ze 339).
Look-alikes and adulterants: Name: Petasites japonicus (Japanese butterbur, fuki); Note: A different species whose young flower stalk contains the PA petasitenine, shown to cause liver tumours in rats (Hirono 1977). Results in this species cannot be assumed for P. hybridus.; Name: Other Petasites species; Note: Methods exist to tell Petasites species apart in products (Avula 2012).
Sustainability and conservation: Not covered by the sources read.

Traditional use and history history, not clinical proof

Tradition or era	Documented use	Source
Ancient to traditional European and Asian medicine	A 2022 review says butterbur species have been used since ancient times for migraine, asthma, hay fever, bronchitis, spastic cough, high blood pressure, ulcers and period pain.	p35427728
Traditional use of rhizomes	Fiebich 2005 says butterbur rhizomes have been used since ancient times for inflammatory diseases.	p15678367
1988 to 2002 (Germany)	A 2019 review says the extraction solvent for a root extract was changed in 1988 to better remove PAs and that in 2002 the German agency BfArM found the product no longer the same as the original and removed it from the market.	p31083451
2002 to 2005	Controlled trials of a leaf extract in hay fever (BMJ 2002, Arch Otolaryngol 2004) and of root extract in migraine (Neurology 2004).	p11799030
2012	UK MHRA consumer alert about unlicensed butterbur and the AAN/AHS guideline listing butterbur Level A.	s-medsafe-butterbur-2012
2015	NCCIH says the AAN stopped recommending butterbur for migraine because of safety concerns.	s-nccih-butterbur
Ancient Greek and Roman uses	Wikipedia-style claims about plague and fever use in ancient Greece and Rome were NOT verified in any source read and are not repeated.	none

Traditional use shows how a plant was used. It does not show that a supplement works or is safe today.

What is inside, and how it may work

What the plant contains and how those compounds act in cells and animals. Laboratory and animal mechanisms are not proven effects in people.

Main compounds

Compound	Family	Where it is concentrated	Activity studied	Level of evidence
Petasin PubChem 5281526	sesquiterpene ester (eremophilane)	Rhizome and leaf (3.3 to 11.4 mg/g in leaves; 7.4 to 15.3 mg/g in rhizomes in wild Swiss plants) Varies by product: 0.02 to 11.6 mg per dosage form in 15 supplements; absent from six.	Reduced CGRP release in rodent tissue (with isopetasin); the petasins are the proposed anti-migraine ingredients. Pure petasin did not block COX enzymes. ^{22 45 4 13}	Laboratory research
Isopetasin PubChem 5318627	sesquiterpene ester (eremophilane)	Rhizome Not stated beyond extract ratios (for example 0.1 to 9.4% in four lab extracts).	Activates then desensitises TRPA1 channels and reduces CGRP release in rodent tissue. ^{5 22 13}	Laboratory research
S-petasin PubChem 16219858	sulfur-containing sesquiterpene ester	Root and rhizome Not stated in the sources read.	Not covered in the abstracts read. ³⁴	Not assessed
Senecionine PubChem 5280906	pyrrolizidine alkaloid	Present in the plant; one of the PAs tested for in leaf extract Total PAs 5 to 90 ppm in rhizomes and 0.02 to 1.50 ppm in leaves of wild Swiss plants.	Toxic to the liver (class warning). ^{42 45}	Authoritative guidance
Senkirkine PubChem 5281752	pyrrolizidine alkaloid (otonecine type)	Tested for in Ze 339 leaf extract Not detected in tested batches of Ze 339 (limit 2 ppb).	Hepatotoxic class. ⁴²	Authoritative guidance
Petasitenine PubChem 5281741	pyrrolizidine alkaloid (otonecine type)	Young flower stalk of Petasites japonicus (a different species) Not stated.	Caused liver tumours in rats at 0.01% in drinking water. ¹⁹	Animal research

Mechanisms, receptors and pathways

Target or pathway	Gene or receptor	What was found	Level
TRPA1 pain channel	TRPA1	Isopetasin activates TRPA1 and then desensitises it, reducing CGRP release and migraine-like responses in rodents. ⁵	Laboratory research
TRPV1 pain channel and CGRP release	TRPV1; CGRP	Petasin and isopetasin reduced CGRP release provoked by TRPA1 and TRPV1 triggers in rat and mouse tissue, confirmed in channel-deleted mice. ²²	Laboratory research
COX-2 and prostaglandin E2	PTGS2 (COX-2)	Rhizome extracts blocked COX-2 and PGE2 release in rat microglia, independent of petasin content; pure petasin and isopetasin did not. ¹³	Laboratory research
Leukotriene biosynthesis (cell work)	Cysteinyl leukotrienes	Trial abstracts state butterbur shows in-vitro inhibition of cysteinyl leukotriene synthesis; the original studies were not read. ²⁵	Laboratory research
Antihistamine effect	Histamine H1	In people, butterbur did not reduce the histamine skin reaction while fexofenadine did, arguing against a direct antihistamine effect. ²¹	Human data
Nasal AMP challenge response	Not applicable	Butterbur reduced the nasal airflow fall after AMP challenge in grass-pollen and house-dust-mite allergy. ^{25 26}	Human data
Bile acid transport (liver)	Bile acid transporters (toxicogenomics)	Petasin-rich extracts at far above therapeutic doses were suggested to impair bile acid transport and lipid and protein metabolism in human liver cells. ²	Laboratory research
Liver enzymes (CYP)	CYP2B, CYP3A1 (rat)	Two oily PA-free products raised CYP2B more than 10-fold in rats. ¹⁸	Animal research

Absorption, metabolism and elimination

Absorption: Petasins showed high permeability across a gut-cell model and absorption in rat intestine (Disch 2018, manufacturer-employed authors).
Distribution: No human tissue distribution data were found.
Metabolism: High metabolism in gut cells and rat intestine; liver S9 fractions of rats and humans metabolised petasins faster than intestinal fractions.
Elimination: No data found.
Bioavailability: No human blood-level data were found in the sources read. A therapeutic peak blood level of about 60 ng/mL for petasins is an estimate used in a rat and cell study (Anderson 2009).

^{11 2}

Isolated compounds versus whole extracts

Whole extracts and pure compounds act differently: extracts blocked COX-2 but pure petasin and isopetasin did not. Petasin-rich extracts were the most toxic in human liver cells. Root, rhizome and leaf extracts differ in petasin and PA content, and two PA-free oily products gave different liver effects in rats than a powder. Results from one product should not be applied to another.

Nutritional composition

Butterbur is a medicinal plant, not a nutritional food. Young flower stalks of the related Petasites japonicus are eaten in parts of Asia, but they contain a PA shown to cause liver tumours in rats (Hirono 1977). No vitamin or mineral data were read.

13benefits and uses looked at

45studies and sources checked

Low certaintybest evidence rating found

8lab markers tracked in trials

1 mixed findings4 low certainty5 very low certainty3 not enough evidence

What was looked at

The 13 benefits and uses checked for Butterbur, each with how strong the evidence is. Tap one to open it.

Preventing migraine attacks: what the reviews concludeLow certainty
Adults: the two placebo-controlled trials of root extractLow certainty
Children and teenagers with migraineVery low certainty
Real-world use in a Brazilian clinicVery low certainty
How it might calm migraine pain pathways (lab and animal work only)Not enough evidence
Hay fever (allergic rhinitis): what the reviews concludeLow certainty
Hay fever: butterbur leaf extract compared with antihistaminesLow certainty
Hay fever: placebo-controlled trials with mixed resultsMixed findings
Nasal allergy challenge tests and year-round (perennial) rhinitisVery low certainty
Skin allergy reaction (wheal and flare)Very low certainty
AsthmaVery low certainty
General inflammation (lab work only)Not enough evidence
Other traditional uses (spasms, cough, blood pressure, stomach ulcers, period pain)Not enough evidence

Lab markers tracked in the studies

Migraine attacks per monthPeople with at least half as many attacks (responders)CGRP release from nerve endings (lab)Total nasal symptom scoreNasal airflow after allergy challenge (peak nasal inspiratory flow)Skin wheal and flare to histamine or allergenAirway inflammation markers in asthma (exhaled nitric oxide, eosinophils)Airway sensitivity (AMP challenge PC20)

What is Butterbur good for?

The main proposed uses and how strong the human evidence is for each. "Low" is the best rating any butterbur use reached here. "Very low", "mixed" and "insufficient" mean few, small, uncontrolled or conflicting studies, and none of these is a proven treatment for you. Migraine trials used a root extract and hay fever trials a leaf extract, so the results cannot be swapped. Ratings are this profile's own reading unless a review's own rating is quoted. Weigh every use against the liver-safety warning on the Safety tab.

Preventing migraine: what the reviews sayLow certaintyReviews are cautiously positive for a special root extract, based on only two small trials of one product. A 2012 US guideline once gave butterbur its top rating; NCCIH says that recommendation was dropped in 2015 over safety concerns, mainly the liver. Because of that, this is the use where the safety warnings matter most. Migraine in adults: the two placebo-controlled trialsLow certaintyIn the larger trial (245 adults, 4 months), attacks fell 48% on the higher dose and 26% on placebo, but the lower dose did not beat placebo. A second small trial (60 people) also favoured the extract. Both trials tested one root extract and some authors are linked to the maker. Migraine in children and teenagersVery low certaintyOne small randomised trial found a benefit over placebo only at a 6-month follow-up, and an open study with no placebo group reported fewer attacks. Reviewers call the evidence in children poor, and liver injury has been reported, so this needs a paediatrician. Migraine in routine clinic careVery low certaintyA 2026 record review of 120 adults reported fewer headache days after 12 weeks on a butterbur product said to be free of pyrrolizidine alkaloids. With no comparison group, it cannot show that the plant did the work. Calming pain pathways (lab work only)Not enough evidenceIn cells and rodents, butterbur compounds quiet pain nerve endings and reduce release of CGRP, a migraine-linked molecule. Nobody has shown this happens in people with migraine. Hay fever: what the reviews sayLow certaintyReviews are encouraging for one leaf extract (Ze 339) but note that the manufacturer paid for the three big trials. A 2024 pooled review of single plants rates evidence for plants against placebo as very low to low. Hay fever compared with antihistaminesLow certaintyIn two manufacturer-linked trials the leaf extract worked about as well as cetirizine or fexofenadine, and beat placebo in the larger three-arm trial. The cetirizine trial had no placebo group. Hay fever against placeboMixed findingsA 186-person trial found the leaf extract beat placebo, while a small independent crossover trial in Scotland found no benefit on nasal airflow or symptoms. Nasal allergy challenge tests and year-round rhinitisVery low certaintyTwo tiny crossover studies (20 and 16 people) found butterbur reduced the nasal response to a lab irritant and eased symptoms in house-dust-mite allergy. These are short surrogate tests. Allergic skin reactionsVery low certaintyIn a small crossover study butterbur did not reduce the skin reaction to histamine or allergen, while an antihistamine did. It may not help allergic skin disease. AsthmaVery low certaintyA tiny crossover trial found lower airway inflammation markers over one week in people already on inhaled steroids, and an uncontrolled open study found fewer attacks. Asthma needs medical care; do not replace a prescribed inhaler. General inflammation (lab work only)Not enough evidenceExtracts blocked an inflammation enzyme (COX-2) in rat cells, but pure petasin did not. No human trial tested this. Other traditional uses (cough, spasms, blood pressure, ulcers, period pain)Not enough evidenceThese are listed in a review of traditional use. No human trials were found, so they are not rated as proven.

Benefits, one by one

Each card is a research question, not a promise. Tap a goal to filter. Open a card to see the studies, doses and limits.

Preventing migraine attacks: what the reviews conclude

Low certainty

Reviews agree the evidence comes from very few trials, nearly all of one special root extract (Petadolex). They lean modestly positive, but the trials are small and some authors are linked to the manufacturer. A US neurology guideline once gave butterbur its top rating, then stopped recommending it in 2015 because of liver safety worries.

7 studiesTrials used standardised root extracts (Petadolex-type). Raw root and…

Studies, doses and limits

What the trials found

Agosti 2006 found only two randomised trials (60 and 233 people), both of Petadolex. It judged the evidence moderate for a higher than recommended dose and asked for more rigorous long-term safety data. Lopresti 2020 reviewed 19 herbal migraine trials and called the evidence for butterbur positive but limited, with high risk of bias in many studies. The 2012 American Academy of Neurology and American Headache Society guideline rated butterbur Level A (effective). NCCIH (Nov 2024) says the AAN stopped recommending butterbur in 2015 because of safety concerns; the AAN notice was not opened. For children, Orr 2014 judged the nutraceutical evidence overall to be poor.

Results

A modest cut in attack frequency in two small trials of one root extract. No independent large trial was found, and the safety record changed the guideline picture.

Effect sizes as reported

Largest trial (Lipton 2004): attacks fell 48% on the higher dose vs 26% on placebo (per-protocol). Smaller trial (Diener 2004 reanalysis): 3.4 to 1.8 attacks a month on butterbur vs 2.9 to 2.6 on placebo.

Why this rating

Only two small placebo-controlled trials of one proprietary extract, with manufacturer links among the authors, and a flawed original analysis in one of them. Agosti rated the evidence moderate for the higher dose; Lopresti rated it limited with high bias risk; the AAN once rated it Level A and later withdrew the recommendation. These do not fully agree, so low is the compiler's reading.

Form and dose studied

Trials used standardised root extracts (Petadolex-type). Raw root and unprocessed products are different and carry a poisoning risk (see the safety tab). Ask a doctor or pharmacist about amounts.

Who was studied

Adults with migraine, 2 to 6 attacks a month, in the main trials; some children's studies.

Limits

Few trials, one product, industry links, large placebo response, and no long-term safety follow-up in trials.

Why researchers looked

Migraine is common and disabling, and many people want an alternative to daily prescription preventers. Butterbur root has a long history as a remedy for headaches and spasms.

Traditional use

Reviews say butterbur species have been used since ancient times in Europe and Asia for migraine, asthma, hay fever, cough, spasms and other problems. That is traditional use, not proof.

Proposed mechanism Preliminary mechanism

In lab and animal work, butterbur compounds (petasin, isopetasin) calm pain-signalling nerve endings and reduce release of CGRP, a molecule involved in migraine. This has not been shown to be how it works in people with migraine.

A proposed mechanism does not show that the effect happens in people.

Raw or unprocessed butterbur contains pyrrolizidine alkaloids that can harm the liver. Even PA-reduced extracts have been linked to liver injury in case reports. See the safety tab.

^{1 30 20 37 29 10 35}

Adults: the two placebo-controlled trials of root extract

Low certainty

In the larger trial, 245 adults took a special root extract or placebo for 4 months. Attacks fell about 48% on the higher dose and 26% on placebo. The lower dose did not beat placebo. A second, small trial (60 people) also favoured the extract.

2 studiesSpecial standardised root extract (Petadolex). Do not assume other bu…

Studies, doses and limits

What the trials found

Lipton 2004 (n=245): adults with 2 to 6 attacks a month took the extract twice daily or placebo. Per-protocol, attacks fell 48% on the higher dose (p=0.0012 vs placebo), 36% on the lower dose (p=0.127) and 26% on placebo. At least a 50% fall in attacks happened in 68% on the higher dose and 49% on placebo (p<0.05). Diener 2004 reanalysed an earlier 60-person trial of Petadolex: attacks fell from 3.4 to 1.8 a month on butterbur and from 2.9 to 2.6 on placebo; responders 45% vs 15%. The authors of that paper say the original analysis had major shortcomings.

Results

Higher dose beat placebo in both trials; the lower dose did not in the larger one.

Effect sizes as reported

Per-protocol: 48% vs 26% fewer attacks (Lipton). Responder rate 68% vs 49% (Lipton) and 45% vs 15% (Diener).

Why this rating

Two small trials, one with a re-done analysis, one product, manufacturer links among authors, and per-protocol reporting in the abstract.

Form and dose studied

Special standardised root extract (Petadolex). Do not assume other butterbur products behave the same.

Who was studied

Adults 18 to 65 with migraine.

Limits

Per-protocol analysis; placebo group improved 26%; 4 months or less; no follow-up after stopping.

Why researchers looked

These were the trials that put butterbur on the map for migraine prevention.

Proposed mechanism Preliminary mechanism

Lab work suggests the active petasins quiet TRPA1 and TRPV1 pain channels and CGRP release. This is not shown in migraine patients.

A proposed mechanism does not show that the effect happens in people.

Mild stomach events, mostly burping, were the most common side effect. Liver safety could not be settled in trials this small; see the safety tab.

^{29 10}

Children and teenagers with migraine

Very low certainty

One small randomised trial (58 children) and one open study (108 children and teens) tested root extract. The randomised trial found a benefit over placebo only at the 6-month follow-up. The open study had no placebo group. Reviews call the evidence in children poor.

3 studiesRoot extract, with the amount scaled by age in the open study. Ask a …

Studies, doses and limits

What the trials found

Oelkers-Ax 2008: primary-school children took the root extract (n=19), had music therapy (n=20) or placebo (n=19) for 12 weeks. Right after treatment only music therapy beat placebo (p=0.005); at the 6-month follow-up both music therapy (p=0.018) and butterbur (p=0.044) did. All groups improved even before treatment. Pothmann 2005: 108 children and teens aged 6 to 17 in an open trial; 77% had at least 50% fewer attacks and attacks fell 63% on average; side effects 7.4%, mainly burping. Orr 2014 rated the quality of nutraceutical evidence in pediatric migraine as poor.

Results

Promising but weak: small, mostly uncontrolled, large placebo response in children.

Effect sizes as reported

Open study: 63% fewer attacks, 77% responders. Randomised trial: no numbers beyond p-values in the abstract.

Why this rating

One small explorative randomised trial plus an uncontrolled study; the reviewers call the evidence poor.

Form and dose studied

Root extract, with the amount scaled by age in the open study. Ask a paediatrician before any use.

Who was studied

Children aged 6 to 17.

Limits

No placebo in the larger study, completers-only analysis in the randomised trial, and no long-term liver follow-up in children.

Why researchers looked

Children have few safe preventers, so families look for alternatives.

Traditional use

No children's tradition is documented in the sources read.

Proposed mechanism Preliminary mechanism

Same lab-only mechanisms as in adults; not studied in children.

A proposed mechanism does not show that the effect happens in people.

Liver injury has been reported with butterbur extracts; do not give to a child without a paediatrician's advice.

^{36 38 37}

Real-world use in a Brazilian clinic

Very low certainty

A 2026 clinic record study of 120 adults found fewer headache days after 12 weeks on a butterbur product said to be free of pyrrolizidine alkaloids. With no comparison group it cannot show the plant did the work.

1 studiesOne PA-free branded product (Petamig); dose and plant part not given …

Studies, doses and limits

What the trials found

Silva-Neto 2026 (retrospective, uncontrolled, one centre): 72 people with episodic and 48 with chronic migraine. Headache days per month fell from 6.0 to 2.6 (episodic) and from 26.3 to 12.7 (chronic). 59.2% had at least a 50% cut. Side effects occurred in 28.3% (bitter taste or burping, lasting about 6 days). The author states the product is PA-free; this paper did not test that.

Results

Fewer headache days, but no control group.

Effect sizes as reported

Mean headache days: 6.0 to 2.6 and 26.3 to 12.7 (p<0.0001 before vs after).

Why this rating

Retrospective, uncontrolled, single centre; natural change and placebo effects cannot be ruled out.

Form and dose studied

One PA-free branded product (Petamig); dose and plant part not given in the abstract.

Who was studied

Adults 18 to 60 with episodic or chronic migraine.

Limits

No control group; PA-free status stated, not tested; short follow-up.

Why researchers looked

To see how a PA-free product performs in routine care.

Proposed mechanism Preliminary mechanism

Same lab-only mechanisms as above.

A proposed mechanism does not show that the effect happens in people.

Bitter taste and burping were common. No liver testing results are given in the abstract.

⁴³

How it might calm migraine pain pathways (lab and animal work only)

Not enough evidence

In cells and rodents, butterbur compounds first stimulate and then quiet pain nerve endings and reduce CGRP release. This is a lab idea, not a proven benefit in people.

4 studiesPure compounds and extracts at lab concentrations that cannot be comp…

Studies, doses and limits

What the trials found

No human trial measured these pathways.

Results

Lab findings only.

Why this rating

Cell and animal work only; the extract content did not predict the COX-2 effect.

Form and dose studied

Pure compounds and extracts at lab concentrations that cannot be compared with doses people take.

Who was studied

Cells, mice and rats.

Limits

Lab only; some funded by the manufacturer.

Why researchers looked

To explain why a plant extract might prevent migraine.

Proposed mechanism Laboratory research

Isopetasin activates and then desensitises TRPA1 channels on pain nerves; petasin and isopetasin reduced CGRP release triggered through TRPA1 and TRPV1 in rodent tissue. Whole extracts also blocked the COX-2 inflammation enzyme in rat cells, but pure petasin did not.

A proposed mechanism does not show that the effect happens in people.

Not applicable.

^{5 22 13 6}

Hay fever (allergic rhinitis): what the reviews conclude

Low certainty

Reviews lean positive for one leaf extract (Ze 339) in hay fever, but most large trials were paid for by its maker and the newest pooled analysis rates the plant evidence as very low to low certainty.

4 studiesThe positive trials used butterbur leaf CO2 extract (Ze 339), not roo…

Studies, doses and limits

What the trials found

Guo 2007 (16 trials of 10 herbs): six butterbur trials suggested it beat placebo or matched non-sedating antihistamines in intermittent hay fever; the authors asked for independent replication because the manufacturer paid for the three large trials. Lim 2024 pooled 29 trials of single plants (butterbur the most studied, five studies): compared with placebo, plants may improve total nasal symptoms (SMD -0.31, 95% CI -0.59 to -0.02; 249 people), very-low to low certainty, and showed no clear difference from antihistamines (moderate certainty). These figures cover all plants, not butterbur alone. Braunwalder 2025 (14 studies) found butterbur the most studied plant, with benefits on immune markers, symptoms and nasal airflow; about half the studies were of sufficient quality. Merk 2025, written by manufacturer employees, calls Ze 339 a promising alternative to antihistamines.

Results

Moderately encouraging for one leaf extract over about 2 weeks; independent confirmation is thin.

Effect sizes as reported

Plants combined versus placebo: SMD -0.31 (95% CI -0.59 to -0.02); versus antihistamine: SMD -0.14 (95% CI -0.46 to 0.18).

Why this rating

Lim 2024 rates the plant-versus-placebo evidence very low to low; Guo 2007 calls it encouraging but manufacturer-funded; Braunwalder 2025 is positive but qualitative. Low is the compiler's reading.

Form and dose studied

The positive trials used butterbur leaf CO2 extract (Ze 339), not root extract. The root extract used for migraine has a different make-up. Ask a doctor or pharmacist about amounts.

Who was studied

Adults with seasonal or intermittent hay fever; trials mostly 2 weeks.

Limits

Short trials, manufacturer funding, one extract, and a null placebo-controlled trial by an independent group.

Why researchers looked

Hay fever is common and antihistamines can cause drowsiness, so people look for alternatives.

Traditional use

Reviews list hay fever and asthma among the traditional uses of butterbur species.

Proposed mechanism Preliminary mechanism

Lab work suggests petasins may reduce leukotriene and histamine-related signalling in cells. One human skin test found no antihistamine-type effect (see the skin test entry).

A proposed mechanism does not show that the effect happens in people.

Raw butterbur plant contains liver-toxic alkaloids; leaf extracts tested in the trials were tested for absence of them. See the safety tab.

^{17 28 7 32}

Hay fever: butterbur leaf extract compared with antihistamines

Low certainty

Two manufacturer-linked trials found the leaf extract worked about as well as cetirizine or fexofenadine. In the larger one it also beat placebo.

3 studiesZe 339 leaf CO2 extract. Do not assume other butterbur products match.

Studies, doses and limits

What the trials found

Schapowal 2002 (n=125): leaf extract (Ze 339) four times daily vs cetirizine for 2 weeks; quality of life and global scores were similar. There was no placebo group. About two thirds of side effects on cetirizine were drowsiness or fatigue. Schapowal 2005 (n=330): Ze 339 and fexofenadine both beat placebo (p<0.001) and did not differ from each other (p=0.37). Käufeler 2006 (open, 580 people): symptoms improved in 90% over 2 weeks and side effects were 3.8%; those who also took allergy medicine did no better than butterbur alone.

Results

Similar to antihistamines, better than placebo in one three-arm trial.

Effect sizes as reported

Not reported as numbers in the abstracts beyond p-values and global scores.

Why this rating

Two randomised trials, both manufacturer-linked, short, plus an open surveillance study.

Form and dose studied

Ze 339 leaf CO2 extract. Do not assume other butterbur products match.

Who was studied

Adults with hay fever.

Limits

No placebo in the cetirizine trial; manufacturer funding; 2 weeks only.

Why researchers looked

To see whether a herbal option can match antihistamines without drowsiness.

Proposed mechanism Preliminary mechanism

Not shown. The petasins may act on leukotrienes in cells; one skin test found no antihistamine effect.

A proposed mechanism does not show that the effect happens in people.

Mild stomach complaints were typical.

^{39 41 24}

Hay fever: placebo-controlled trials with mixed results

Mixed findings

A dose-response trial found the leaf extract beat placebo. A small independent crossover trial in Scotland found no benefit for symptoms or nasal airflow.

2 studiesLeaf extract in the positive trial; product not named in the null tri…

Studies, doses and limits

What the trials found

Schapowal 2004 (n=186): Ze 339 high dose, low dose or placebo for 2 weeks; both butterbur groups improved more than placebo, with more benefit on the higher dose; side effects matched placebo. Gray 2004 (n=35 crossover, grass pollen season): butterbur 2 weeks vs placebo had no significant effect on nasal airflow, symptoms or quality of life (total nasal symptom score 3.4 and 3.5 vs 3.7 and 3.8, out of 12). The product in Gray 2004 was not named in the abstract.

Results

One positive and one null placebo-controlled trial.

Effect sizes as reported

Gray 2004: no significant difference. Schapowal 2004: significant but numbers not in abstract.

Why this rating

The two placebo-controlled trials disagree; products and designs differ.

Form and dose studied

Leaf extract in the positive trial; product not named in the null trial.

Who was studied

Adults with intermittent hay fever.

Limits

Small crossover trial; 2 weeks; different products.

Why researchers looked

To test butterbur against placebo, the fairest comparison.

Proposed mechanism Preliminary mechanism

Not shown.

A proposed mechanism does not show that the effect happens in people.

No difference in side effects from placebo in the larger trial.

^{40 16}

Nasal allergy challenge tests and year-round (perennial) rhinitis

Very low certainty

Three very small crossover studies found that butterbur reduced the nasal response to a lab irritant (AMP) and eased symptoms in house-dust-mite allergy. They are lab-style tests with 16 to 20 people.

2 studiesButterbur (product not named in the abstracts).

Studies, doses and limits

What the trials found

Lee 2003 (n=20, grass pollen hay fever, 2 weeks): recovery after nasal AMP challenge was smaller on butterbur (584 vs 1438; mean difference 854, 95% CI 95 to 1614) and the maximum fall in airflow was 30% vs 43%. Lee 2004 (n=16, house dust mite year-round rhinitis, 1 week): maximum airflow fall 34% on butterbur vs 46% on placebo, and nasal symptom score 1.8 vs 2.8, similar to fexofenadine.

Results

Positive on challenge tests and symptoms in tiny short studies.

Effect sizes as reported

See above (differences with 95% CI for Lee 2003).

Why this rating

Very small crossover studies of 1 to 2 weeks with surrogate outcomes; products not named.

Form and dose studied

Butterbur (product not named in the abstracts).

Who was studied

Adults with allergic rhinitis.

Limits

Tiny samples; surrogate lab outcomes; short.

Why researchers looked

Objective tests avoid the guesswork of symptom scores.

Proposed mechanism Preliminary mechanism

Possibly reduced leukotriene-type inflammation; not shown directly.

A proposed mechanism does not show that the effect happens in people.

None highlighted in the abstracts.

^{25 26}

Skin allergy reaction (wheal and flare)

Very low certainty

In a small crossover study, butterbur did not reduce the skin reaction to histamine or allergen, while an antihistamine did. The authors conclude it may not help allergic skin disease.

1 studiesButterbur (product not named).

Studies, doses and limits

What the trials found

Jackson 2004 (20 atopic adults completed): after 1 week, butterbur did not change the skin wheal or flare to histamine or allergen compared with placebo; fexofenadine did.

Results

No effect on the skin test.

Effect sizes as reported

Allergen wheal 50.4 mm2 on butterbur vs 65.4 mm2 on placebo, not significant (fexofenadine 31.1 mm2, significant).

Why this rating

One tiny short crossover study; null result.

Form and dose studied

Butterbur (product not named).

Who was studied

Atopic adults.

Limits

Small, short; skin test not a symptom outcome.

Why researchers looked

To test whether butterbur acts like an antihistamine.

Proposed mechanism Preliminary mechanism

Argues against a direct antihistamine effect in people.

A proposed mechanism does not show that the effect happens in people.

None.

²¹

Asthma

Very low certainty

Two small studies suggest butterbur may lower airway inflammation markers or reduce asthma medicine use, but there is no strong trial. Asthma needs medical care; do not replace a prescribed inhaler.

2 studies25 mg-per-dose butterbur (product not named) and Petadolex root extra…

Studies, doses and limits

What the trials found

Lee 2004 (n=16, atopic asthma on inhaled corticosteroids, 1 week, crossover): butterbur improved the AMP airway challenge by 0.7 doubling doses, lowered exhaled nitric oxide (-1.2 vs +0.5 ppb) and eosinophil markers versus placebo. Danesch 2004 (open trial, 64 adults and 16 children, 2 months, root extract Petadolex): attacks and symptoms fell, peak flow and FEV1 rose, and over 40% of medicine users reduced their medicines. No control group.

Results

Inflammation markers improved in a tiny study; open trial improved with no control.

Effect sizes as reported

PC20: +0.6 vs -0.1 doubling dilutions (butterbur vs placebo).

Why this rating

One tiny crossover trial on lab markers and one uncontrolled open trial.

Form and dose studied

25 mg-per-dose butterbur (product not named) and Petadolex root extract in the two studies.

Who was studied

Adults with allergic asthma; some children.

Limits

Tiny, short, no asthma-attack outcomes in the randomised study.

Why researchers looked

Butterbur has been traditionally used for asthma and spastic cough.

Traditional use

Reviews list asthma and spastic cough among traditional uses of butterbur species.

Proposed mechanism Laboratory research

Lab work suggests anti-leukotriene action in cells; this is not proven in people.

A proposed mechanism does not show that the effect happens in people.

Do not stop asthma medicine without a doctor.

^{27 9}

General inflammation (lab work only)

Not enough evidence

In cell tests, butterbur extracts block the COX-2 inflammation enzyme. Pure petasin did not, so the active ingredient is unclear. No human trial tested this.

1 studiesLab extracts.

Studies, doses and limits

Results

Lab only.

Why this rating

Cell tests only.

Form and dose studied

Lab extracts.

Who was studied

Rat cells.

Limits

No human data.

Why researchers looked

Butterbur rhizome has long been used for inflammatory complaints.

Traditional use

Fiebich 2005 notes use since ancient times for inflammatory diseases.

Proposed mechanism Laboratory research

Four rhizome extracts inhibited COX-2 and PGE2 release in rat brain immune cells independent of petasin content; pure petasin and isopetasin did not.

A proposed mechanism does not show that the effect happens in people.

Not applicable.

¹³

Other traditional uses (spasms, cough, blood pressure, stomach ulcers, period pain)

Not enough evidence

Butterbur species have traditional uses in Europe and Asia for several conditions. A 2022 review lists them. No human trial for these uses was read, so they are not rated as proven.

1 studiesNot applicable.

Studies, doses and limits

Why this rating

No human trials found.

Limits

Tradition is not proof; raw plant carries liver-poison risk.

Why researchers looked

Historic use.

Traditional use

Kulinowski 2022 lists respiratory (asthma, hay fever, bronchitis, spastic cough), cardiovascular (high blood pressure), stomach (ulcers) and genitourinary (period pain) uses.

Proposed mechanism Pharmacological theory

Not established.

A proposed mechanism does not show that the effect happens in people.

Do not use raw butterbur plant or unprocessed root; see the safety tab.

²³

Connections to the rest of health

How Butterbur connects to the rest of our health knowledge graph. Every link says what kind of connection it is. A safety link (for example to the liver) is a warning, not a benefit. A study that merely mentions a topic never creates a benefit link.

Explore every connection: goals, body systems, organs, symptoms, genes

Kind of connection Connected to Search

A. Wellness goals

Connected to	Kind of connection	Evidence status	Context	Studies
Brain Health	Potential benefit	Low certainty	Preventing migraine attacks: what the reviews conclude: Reviews agree the evidence comes from very few trials, nearly all of one special root extract (Petadolex). They lean modestly positive, but the trials are small and some authors are linked to the manufacturer. A US neurology guideline once gave butterbur its top rating, then stopped recommending it in 2015 because of liver safety worries. Children and teenagers with migraine: One small randomised trial (58 children) and one open study (108 children and teens) tested root extract. The randomised trial found a benefit over placebo only at the 6-month follow-up. The open study had no placebo group. Reviews call the evidence in children poor.	^{1 30 20 37 29 10 35 36 38 43}
Brain Health	Demonstrated clinical effect	Low certainty	Adults: the two placebo-controlled trials of root extract: In the larger trial, 245 adults took a special root extract or placebo for 4 months. Attacks fell about 48% on the higher dose and 26% on placebo. The lower dose did not beat placebo. A second, small trial (60 people) also favoured the extract.	^{29 10}
Brain Health	Preliminary mechanism	Not enough evidence	How it might calm migraine pain pathways (lab and animal work only): In cells and rodents, butterbur compounds first stimulate and then quiet pain nerve endings and reduce CGRP release. This is a lab idea, not a proven benefit in people.	^{5 22 13 6}
Immune Support	Potential benefit	Low certainty	Hay fever (allergic rhinitis): what the reviews conclude: Reviews lean positive for one leaf extract (Ze 339) in hay fever, but most large trials were paid for by its maker and the newest pooled analysis rates the plant evidence as very low to low certainty. Hay fever: butterbur leaf extract compared with antihistamines: Two manufacturer-linked trials found the leaf extract worked about as well as cetirizine or fexofenadine. In the larger one it also beat placebo.	^{17 28 7 32 39 41 24 40 16 25 26 21 27 9}
Inflammation Balance	Potential benefit	Low certainty	Hay fever (allergic rhinitis): what the reviews conclude: Reviews lean positive for one leaf extract (Ze 339) in hay fever, but most large trials were paid for by its maker and the newest pooled analysis rates the plant evidence as very low to low certainty. Asthma: Two small studies suggest butterbur may lower airway inflammation markers or reduce asthma medicine use, but there is no strong trial. Asthma needs medical care; do not replace a prescribed inhaler.	^{17 28 7 32 27 9}
Inflammation Balance	Preliminary mechanism	Not enough evidence	General inflammation (lab work only): In cell tests, butterbur extracts block the COX-2 inflammation enzyme. Pure petasin did not, so the active ingredient is unclear. No human trial tested this.	¹³
Digestive Support	Traditional use	Not enough evidence	Other traditional uses (spasms, cough, blood pressure, stomach ulcers, period pain): Butterbur species have traditional uses in Europe and Asia for several conditions. A 2022 review lists them. No human trial for these uses was read, so they are not rated as proven.	²³
Blood Pressure Support	Traditional use	Not enough evidence	Other traditional uses (spasms, cough, blood pressure, stomach ulcers, period pain): Butterbur species have traditional uses in Europe and Asia for several conditions. A 2022 review lists them. No human trial for these uses was read, so they are not rated as proven.	²³

B. Body systems

Connected to	Kind of connection	Evidence status	Context	Studies
Nervous system (migraine)	Potential benefit	Low certainty	Fewer attacks in two small trials; safety concerns led the AAN to stop recommending it. Preparation or compound: Butterbur root extract (Petadolex-type)	^{29 10 1}
Respiratory system (hay fever, asthma)	Potential benefit	Low certainty	Similar to antihistamines in two trials; independent placebo trial null. Preparation or compound: Butterbur leaf extract (Ze 339); other butterbur	^{17 16}
Nervous System	Biological association	Low certainty	Studied for: Preventing migraine attacks: what the reviews conclude Studied for: Adults: the two placebo-controlled trials of root extract	^{1 30 20 37 29 10 35 36 38 43 5 22 13 6}
Respiratory System	Biological association	Low certainty	Studied for: Hay fever (allergic rhinitis): what the reviews conclude Studied for: Hay fever: butterbur leaf extract compared with antihistamines	^{17 28 7 32 39 41 24 40 16 25 26 27 9 23}
Immune System	Biological association	Low certainty	Studied for: Hay fever (allergic rhinitis): what the reviews conclude Studied for: Hay fever: butterbur leaf extract compared with antihistamines	^{17 28 7 32 39 41 24 21 13}
Integumentary System	Biological association	Very low certainty	Studied for: Skin allergy reaction (wheal and flare)	²¹
Digestive System	Biological association	Not enough evidence	Studied for: Other traditional uses (spasms, cough, blood pressure, stomach ulcers, period pain)	²³
Reproductive System	Biological association	Not enough evidence	Studied for: Other traditional uses (spasms, cough, blood pressure, stomach ulcers, period pain)	²³

C. Body parts and organs

Connected to	Kind of connection	Evidence status	Context	Studies
Liver	Adverse effect or risk	Low certainty	Liver injury reported, with transplants; PAs are hepatotoxic. Preparation or compound: Raw plant (PAs); PA-reduced root extract	^{3 35 45}

F. Functional actions

Connected to	Kind of connection	Evidence status	Context	Studies
Anti-inflammatory	Preliminary mechanism	Not enough evidence	COX-2 block in rat cells, independent of petasin content. Preparation or compound: Butterbur rhizome extracts	¹³

I. Biomarkers

Connected to	Kind of connection	Evidence status	Context	Studies
ALT (liver enzyme)	Adverse effect or risk	Low certainty	Severe cases had ALT 3 to 153 times the upper limit of normal. Preparation or compound: Petadolex in reported cases	³
AST (liver enzyme)	Adverse effect or risk	Low certainty	Severe cases had AST 2 to 104 times the upper limit of normal. Preparation or compound: Petadolex in reported cases	³
Migraine attacks per month	Biological association	Low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	^{29 10 38 36}
People with at least half as many attacks (responders)	Biological association	Low certainty	Measured in trials: raised. A lab change is not the same as a health benefit.	^{29 10 43}
CGRP release from nerve endings (lab)	Biological association	Not enough evidence	Measured in trials: lowered. A lab change is not the same as a health benefit.	^{22 5}
Total nasal symptom score	Biological association	Low certainty	Measured in trials: mixed. A lab change is not the same as a health benefit.	^{16 26 28}
Nasal airflow after allergy challenge (peak nasal inspiratory flow)	Biological association	Very low certainty	Measured in trials: mixed. A lab change is not the same as a health benefit.	^{25 16}
Skin wheal and flare to histamine or allergen	Biological association	Very low certainty	Measured in trials: no-change. A lab change is not the same as a health benefit.	²¹
Airway inflammation markers in asthma (exhaled nitric oxide, eosinophils)	Biological association	Very low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	²⁷
Airway sensitivity (AMP challenge PC20)	Biological association	Very low certainty	Measured in trials: raised. A lab change is not the same as a health benefit.	²⁷

J. Genes, receptors and enzymes

Connected to	Kind of connection	Evidence status	Context	Studies
TRPA1	Preliminary mechanism	In vitro	Isopetasin desensitises this pain channel in cells and rodents.	⁵
TRPV1	Preliminary mechanism	In vitro	Petasins reduce CGRP release triggered via TRPV1 in rodent tissue.	²²
PTGS2 (COX-2)	Preliminary mechanism	In vitro	Extracts block COX-2 in rat cells.	¹³
CYP2B and CYP3A	Preliminary mechanism	Animal	Raised in rats by two oily PA-free products.	¹⁸

K. Biological pathways

Connected to	Kind of connection	Evidence status	Context	Studies
CGRP signalling	Preliminary mechanism	In vitro	Proposed target for migraine prevention; lab only for butterbur.	^{22 5}
Bile acid transport	Preliminary mechanism	In vitro	Possible route of liver injury suggested by gene-activity work at high concentrations.	²

L. Active compounds

Connected to	Kind of connection	Evidence status	Context	Studies
Petasin	Biological association	In vitro	Reduced CGRP release in rodent tissue (with isopetasin); the petasins are the proposed anti-migraine ingredients. Pure petasin did not block COX enzymes. Preparation or compound: Rhizome and leaf (3.3 to 11.4 mg/g in leaves; 7.4 to 15.3 mg/g in rhizomes in wild Swiss plants)	^{22 45 4 13}
Isopetasin	Biological association	In vitro	Activates then desensitises TRPA1 channels and reduces CGRP release in rodent tissue. Preparation or compound: Rhizome	^{5 22 13}
S-petasin	Biological association	Not assessed	Not covered in the abstracts read. Preparation or compound: Root and rhizome	³⁴
Senecionine	Biological association	Authoritative guidance	Toxic to the liver (class warning). Preparation or compound: Present in the plant; one of the PAs tested for in leaf extract	^{42 45}
Senkirkine	Biological association	Authoritative guidance	Hepatotoxic class. Preparation or compound: Tested for in Ze 339 leaf extract	⁴²
Petasitenine	Biological association	Animal	Caused liver tumours in rats at 0.01% in drinking water. Preparation or compound: Young flower stalk of Petasites japonicus (a different species)	¹⁹

Each connection is stored with the ingredient, the connected record, the kind of connection, the evidence status, the supporting studies, the preparation or compound involved, and a review date (last reviewed October 25, 2026).

Evidence at a glance: what changed in 8 lab markers

Tap a number to show only those markers. Counts are markers, not people.

Other3212

The evidence

The research centre for Butterbur. Every record here was checked against PubMed or the publisher's page; we never list a study we have not verified.

Evidence by outcome

Preventing migraine attacks: what the reviews conclude 7 studies · 305 people in trials Low certainty Adults: the two placebo-controlled trials of root extract 2 studies · 305 people in trials Low certainty Children and teenagers with migraine 3 studies · 58 people in trials Very low certainty Real-world use in a Brazilian clinic 1 study Very low certainty How it might calm migraine pain pathways (lab and animal work only) 4 studies Not enough evidence Hay fever (allergic rhinitis): what the reviews conclude 4 studies Low certainty Hay fever: butterbur leaf extract compared with antihistamines 3 studies · 455 people in trials Low certainty Hay fever: placebo-controlled trials with mixed results 2 studies · 221 people in trials Mixed findings Nasal allergy challenge tests and year-round (perennial) rhinitis 2 studies · 36 people in trials Very low certainty Skin allergy reaction (wheal and flare) 1 study · 20 people in trials Very low certainty Asthma 2 studies · 16 people in trials Very low certainty General inflammation (lab work only) 1 study Not enough evidence Other traditional uses (spasms, cough, blood pressure, stomach ulcers, period pain) 1 study Not enough evidence

Bars show how many studies we found for each outcome, not how well it works. People are added up across trials only (reviews are not added, to avoid double counting).

How we rate evidence

High certainty

Several well-run trials or reviews agree; further research is unlikely to change the picture much.

Moderate certainty

Good trials point the same way, but some doubt remains.

Low certainty

Small, short or inconsistent trials. The finding could change with better research.

Very low certainty

Very little or very weak human data. Treat as a possibility only.

Mixed findings

Trials or reviews disagree with each other.

Not enough evidence

Not enough human research to say either way.

Evidence against

Good research found no benefit, or found harm.

We take the reviewers' own risk-of-bias and GRADE statements where they exist, and otherwise judge from study size, design, consistency, directness and whether a product matches what is sold. Many publications do not make evidence high. Method note: We searched PubMed, read the NCCIH page and a Medsafe regulatory note through a summary tool, and recorded each study's design, people, product and result as reported. Only abstracts were read for most studies. The MHRA, BfArM and AAN notices, the EMA pyrrolizidine alkaloid statement, LiverTox and LactMed could not be opened and are marked not verified. Certainty ratings are the compiler's own reading unless a review's own rating is quoted. Several safety and efficacy papers are by authors linked to the manufacturers; this is flagged. No study is cited that we could not verify.

Study library: every study we checked 45

Design Research type Plant part Year from Search

Study	Design	People	Product, dose, length	Result
Agosti 2006 Effectiveness of Petasites hybridus preparations in the prophylaxis of migraine: a systematic review Phytomedicine · PMID 16987643 · DOI	Systematic review	293 Two RCTs of Petadolex in migraine (60 and 233 patients)	Petadolex Petasites root extract root higher dose (reported by the reviewers as 150 mg a day) vs lower dose (100 mg a day) vs placebo 3 to 4 months · vs placebo	The higher dose cut attacks more than the lower dose and placebo. The reviewers rated this moderate evidence for a higher than recommended dose and asked for rigorous long-term safety studies before recommending it as an alternative. Not pooled (results too different) (not applicable) Methods, limits and funding Outcomes: Attack frequency; responders Limitations: Only two trials, both of the same product. Funding and conflicts: Not stated in the abstract Read: abstract
Anderson 2009 Toxicogenomics applied to cultures of human hepatocytes enabled an identification of novel petasites hybridus extracts for the treatment of migraine with improved hepatobiliary safety Toxicol Sci · PMID 19770483 · DOI	In vivo/in vitro animal	Rats (28-day and 6-month studies) and human liver cell cultures	Several Petasites hybridus extracts, including petasin-rich ones root about 200 times the therapeutic dose in rats 28 days and 6 months	At around 200 times the therapeutic dose rats had raised liver enzymes and bilirubin; by the end of 6 months liver function returned to normal. In human liver cells only petasin-rich extracts at far above therapeutic levels were toxic. One extract was characterised with no toxicity signal. Lab and animal findings only (not applicable) Methods, limits and funding Outcomes: Liver enzymes, bilirubin, gene activity Limitations: Animal and cell work. The authors say rat studies did not reliably predict human liver toxicity. Funding and conflicts: Manufacturer Weber and Weber funded authors' earlier studies (2008 to 2010) per Anderson 2019 Read: abstract
Anderson 2019 Hepatobiliary Events in Migraine Therapy with Herbs-The Case of Petadolex, A Petasites Hybridus Extract J Clin Med · PMID 31083451 · DOI	Case series	Spontaneous reports of liver injury linked to Petadolex (48 spontaneous reports; 12 more cases from trials among 397 patients, none serious)	Petadolex butterbur root extract root 50 to 225 mg a day in reports 4 to 730 days in reports	Most patients had mild liver blood test changes; nine had severe injury (the full text table lists ten), with onset around 103 days. Two needed liver transplants. All resolved after stopping. Causality: 3 likely, 13 possible, 8 unlikely, 24 unclassifiable. 22 patients also took other drugs that can harm the liver. The authors call the cases rare and idiosyncratic. No rate is given. Not applicable (not applicable) Methods, limits and funding Outcomes: Liver injury (herbal-induced liver injury) reports Limitations: Spontaneous reports cannot give a frequency. Authors consult for the manufacturer. The full text was only partly read. Funding and conflicts: Manufacturer funded earlier studies by the authors (2008 to 2010) Read: full-text
Avula 2012 Simultaneous determination of sesquiterpenes and pyrrolizidine alkaloids from the rhizomes of Petasites hybridus (L.) G.M. et Sch. and dietary supplements using UPLC-UV and HPLC-TOF-MS methods J Pharm Biomed Anal · PMID 22809670 · DOI	Lab analysis	21 dietary supplements claiming to contain Petasites hybridus, plus plant rhizome	Market supplements and rhizome rhizome	Petasin ranged from 0.02 to 11.6 mg per dosage form in 15 supplements and was not detected in six. Pyrrolizidine alkaloids were detected in seven supplements; seven others had petasin within the label claim and no PAs. Content varies widely (not applicable) Methods, limits and funding Outcomes: Petasin and pyrrolizidine alkaloid content Limitations: Products tested around 2012; the market may have changed. Funding and conflicts: US government funding (research support) Read: abstract
Benemei 2017 The anti-migraine component of butterbur extracts, isopetasin, desensitizes peptidergic nociceptors by acting on TRPA1 cation channel Br J Pharmacol · PMID 28622417 · DOI	Laboratory research animal	Cells, mouse and rat tissue	Isopetasin (a butterbur compound) root acute	Isopetasin first excited and then desensitised pain nerve endings that carry TRPA1, and reduced CGRP release from mouse spinal cord tissue and migraine-like responses in rodents. Lab findings only (not applicable) Methods, limits and funding Outcomes: TRPA1 channel and CGRP release Limitations: Lab and animal work; says nothing about doses or effects in people. Funding and conflicts: Not stated in the abstract Read: abstract
Borlak 2022 Petasites for Migraine Prevention: New Data on Mode of Action, Pharmacology and Safety. A Narrative Review Front Neurol · PMID 35585841 · DOI	Narrative review animal	Narrative review of published literature and study reports on Petadolex	Petadolex butterbur root extract root	Reports a no-observed-adverse-effect level at 15 times the highest clinical dose in rats, bile duct overgrowth at 45 to 90 times that dose, and no relevant abnormal liver tests in the trials it reviewed. Not applicable (not applicable) Methods, limits and funding Outcomes: Mode of action, safety, efficacy Limitations: Narrative review by authors with financial ties to the manufacturer. Funding and conflicts: Declared: manufacturer Weber and Weber funded studies and pays consultancy fees to authors Read: abstract
Braunwalder 2025 Efficacy and Safety of Phytotherapy and Anthroposophic Medicine in Seasonal Allergic Rhinitis: A Systematic Review Int Arch Allergy Immunol · PMID 39084196 · DOI	Systematic review	14 studies of herbs and anthroposophic medicines for hay fever	Phytotherapy including butterbur mixed varied varied · vs varied	Butterbur was the most studied plant and showed benefit on immune markers, symptoms and nasal airflow. About half the studies were rated sufficient quality. Qualitative (not applicable) Methods, limits and funding Outcomes: Seasonal allergic rhinitis Limitations: Qualitative synthesis only. Funding and conflicts: Not stated in the abstract Read: abstract
Danesch 2003 Safety of a patented special butterbur root extract for migraine prevention Headache · PMID 12864764 · DOI	Review	Review of animal toxicity, mutagenicity, trial, post-marketing and pharmacovigilance data on one patented extract	Patented special butterbur root extract (Petadolex) root	Concluded the extract is safe for human treatment. This was written in 2003, before later liver-injury reports were collected. Not applicable (not applicable) Methods, limits and funding Outcomes: Safety Limitations: Authors linked to the manufacturer; older than the liver reports analysed in 2019. Funding and conflicts: Author affiliation: Weber and Weber (manufacturer) Read: abstract
Danesch 2004 Petasites hybridus (Butterbur root) extract in the treatment of asthma--an open trial Altern Med Rev · PMID 15005644	Open label trial	80 64 adults and 16 children with asthma	Petadolex butterbur root extract root 2 months, then 2 optional months	Attacks, peak flow, FEV1 and symptoms improved; over 40% of those using asthma medicine at the start cut their intake. No control group (not given) Methods, limits and funding Outcomes: Asthma attacks, peak flow, FEV1, medicine use Limitations: Open, non-randomised, no placebo, manufacturer-linked author. Funding and conflicts: Author affiliation not stated in the abstract Read: abstract
Diener 2004 The first placebo-controlled trial of a special butterbur root extract for the prevention of migraine: reanalysis of efficacy criteria Eur Neurol · PMID 14752215 · DOI	Controlled trial	60 Adults with migraine (33 on extract, 27 on placebo)	Petadolex butterbur root extract root 25 mg capsules, two capsules twice a day 3 months (after a 4-week baseline) · vs placebo	Attacks per month fell from 3.4 to 1.8 on butterbur (p=0.0024) and from 2.9 to 2.6 on placebo (not significant). Responders: 45% vs 15%. About 1.6 fewer attacks a month on butterbur vs 0.3 on placebo (not given) Methods, limits and funding Outcomes: Attacks per month; share with at least a 50% fall (responders) Limitations: Small trial. The authors say the original protocol and analysis had major shortcomings, so this is a reanalysis done to meet regulatory needs. Funding and conflicts: Danesch is listed with the manufacturer Weber and Weber in a 2003 paper; funding is not stated in the abstract. Treat as industry-linked. Read: abstract
Disch 2018 In Vitro and In Situ Absorption and Metabolism of Sesquiterpenes from Petasites hybridus Extracts Planta Med · PMID 29341029 · DOI	Laboratory research in-vitro	Cell models, rat intestine, human and rat liver fractions	Petasins from Ze 339 leaf	High permeability and high metabolism; liver metabolism was higher than gut metabolism in both rat and human fractions. Lab findings only (not applicable) Methods, limits and funding Outcomes: Absorption and metabolism Limitations: No blood levels in people. Funding and conflicts: Four of five authors employed by the manufacturer Read: abstract
EMA Committee on Herbal Medicinal Products (HMPC) 2016 Public statement on contamination of herbal medicinal products with pyrrolizidine alkaloids EMA (ema.europa.eu) · Source	Authoritative guidance	Herbal medicine makers and regulators	Herbal products containing pyrrolizidine alkaloids unspecified	Only the document title and date (8 June 2016) could be read; the text was unreadable. Secondary web summaries say the statement and a later revision (2021) set an acceptable intake of 1 microgram a day of PAs for adults. NOT verified. Methods, limits and funding Limitations: Body text not read. No EMA monograph specific to Petasites hybridus was found. Read: summary-only
Fiebich 2005 Petasites hybridus extracts in vitro inhibit COX-2 and PGE2 release by direct interaction with the enzyme and by preventing p42/44 MAP kinase activation in rat primary microglial cells Planta Med · PMID 15678367 · DOI	Laboratory research in-vitro	Rat brain immune cells and enzyme tests	Four lipophilic Petasites hybridus rhizome extracts rhizome acute	Extracts blocked COX-2 more than COX-1. This did not depend on petasin or isopetasin content, and the pure compounds did not block either enzyme. Lab findings only (not applicable) Methods, limits and funding Outcomes: COX-1 and COX-2 enzyme activity; PGE2 release Limitations: Lab only. Funding and conflicts: Not stated in the abstract Read: abstract
Forsch 2020 In vitro hepatotoxicity of Petasites hybridus extract (Ze 339) depends on the concentration, the cytochrome activity of the cell system, and the species used Phytother Res · PMID 31631423 · DOI	Laboratory research in-vitro	Human and rodent liver cell lines and liver fractions	Ze 339 butterbur leaf extract, its constituents and metabolites leaf acute	Toxicity depended on the concentration, the enzyme activity of the cell type and the species. Lab findings only (not applicable) Methods, limits and funding Outcomes: Cell toxicity; reactive metabolites Limitations: Lab only; cannot give a risk in people. Funding and conflicts: Authors are employees of Max Zeller Soehne AG Read: abstract
GBIF Secretariat 2026 GBIF backbone: Petasites hybridus (L.) G.Gaertn., B.Mey. & Scherb. GBIF species API (api.gbif.org), retrieved 2026-10-25 · Source	Database record	Plant taxonomy	Petasites hybridus unspecified	Accepted species, key 9490132, family Asteraceae, genus Petasites, order Asterales. Synonyms listed: Petasites major, Petasites ovalis, Petasites reuterianus. Methods, limits and funding Limitations: Database record. Read: full-text
Gray 2004 Effects of butterbur treatment in intermittent allergic rhinitis: a placebo-controlled evaluation Ann Allergy Asthma Immunol · PMID 15281472 · DOI	Controlled trial	35 Adults with intermittent allergic rhinitis in the grass pollen season	Butterbur (product not named in the abstract) 50 mg twice a day 2 weeks (crossover) · vs placebo	No significant effect on any outcome. For example total nasal symptom score 3.4 and 3.5 (morning, evening) on butterbur vs 3.7 and 3.8 on placebo (maximum 12). No difference from placebo (not given) Methods, limits and funding Outcomes: Peak nasal inspiratory flow; nasal and eye symptoms; quality of life Limitations: Small crossover trial; a different product from the Ze 339 trials may explain the difference. Funding and conflicts: Not stated in the abstract Read: abstract
Guo 2007 Herbal medicines for the treatment of allergic rhinitis: a systematic review Ann Allergy Asthma Immunol · PMID 18219828 · DOI	Systematic review	16 double-blind RCTs of herbal medicines for allergic rhinitis; six on butterbur	Petasites hybridus extract and other herbs mixed varied varied · vs placebo or antihistamine	Six butterbur trials suggested it beat placebo or matched non-sedating antihistamines for intermittent hay fever. The authors want independent replication because the manufacturer paid for the three large trials. Not pooled (not applicable) Methods, limits and funding Outcomes: Allergic rhinitis symptoms Limitations: Manufacturer funding of the large trials. Funding and conflicts: Noted: manufacturer financial support for 3 large trials Read: abstract
Hashida 2023 Application to Butterbur Products of a Suggested Daily Intake-Based Safety Evaluation of Individual Herbal Supplements with Cytochrome P450 Expression as a Major Index J Nutr Sci Vitaminol (Tokyo) · PMID 37394426 · DOI	Animal research animal	Rats fed three PA-free butterbur products	Three commercial PA-free butterbur products (two oily, one powder) 100 times a human suggested daily intake 8 days	Two oily products raised CYP2B gene activity more than 10-fold and CYP3A1 less than 4-fold, with liver enlargement. The powder did not. Product-dependent effects in rats (not applicable) Methods, limits and funding Outcomes: Liver enzyme gene activity; kidney Limitations: Rats; short; relevance for people unknown. Funding and conflicts: Not stated in the abstract Read: abstract
Hirono 1977 Carcinogenic activity of petasitenine, a new pyrrolizidine alkaloid isolated from Petasites japonicus Maxim J Natl Cancer Inst · PMID 191625 · DOI	Animal research animal	ACI rats	Petasitenine (a PA from Petasites japonicus, a different species) flower stalk 0.05% or 0.01% in drinking water up to over 160 days · vs untreated rats	At the higher level all rats died or were killed within 72 days with liver necrosis; at the lower level 8 of 10 survivors beyond 160 days developed liver tumours. Carcinogenic in rats (not applicable) Methods, limits and funding Outcomes: Liver damage and tumours Limitations: Different species, one purified PA at high dose in rats. Funding and conflicts: Not stated in the abstract Read: abstract
Holland 2012 Evidence-based guideline update: NSAIDs and other complementary treatments for episodic migraine prevention in adults: report of the Quality Standards Subcommittee of the American Academy of Neurology and the American Headache Society Neurology · PMID 22529203 · DOI	Authoritative guidance	Adults with episodic migraine (guideline panel reviewed 284 abstracts, 49 Class I or II articles)	Guideline on NSAIDs and complementary treatments for migraine prevention; butterbur = Petasites root	Rated Petasites (butterbur) Level A (effective, should be offered) and MIG-99 feverfew, magnesium and riboflavin Level B. NCCIH (Nov 2024) says the AAN stopped recommending butterbur in 2015 because of safety concerns; the AAN notice itself could not be opened. Level A at the time (not applicable) Methods, limits and funding Outcomes: Evidence-based recommendations Limitations: Based on studies from 1999 to 2009. Built on the same few Petadolex trials. The Level A rating for butterbur was withdrawn in 2015 according to NCCIH, so do not read this recommendation as current advice. Funding and conflicts: Not stated in the abstract Read: abstract
Jackson 2004 The effects of butterbur on the histamine and allergen cutaneous response Ann Allergy Asthma Immunol · PMID 14989395 · DOI	Controlled trial	20 Atopic adults (completers)	Butterbur (product not named in the abstract) 50 mg twice a day 1 week (crossover) · vs fexofenadine, montelukast, placebo	Butterbur did not change the skin reaction compared with placebo; fexofenadine did. The authors conclude butterbur may not work in allergic skin disease. No effect on skin reaction (not applicable) Methods, limits and funding Outcomes: Skin wheal and flare to histamine and allergen Limitations: Small, short; a skin-prick test, not a symptom outcome. Funding and conflicts: Not stated in the abstract Read: abstract
Kleeberg-Hartmann 2021 Petasin and isopetasin reduce CGRP release from trigeminal afferents indicating an inhibitory effect on TRPA1 and TRPV1 receptor channels J Headache Pain · PMID 33849430 · DOI	Laboratory research animal	Rat and mouse skull and nerve preparations	Petasin and isopetasin root acute · vs vehicle	Both compounds reduced CGRP release provoked by TRPA1 and TRPV1 triggers in a roughly dose-dependent way; confirmed in mice lacking those channels. Lab findings only (not applicable) Methods, limits and funding Outcomes: CGRP release Limitations: Lab only. Funding and conflicts: No competing interests declared Read: abstract
Kulinowski Ł 2022 A review on the ethnobotany, phytochemistry, pharmacology and toxicology of butterbur species (Petasites L.) J Ethnopharmacol · PMID 35427728 · DOI	Narrative review	Review of the genus Petasites	Petasites species mixed	Lists traditional uses (migraine, asthma, hay fever, bronchitis, spastic cough, high blood pressure, ulcers, period pain) and reports over 200 sesquiterpenes and pyrrolizidine alkaloids in the genus. Not applicable (not applicable) Methods, limits and funding Outcomes: Traditional use, chemistry, toxicology Limitations: Review. Funding and conflicts: Not stated in the abstract Read: abstract
Käufeler 2006 Efficacy and safety of butterbur herbal extract Ze 339 in seasonal allergic rhinitis: postmarketing surveillance study Adv Ther · PMID 16751170 · DOI	Observational study	580 Adults with hay fever in routine care	Butterbur leaf CO2 extract Ze 339 (8 mg petasines per tablet) leaf average 2 tablets a day 2 weeks	Symptoms improved in 90% of people. Side effects 3.8%, mostly stomach complaints. 44% also took allergy medicine and did no better than butterbur alone. No control group (not given) Methods, limits and funding Outcomes: Symptom visual scales; satisfaction Limitations: Open, no placebo; manufacturer-linked authors. Funding and conflicts: Authors linked to the manufacturer Read: abstract
Lee 2003 Butterbur, a herbal remedy, attenuates adenosine monophosphate induced nasal responsiveness in seasonal allergic rhinitis Clin Exp Allergy · PMID 12859442 · DOI	Controlled trial	20 Adults with grass-pollen hay fever	Butterbur (product not named in the abstract) 50 mg twice a day 2 weeks (crossover) · vs placebo	Recovery (area under the curve) was smaller on butterbur: 584 vs 1438; maximum fall in nasal airflow 30% vs 43%. Mean difference 854 (area) and 13 percentage points (maximum fall) (95% CI 95 to 1614; 1 to 25) Methods, limits and funding Outcomes: Nasal response to AMP challenge Limitations: Small crossover challenge study. Funding and conflicts: Not stated in the abstract Read: abstract
Lee 2004 A placebo-controlled evaluation of butterbur and fexofenadine on objective and subjective outcomes in perennial allergic rhinitis Clin Exp Allergy · PMID 15080820 · DOI	Controlled trial	16 Adults with year-round allergic rhinitis and house dust mite allergy	Butterbur (product not named in the abstract) 50 mg twice a day 1 week (crossover) · vs fexofenadine and placebo	Maximum fall in nasal airflow after challenge was smaller on butterbur (34%) than placebo (46%); nasal symptom score 1.8 vs 2.8. Butterbur and fexofenadine did not differ. Better than placebo, similar to fexofenadine (P<0.05 vs placebo) Methods, limits and funding Outcomes: Nasal response to AMP challenge; nasal symptoms Limitations: Very small, one week; challenge test. Funding and conflicts: Not stated in the abstract Read: abstract
Lee 2004 Butterbur, a herbal remedy, confers complementary anti-inflammatory activity in asthmatic patients receiving inhaled corticosteroids Clin Exp Allergy · PMID 14720270 · DOI	Controlled trial	16 Adults with allergic asthma already on inhaled corticosteroids	Butterbur (product not named in the abstract) 25 mg twice a day 1 week (crossover) · vs placebo	PC20 improved by 0.6 doubling doses on butterbur vs -0.1 on placebo; exhaled nitric oxide fell 1.2 ppb vs a rise of 0.5; eosinophil cationic protein fell 3.9 vs rose 3.3 microg/L; blood eosinophils fell 31 vs rose 38 million per litre. 0.7 doubling-dose difference in PC20 (not given) Methods, limits and funding Outcomes: AMP bronchial challenge (PC20); exhaled nitric oxide; eosinophils Limitations: Very small and only one week; these are lab markers, not asthma attacks. Funding and conflicts: Not stated in the abstract Read: abstract
Lim 2024 Medicinal plants for allergic rhinitis: A systematic review and meta-analysis PLoS One · PMID 38603736 · DOI	Meta-analysis	1879 29 RCTs of single medicinal plants for allergic rhinitis (butterbur was the most studied, 5 studies)	Single medicinal plants (all plants combined) mixed varied up to 8 weeks · vs placebo or antihistamine	Compared with placebo, plants may improve total nasal symptoms (SMD -0.31, 95% CI -0.59 to -0.02; 249 people, 5 studies) with very-low to low certainty. Versus antihistamines: no clear difference (SMD -0.14, 95% CI -0.46 to 0.18), moderate certainty. These figures cover all plants, not butterbur alone. SMD -0.31 vs placebo (-0.59 to -0.02) Methods, limits and funding Outcomes: Total nasal symptoms; quality of life Limitations: Figures are for all plants pooled. Authors say clear beneficial evidence is still lacking. Funding and conflicts: No competing interests declared Read: abstract
Lipton 2004 Petasites hybridus root (butterbur) is an effective preventive treatment for migraine Neurology · PMID 15623680 · DOI	Controlled trial	245 Adults aged 18 to 65 with migraine (2 to 6 attacks a month)	Special standardised Petasites hybridus root extract root 75 mg twice a day or 50 mg twice a day 4 months · vs placebo	Per-protocol: attacks fell 48% on the higher dose (p=0.0012 vs placebo), 36% on the lower dose (p=0.127, not significant) and 26% on placebo. At least a 50% cut in attacks: 68% on the higher dose vs 49% on placebo (p<0.05). Mild stomach events, mostly burping, were the most common side effect. Higher dose beat placebo; lower dose did not (p=0.0012 (higher dose, per-protocol); p=0.127 (lower dose)) Methods, limits and funding Outcomes: Change in migraine attacks per month (percent change from baseline) Limitations: The abstract reports the per-protocol analysis. Only the higher dose reached significance. Placebo group still improved by 26%. Funding and conflicts: Not stated in the abstract Read: abstract
Lopresti 2020 Herbal treatments for migraine: A systematic review of randomised-controlled studies Phytother Res · PMID 32310327 · DOI	Systematic review	Adults (and some children) in 19 randomised trials of nine single herbs for migraine	Single-herb treatments incl. butterbur and feverfew mixed varied varied · vs placebo/active	Findings for feverfew were mixed; there was positive but limited evidence for butterbur. Risk of bias was high for many studies. Not pooled (not applicable) Methods, limits and funding Outcomes: Migraine prevention or acute treatment Limitations: Many included trials were small with high risk of bias. Funding and conflicts: Not stated in the abstract Read: abstract
Medsafe 2012 Butterbur and liver toxicity (Medsafe Prescriber Update) Medsafe (medsafe.govt.nz) · Source	Authoritative guidance	Health professionals	Butterbur Petasites hybridus products unspecified	June 2012 note. Says the MHRA issued a consumer alert about liver toxicity from unlicensed butterbur products, and cites 40 literature cases including nine of acute hepatitis and two of liver failure needing transplant. Reports VigiBase (WHO) liver reports without a count. Does not mention Germany, Switzerland or Petadolex. Methods, limits and funding Limitations: Marked as more than five years old; read through a page summary tool. Read: summary-only
Merk 2025 Efficacy and Safety of P. hybridus Leaf Extract Ze 339 for the Treatment of Allergic Rhinitis Adv Respir Med · PMID 40558112 · DOI	Narrative review	Review of preclinical, clinical, real-world and pharmacovigilance data on Ze 339	Butterbur leaf extract Ze 339 leaf	Concludes Ze 339 is a promising alternative to antihistamines with a favourable safety profile and no sleepiness. Not applicable (not applicable) Methods, limits and funding Outcomes: Efficacy and safety in allergic rhinitis Limitations: Three of four authors work for the manufacturer. Funding and conflicts: Three authors employed by Max Zeller Soehne AG, maker of Ze 339 Read: abstract
NCBI 2026 NCBI Taxonomy: Petasites hybridus NCBI Taxonomy (eutils), retrieved 2026-10-25 · Source	Database record	Plant taxonomy	Petasites hybridus unspecified	Taxonomy ID 278673, rank species. Methods, limits and funding Limitations: Database record. Read: full-text
NCBI PubChem 2026 PubChem compound records PubChem PUG REST, retrieved 2026-10-25 · Source	Database record	Chemistry	Petasin, isopetasin, S-petasin, senecionine, petasitenine, senkirkine unspecified	CIDs: (+)-petasin 5281526 (C20H28O3); isopetasin 5318627 (C20H28O3); S-petasin 16219858 (C19H26O3S); senecionine 5280906 (C18H25NO5); petasitenine 5281741 (C19H27NO7); senkirkine 5281752 (C19H27NO6). Methods, limits and funding Limitations: Database records. Read: full-text
National Center for Complementary and Integrative Health 2024 Butterbur: Usefulness and Safety NCCIH (nccih.nih.gov) · Source	Authoritative guidance	General public	Butterbur unspecified	Last updated Nov 2024. Says butterbur root extract may reduce migraine frequency in adults and children and that a leaf extract may help hay fever symptoms. Says the AAN recommended butterbur in 2012 but stopped in 2015 over safety concerns; PAs can damage liver and lungs and may cause cancer; only PA-free products should be considered; PA-free products seemed safe for up to 16 weeks but rare liver injury has been reported with products said to be PA-free; PA-containing products should not be used in pregnancy or breastfeeding; little is known about PA-free products in these cases. Methods, limits and funding Limitations: Read through a page summary tool, not word for word. Read: summary-only
Oelkers-Ax 2008 Butterbur root extract and music therapy in the prevention of childhood migraine: an explorative study Eur J Pain · PMID 17659990 · DOI	Controlled trial	58 Primary-school children with migraine (19 butterbur, 20 music therapy, 19 placebo)	Petadolex butterbur root extract root 12 weeks of treatment, then follow-up · vs placebo (and music therapy as a second active arm)	Right after treatment only music therapy beat placebo (p=0.005). At follow-up both music therapy (p=0.018) and butterbur (p=0.044) beat placebo. All groups improved a lot even before treatment. Butterbur beat placebo only at the follow-up (p=0.044 at follow-up (completers)) Methods, limits and funding Outcomes: Fall in headache frequency after treatment and 6 months later Limitations: Explorative, small, partly double-blind, completers-only analysis; big improvement in the baseline phase. Funding and conflicts: Not stated in the abstract Read: abstract
Orr 2014 Nutraceuticals in the prophylaxis of pediatric migraine: Evidence-based review and recommendations Cephalalgia · PMID 24443395 · DOI	Systematic review	30 studies of six nutraceuticals for migraine prevention in children (butterbur, riboflavin, ginkgolide B, magnesium, CoQ10, omega-3)	Several nutraceuticals incl. butterbur mixed varied varied · vs varied	The authors judged the overall quality of evidence for nutraceuticals in children to be poor. Not pooled (not applicable) Methods, limits and funding Outcomes: Pediatric migraine prevention Limitations: Poor quality evidence base. Funding and conflicts: Not stated in the abstract Read: abstract
Pothmann 2005 Migraine prevention in children and adolescents: results of an open study with a special butterbur root extract Headache · PMID 15836592 · DOI	Open label trial	108 Children and teenagers aged 6 to 17 with migraine	Butterbur root extract (special extract) root 50 to 150 mg a day depending on age 4 months	77% had at least a 50% fewer attacks; attacks fell 63% on average. Side effects 7.4%, mostly burping; none serious. 63% fewer attacks (no control group) (not given) Methods, limits and funding Outcomes: Attack frequency in migraine diaries Limitations: No placebo group, so improvement cannot be credited to butterbur alone; migraine improves with placebo in children. Funding and conflicts: Not stated in the abstract Read: abstract
Schapowal 2002 Randomised controlled trial of butterbur and cetirizine for treating seasonal allergic rhinitis BMJ · PMID 11799030 · DOI	Controlled trial	125 Adults with hay fever (seasonal allergic rhinitis), 61 butterbur and 64 cetirizine	Butterbur CO2 extract tablets (Ze 339, leaf) leaf one tablet four times a day 2 weeks · vs cetirizine	Improvement was similar in both groups; global scores were the same (median 3). About two thirds of side effects on cetirizine were drowsiness or fatigue. Similar to cetirizine; no placebo arm (not given) Methods, limits and funding Outcomes: Quality of life (SF-36) and clinical global impression Limitations: No placebo group, so it shows butterbur was similar to an antihistamine, not that either beat no treatment. Guo 2007 says the three large butterbur trials had manufacturer funding. Funding and conflicts: Manufacturer support for the large trials per Guo 2007 Read: abstract
Schapowal 2004 Butterbur Ze339 for the treatment of intermittent allergic rhinitis: dose-dependent efficacy in a prospective, randomized, double-blind, placebo-controlled study Arch Otolaryngol Head Neck Surg · PMID 15611396 · DOI	Controlled trial	186 Adults with intermittent allergic rhinitis (60 high dose, 65 low dose, 61 placebo)	Butterbur Ze 339 leaf extract tablets leaf one tablet three times a day (high) or twice a day (low) 2 weeks · vs placebo	Both butterbur groups improved more than placebo, with more benefit on the higher dose. Side effects were the same as on placebo. The abstract gives no numbers. Better than placebo (numbers not in abstract) (not given in abstract) Methods, limits and funding Outcomes: Change in daytime symptoms Limitations: Manufacturer-funded per Guo 2007. Only 2 weeks. Funding and conflicts: Manufacturer support per Guo 2007 Read: abstract
Schapowal 2005 Treating intermittent allergic rhinitis: a prospective, randomized, placebo and antihistamine-controlled study of Butterbur extract Ze 339 Phytother Res · PMID 16114089 · DOI	Controlled trial	330 Adults with intermittent allergic rhinitis	Butterbur Ze 339 leaf extract (8 mg total petasine per tablet) leaf one tablet three times a day · vs fexofenadine and placebo	Both butterbur and fexofenadine beat placebo (p<0.001); butterbur and fexofenadine did not differ (p=0.37). Butterbur matched fexofenadine and beat placebo (p<0.001 vs placebo) Methods, limits and funding Outcomes: Change in daytime symptoms Limitations: Manufacturer-funded per Guo 2007. Funding and conflicts: Manufacturer support per Guo 2007 (three large trials) Read: abstract
Schenk 2015 UPLC TOF MS for sensitive quantification of naturally occurring pyrrolizidine alkaloids in Petasites hybridus extract (Ze 339) J Chromatogr B Analyt Technol Biomed Life Sci · PMID 26094206 · DOI	Lab analysis	Batches of one butterbur leaf extract	Ze 339 butterbur leaf CO2 extract leaf	No PAs were found in the tested batches; the test could detect 2 parts per billion. PAs absent in tested batches (not applicable) Methods, limits and funding Outcomes: Pyrrolizidine alkaloid levels Limitations: Manufacturer authors; only that one product. Funding and conflicts: Manufacturer-based authors (Zeller) Read: abstract
Silva-Néto 2026 Efficacy of Petasites hybridus in migraine prophylaxis: the first real-world study Front Neurol · PMID 41908273 · DOI	Observational study	120 Adults with episodic (72) or chronic (48) migraine, single clinic in Brazil	Petasites hybridus (product Petamig, described as free of pyrrolizidine alkaloids by the author) 12 weeks	Headache days fell from 6.0 to 2.6 (episodic) and 26.3 to 12.7 (chronic). 59% had at least a 50% cut. Side effects in 28.3% (bitter taste or burping). Fewer headache days (no control group) (p<0.0001 (before vs after)) Methods, limits and funding Outcomes: Headache days per month Limitations: Retrospective, uncontrolled, one clinic. Cannot separate drug from placebo or natural change. PA-free claim comes from the author, not from testing in this paper. Funding and conflicts: No competing interests declared Read: abstract
UK Medicines and Healthcare products Regulatory Agency 2012 MHRA advice against unlicensed butterbur products (as reported by secondary sources) Reported by Medsafe, ConsumerLab and trade press · Source	Authoritative guidance	UK consumers	Unlicensed Petasites hybridus products unspecified	Early 2012: advised consumers not to take unlicensed butterbur products because of liver toxicity risk linked to pyrrolizidine alkaloids, and to stop and speak to a GP or pharmacist if worried. A voluntary withdrawal of unlicensed products was reported by one trade source. The gov.uk notice itself was NOT opened. Methods, limits and funding Limitations: NOT verified against the MHRA page. Anderson 2019 also says MHRA declined a manufacturer's 2011 application for a traditional herbal registration on safety grounds. Read: summary-only
Wildi 1998 Quantitative analysis of petasin and pyrrolizidine alkaloids in leaves and rhizomes of in situ grown Petasites hybridus plants Planta Med · PMID 17253240 · DOI	Lab analysis	Wild Petasites hybridus plants in Switzerland	Petasites hybridus leaves and rhizomes leaf and rhizome	Petasin 3.3 to 11.4 mg/g in leaves and 7.4 to 15.3 mg/g in rhizomes. PAs 5 to 90 ppm in rhizomes and 0.02 to 1.50 ppm in leaves. Leaves had far less PA (not applicable) Methods, limits and funding Outcomes: Petasin and PA content Limitations: Swiss wild populations only. Funding and conflicts: Not stated in the abstract Read: abstract

What we do not know yet

What the evidence supports: low-certainty reductions in migraine attacks on the higher dose of a root extract and hay fever symptom relief from a leaf extract. Nothing is rated above low.
What is uncertain: the trials are small, short, mostly of two manufacturer extracts, and one independent hay fever trial was null.
No independent large long trial of butterbur for migraine or hay fever was found.
The frequency of liver injury is unknown; the case-series authors have manufacturer ties and no rate was given.
The regulator and guideline notices (MHRA, German BfArM, AAN) were read only through secondary sources; no Swiss action and no EU monograph for Petasites hybridus were found. The LiverTox, LactMed and FDA pages were not opened.
No human interaction studies and no pregnancy data for PA-free products were found.
Only abstracts were read for the studies cited. Amounts and product comparison are not covered here; ask a doctor or pharmacist.

How new research is added

New studies are found by a saved PubMed search and are placed in a review queue. A person reads each one before it changes any evidence rating or safety wording. New publications never change ratings automatically.

Lab markers

What changed in blood or saliva tests during trials. A changed number is not the same as feeling or being healthier, and we give no "normal ranges".

Other 8

▼

Migraine attacks per month

Lowered in trials

How many separate migraine attacks a person has in a month, recorded in a diary.

Low certainty

Results and what it means

Why it was measured

The main outcome in migraine prevention trials.

Statistical vs. clinical meaning

Statistically significant in the adult trials for the higher dose; the placebo group also improved a lot.

Safety implication

No direct link; see the safety tab for liver warnings.

Study	Result	People	Product
²⁹	Per-protocol: attacks fell 48% (higher dose), 36% (lower dose, not significant) and 26% (placebo) over 4 months.	245 adults with migraine	Special root extract twice daily vs placebo, 4 months
¹⁰	3.4 to 1.8 attacks a month on butterbur (p=0.0024); 2.9 to 2.6 on placebo (not significant).	60 adults with migraine	Petadolex root extract, 3 months
³⁸	Attack frequency fell 63% in an open study.	108 children and teens	Root extract, 4 months, no control group
³⁶	Butterbur beat placebo only at the 6-month follow-up (p=0.044); all groups improved at baseline.	58 primary-school children	Petadolex, 12 weeks

^{29 10 38 36}

▲

People with at least half as many attacks (responders)

Raised in trials

The share of participants whose attacks fall by 50% or more.

Low certainty

Results and what it means

Why it was measured

Shows how many people notice a real change, not just the average.

Statistical vs. clinical meaning

Meaningful on paper, but placebo response was also high (49%).

Safety implication

None.

Study	Result	People	Product
²⁹	68% on the higher dose vs 49% on placebo (p<0.05).	245 adults	Special root extract, 4 months
¹⁰	45% on butterbur vs 15% on placebo.	60 adults	Petadolex, 3 months
⁴³	59.2% had a headache-day reduction of at least 50% (no control group).	120 adults in one clinic	PA-free product, 12 weeks

^{29 10 43}

▼

CGRP release from nerve endings (lab)

Lowered in trials

CGRP is a small protein released by trigeminal nerves that is linked to migraine pain.

Not enough evidence

Results and what it means

Why it was measured

Used in the lab to test whether a compound might act on migraine pathways.

Statistical vs. clinical meaning

Lab finding only. CGRP was not measured in butterbur trial participants in the sources read.

Safety implication

None.

Study	Result	People	Product
²²	Petasin and isopetasin reduced CGRP release provoked by TRPA1 and TRPV1 triggers in rodent tissue.	Rat and mouse tissue	Pre-incubation with petasin or isopetasin
⁵	Isopetasin desensitised and reduced CGRP release from mouse spinal cord tissue.	Mouse tissue and cells	Isopetasin

^{22 5}

↕

Total nasal symptom score

Mixed results in trials

A score adding up sneezing, runny nose, blockage and itching, usually from daily diary entries.

Low certainty

Results and what it means

Why it was measured

The main symptom outcome in hay fever trials.

Statistical vs. clinical meaning

Small average gains; one null trial; plant figure is not butterbur alone.

Safety implication

None.

Study	Result	People	Product
¹⁶	Butterbur 3.4 (morning) and 3.5 (evening) vs placebo 3.7 and 3.8 (maximum 12); not significant.	35 adults with hay fever	Butterbur 2 weeks crossover
²⁶	Butterbur 1.8 vs placebo 2.8 (p<0.05).	16 adults with year-round rhinitis	Butterbur 1 week crossover
²⁸	All single plants pooled vs placebo: SMD -0.31 (95% CI -0.59 to -0.02), 5 studies, 249 people.	Adults and children with allergic rhinitis	Single medicinal plants

^{16 26 28}

↕

Nasal airflow after allergy challenge (peak nasal inspiratory flow)

Mixed results in trials

How hard a person can breathe in through the nose; a fall after a challenge test shows nasal blockage.

Very low certainty

Results and what it means

Why it was measured

Objective measure of nasal blockage.

Statistical vs. clinical meaning

Mixed: positive after a lab challenge, none at home.

Safety implication

None.

Study	Result	People	Product
²⁵	Maximum fall 30% on butterbur vs 43% on placebo (difference 13, 95% CI 1 to 25).	20 adults with grass pollen hay fever	Butterbur 2 weeks crossover
¹⁶	Morning and evening values 107 and 114 L/min on butterbur vs 105 and 117 on placebo; no effect.	35 adults	Butterbur 2 weeks crossover

^{25 16}

●

Skin wheal and flare to histamine or allergen

No clear change in trials

The raised patch and redness on the skin after a skin-prick test.

Very low certainty

Results and what it means

Why it was measured

Shows whether a medicine blocks histamine or allergic skin reactions.

Statistical vs. clinical meaning

No effect in this test.

Safety implication

None.

Study	Result	People	Product
²¹	Histamine wheal 16.4 mm2 on butterbur vs 15.5 on placebo (not significant); fexofenadine 9.4 (significant).	20 atopic adults	Butterbur 1 week crossover

²¹

▼

Airway inflammation markers in asthma (exhaled nitric oxide, eosinophils)

Lowered in trials

Exhaled nitric oxide and blood eosinophil counts are lab signs of allergic airway inflammation.

Very low certainty

Results and what it means

Why it was measured

Surrogate markers of inflammation in asthma.

Statistical vs. clinical meaning

Lab markers in 16 people for one week; not asthma control.

Safety implication

None.

Study	Result	People	Product
²⁷	Exhaled nitric oxide changed by -1.2 ppb on butterbur vs +0.5 on placebo; blood eosinophils -31 vs +38 million per litre; eosinophil cationic protein -3.9 vs +3.3 microg/L.	16 adults with atopic asthma on inhaled steroids	Butterbur 25 mg twice daily, 1 week crossover

²⁷

▲

Airway sensitivity (AMP challenge PC20)

Raised in trials

PC20 is the dose of a trigger needed to cut lung airflow by 20%; a higher number means less sensitive airways.

Very low certainty

Results and what it means

Why it was measured

Objective lung test in asthma trials.

Statistical vs. clinical meaning

Small change in a tiny trial.

Safety implication

None.

Study	Result	People	Product
²⁷	PC20 improved by 0.6 doubling dilutions on butterbur vs -0.1 on placebo (0.7 difference).	16 adults with atopic asthma	Butterbur 1 week crossover

²⁷

Safety

What studies and official sources say about side effects, medicines and who should check first. Each point shows what kind of evidence it comes from.

Main points to know 7

Raw butterbur and unprocessed products contain pyrrolizidine alkaloids (PAs), which can poison the liver. Butterbur plants naturally contain PAs. NCCIH says PAs can damage the liver and lungs and may cause cancer. In wild Swiss plants PAs were 5 to 90 parts per million in rhizomes and 0.02 to 1.50 in leaves. A purified PA from a related species (Petasites japonicus) caused liver tumours in rats. For this reason only products tested and certified free of PAs should even be considered, and raw plant, tea or unprocessed root products should be avoided. Authoritative guidance ^{35 45 19}
Even PA-reduced or PA-free extracts have been linked to liver injury. NCCIH says PA-free products seemed safe for up to 16 weeks, but there have been rare liver injury cases with products reported to be PA-free. A review of reports about one root extract (Petadolex) found 48 spontaneous liver reports: 3 likely, 13 possible, 8 unlikely and 24 unclassifiable as caused by the extract; nine were severe (ten in the full text table) and two people needed liver transplants. The authors, who consult for the manufacturer, call the cases rare and often confounded by other drugs. No rate can be calculated from spontaneous reports. Case series ^{3 35}
Regulators warned against unlicensed butterbur products in 2012. Medsafe (New Zealand, June 2012) reports that the UK MHRA issued a consumer alert about liver toxicity from unlicensed butterbur products, citing 40 literature cases including nine of acute hepatitis and two of liver failure needing transplant. The MHRA page itself was not opened. Authoritative guidance ^{31 44}
A leading US neurology guideline stopped recommending butterbur. The 2012 AAN and American Headache Society guideline rated butterbur Level A for migraine prevention. NCCIH (Nov 2024) says the AAN stopped recommending butterbur in 2015 because of safety concerns. The AAN notice itself could not be opened. Authoritative guidance ^{20 35}
Pregnancy and breastfeeding: do not use PA-containing products; little is known about PA-free ones. NCCIH says products containing PAs should not be used in pregnancy or while breastfeeding, and that little is known about PA-free products in these situations. No controlled pregnancy data on butterbur were found. Authoritative guidance ³⁵
Butterbur is a daisy-family plant (Asteraceae). People with allergies to ragweed, daisies, chrysanthemums or marigolds may want to ask a doctor first. No butterbur-specific allergy case was found in the sources read; this is a general precaution for the plant family. Pharmacological theory ¹⁵
Burping and a bitter taste are the most common effects. In the main migraine trial the most frequent possibly-related events were mild stomach events, mostly burping. In a clinic series, 28.3% had a bitter taste or burping lasting about 6 days. In a children's open study 7.4% had undesired effects, mostly burping. Controlled trial ^{29 43 38}

Side effects 4

Effect	How often	Severity and recovery	Evidence
Burping (eructation) and bitter taste Burping lasted about 6 days on average in the clinic series.	7.4% in children's open study (Pothmann); 28.3% in a clinic series (Silva-Neto); not given as a rate in the Lipton abstract	mild reversible	Controlled trial ^{29 38 43}
Stomach and digestive complaints Mostly non-specific digestive complaints.	3.8% of 580 people with hay fever on a leaf extract (Kaufeler 2006)	mild reversible	Observational ²⁴
Side effects similar to placebo in a 186-person hay fever trial Leaf extract for 2 weeks.	Not given in the abstract	mild reversible	Controlled trial ⁴⁰
Liver injury (herbal-induced liver injury) Onset averaged about 103 days in the severe cases. The reports concern a root extract.	Unknown. Cases are rare per the authors, but a rate was not calculated; 12 trial cases among 397 patients were not serious.	severe Resolved after stopping in reported cases; two people needed transplants	Case series ³

Liver

Liver injury is the main safety issue. Raw butterbur contains pyrrolizidine alkaloids that are liver poisons, and in animals one such alkaloid from a related species caused liver tumours. Products are supposed to remove them. A review of 48 spontaneous reports about the root extract Petadolex found mostly mild liver blood test changes, nine severe cases and two liver transplants; the authors rated 3 as likely, 13 possible and 8 unlikely linked to the extract, 24 could not be classified, and many patients were also taking drugs that can harm the liver. The authors, who have ties to the manufacturer, call the cases rare and idiosyncratic. At very high doses in rats (45 to 90 times the clinical dose) there was bile duct overgrowth; at about 200 times the dose liver enzymes rose and recovered by 6 months. Cell studies suggest petasin-rich extracts can be toxic at far above therapeutic levels. A rat study of three PA-free commercial products found two oily products strongly raised liver enzyme-gene activity with liver enlargement while a powder did not, so products differ. ^{3 2 14 18 6}

How often. No frequency of liver injury could be found. The authors say the cases are rare but give no rate. The LiverTox entry could not be opened.

Monitoring. No agency monitoring advice was found. Stop and get medical care for yellow skin or eyes, dark urine, pale stools, severe tiredness or upper-right belly pain. Anyone with liver disease, or who drinks heavily or takes other liver-affecting medicines, should ask a doctor first.

This is a safety connection. It is not evidence that Butterbur supports liver health.

Medicines and supplements 5

With	Why it matters	Importance	Evidence	What to do
Paracetamol (acetaminophen), ibuprofen and other pain relievers often used for migraine over-the-counter	22 of the people in the liver-injury reports also took drugs that can harm the liver. In human liver cells, at about 49 times usual blood levels of the butterbur compounds plus ordinary levels of ibuprofen, paracetamol and naratriptan, liver enzyme and enzyme-activity changes were seen.	Possible added liver strain; the reports cannot show which drug caused injury.	Laboratory research Very low certainty ³	Ask a doctor or pharmacist before combining butterbur with regular pain relievers; keep a list of everything you take.
Medicines handled by liver enzymes (CYP enzymes) prescription	In rats, two oily PA-free butterbur products raised CYP2B gene activity more than 10-fold and CYP3A1 less than 4-fold. Cell experiments found that toxicity depended on enzyme activity of the cell type.	Theoretical: if the same happened in people, drug levels could change. No human interaction study was found.	Animal research Very low certainty ^{18 14}	Tell your prescriber and pharmacist if you take any regular medicine.
Other herbs or supplements that can harm the liver (for example herbs that contain pyrrolizidine alkaloids) herb	Added liver toxicity is plausible if both are taken.	Not studied.	Pharmacological theory Very low certainty ³⁵	Do not combine butterbur with other products with liver-safety warnings.
Alcohol alcohol	Alcohol is a liver stressor; combined effect with butterbur not studied.	Unknown.	Pharmacological theory Very low certainty ³⁵	Heavy drinkers should ask a doctor before use.
Antihistamines for hay fever prescription	In an open study 44% of people also took allergy medicine and did no better than butterbur alone; no harmful combination was reported.	No interaction signal found, but studies were not designed to look for one.	Observational Very low certainty ²⁴	Tell your doctor or pharmacist about all allergy medicines you use.

Who should check first 6

Situation	Advice	Evidence
Pregnancy and breastfeeding	NCCIH: PA-containing products should not be used. For PA-free products little is known, so avoid unless a doctor agrees.	Authoritative guidance ³⁵
Liver disease, unexplained liver test results, or heavy alcohol use	Avoid; butterbur has been linked to liver injury.	Authoritative guidance ^{35 3}
Any raw butterbur plant, tea, unprocessed root, or product that does not state it is certified free of pyrrolizidine alkaloids	Avoid.	Authoritative guidance ^{35 4}
Skin use of PA-containing products	NCCIH says PA-containing products may not be safe on the skin because PAs could be absorbed through cuts or scrapes.	Authoritative guidance ³⁵
Allergy to ragweed or other daisy-family plants	Ask a doctor first; cross-reaction is a general precaution for the plant family and not documented for butterbur in the sources read.	Pharmacological theory ¹⁵
Children	Used in small studies, but liver injury has been reported with the extract and the evidence is poor. Only under a paediatrician's advice.	Systematic review ^{37 38}

Overdose, product quality and regulators

Summary. The dose-limiting concern is the liver, from pyrrolizidine alkaloids in the plant and possibly from other constituents of extracts. Leaves have far less PA than rhizomes, and extraction removes PA to trace amounts in some products.

Animal. Petasitenine (a PA from Petasites japonicus, a different species) caused death within 72 days at 0.05% in drinking water and liver tumours in 8 of 10 survivors at 0.01%. For a root extract, the no-observed-adverse-effect level in rats was 15 times the top clinical dose; bile duct overgrowth appeared at 45 to 90 times; at about 200 times, enzymes rose then recovered by 6 months. This was industry-linked work.

Human doses studied. Trials ran up to 4 months (adult migraine) and 2 weeks (hay fever). Longer human safety is unknown.

High dose or long term signals. Liver injury cases occurred between 4 and 730 days of use in reports.

Overdose response. No overdose data found.

Genotoxicity. PAs are described as hepatotoxic, genotoxic and possibly carcinogenic by regulators (as quoted in a 2015 testing paper). Manufacturer paper reports mutagenicity studies of one extract with no concern; not independently verified.

Toxic plant parts. Rhizome (underground stem) and root hold the most PAs in wild plants (5 to 90 ppm in Swiss samples); leaves much less (0.02 to 1.50 ppm). Other Petasites species (for example P. japonicus) contain the PAs petasitenine and senkirkine, which are liver toxic.

Quality varies greatly. In a lab test of 21 market supplements, petasin ranged from 0.02 to 11.6 mg per dose in 15, was absent from six, and PAs were found in seven. Because of this, a label that says butterbur is not proof of safety.

PAs in market supplements. Pyrrolizidine alkaloids were detected in seven of 21 products claiming to contain Petasites hybridus; seven others had petasin within label claim and no PAs. ⁴

PA-free leaf extract. No PAs were detected in tested batches of one leaf extract (Ze 339), with a detection limit of 2 parts per billion. Manufacturer-based authors. ⁴²

Products differ even when PA-free. Three PA-free commercial products tested in rats gave very different liver effects; two oily ones were worse than a powder. ¹⁸

Body	Action	Date
NCCIH (US National Institutes of Health) (United States)	Butterbur page: usefulness and safety. Says root extract may reduce migraine frequency, leaf extract may help hay fever, PAs can damage liver and lungs and may cause cancer, only PA-free products should be considered, rare liver injury reported even with PA-free products, avoid PA products in pregnancy and breastfeeding, and that the AAN stopped recommending butterbur in 2015. source	2024-11
Medsafe (New Zealand)	June 2012 Prescriber Update on butterbur liver toxicity: cites 40 literature cases (nine acute hepatitis, two liver failure needing transplant) and WHO VigiBase reports. source	2012-06

When to get medical help

Yellow skin or eyes, dark urine or pale stools: stop and get medical care today.
Severe tiredness, nausea or vomiting with pain in the upper right belly: stop and get medical care.
Swelling of the face, lips or throat, or trouble breathing (allergic reaction): call emergency services.
Itching all over or a new rash after starting: stop and ask a doctor.
Any new unexplained bruising or bleeding: stop and get checked.
Asthma getting worse while using butterbur: use your prescribed rescue medicine and see a doctor.

Stop the supplement and contact a clinician or emergency service. In the US, Poison Control is 1-800-222-1222.

General information only, not medical advice. Ask a clinician or pharmacist before use, especially with medicines, in pregnancy or with a health condition. See the Natural Remedy Safety guide.

Frequently asked questions

Does butterbur prevent migraine?

Possibly a little, but the evidence is low certainty. Two small trials of one special root extract found fewer attacks on the higher dose than on placebo (48% vs 26% fewer in the larger trial), while the lower dose did not beat placebo. Some authors are linked to the manufacturer. A 2006 review rated the evidence moderate for a higher dose; a 2020 review called it positive but limited with high risk of bias.

^{29 10 1 30}

Is butterbur safe for the liver?

Not reliably. Raw plant contains pyrrolizidine alkaloids that can poison the liver. Even products made to remove them have been linked to liver injury: a review of reports about one root extract found 48 spontaneous liver reports, nine severe and two needing a transplant, though the authors, who have ties to the maker, call the cases rare and often confounded by other drugs. NCCIH says PA-free products have seemed safe for up to 16 weeks, but rare liver injury has been reported even with those.

^{3 35}

Why did a neurology guideline stop recommending butterbur?

The 2012 American Academy of Neurology and American Headache Society guideline gave butterbur its top rating (Level A) for preventing migraine. NCCIH says the AAN stopped recommending it in 2015 because of safety concerns. The AAN notice itself could not be opened.

^{20 35}

What are pyrrolizidine alkaloids (PAs)?

PAs are natural plant chemicals. NCCIH says they can damage the liver and lungs and may cause cancer. One PA from a related butterbur species caused liver tumours in rats. In wild Swiss plants rhizomes held 5 to 90 parts per million and leaves 0.02 to 1.50. Regulated extracts try to remove them to trace amounts.

^{35 19 45}

Are PA-free butterbur products safe?

They are safer than raw plant, but not proven safe. NCCIH says PA-free products have seemed safe for up to 16 weeks, with rare liver injury reported even so. Products also vary: a lab test of 21 market supplements found PAs in seven, and a rat study of three PA-free products found one powder had no liver effect while two oily ones raised liver enzyme-gene activity.

^{35 4 18}

What did regulators say?

The UK MHRA advised consumers in 2012 not to take unlicensed butterbur products because of liver toxicity concerns; this is reported by the New Zealand regulator Medsafe, and the MHRA notice itself was not opened. A 2019 review says a German agency removed one product from the market in 2002 because of a change in its make-up. No Swiss action and no EU herbal monograph for Petasites hybridus were found.

^{31 44 3}

Does butterbur help hay fever?

Perhaps for one leaf extract. In manufacturer-linked trials it matched cetirizine and fexofenadine and beat placebo in the larger one. But a small independent placebo-controlled trial found no benefit, and a 2024 pooled review of single plants rates the placebo comparison as very low to low certainty.

^{39 41 16 28}

Can children take butterbur?

Small studies exist, but the evidence is poor and liver injury has been reported with the extract. One randomised trial in 58 children found a benefit over placebo only at the 6-month follow-up. Only use it under a paediatrician's advice.

^{36 38 37}

Is butterbur safe in pregnancy or breastfeeding?

Avoid it. NCCIH says products containing PAs should not be used in pregnancy or while breastfeeding, and that little is known about PA-free products in those situations.

³⁵

What are the common side effects?

Burping and a bitter taste were the most common. In a clinic series 28.3% had them for about 6 days, in a children's study 7.4% had undesired effects mostly burping, and in a hay fever surveillance study of 580 people 3.8% had mostly digestive complaints.

^{43 38 24}

Does butterbur interact with medicines?

No human interaction study was found. In liver cells, high levels of butterbur compounds with paracetamol, ibuprofen or naratriptan changed liver measures, and in rats two oily products activated liver drug-handling genes. Many reported liver cases also involved other liver-affecting drugs. Tell your doctor or pharmacist about everything you take.

^{3 18}

Can butterbur replace my asthma or hay fever medicine?

No. Evidence for asthma is very low (a tiny one-week trial and an open study), and for hay fever it is low. Do not stop prescribed medicines. Ask a doctor or pharmacist about amounts and about anything you plan to combine.

^{27 9 17}

Explore by health goal

Immune Support Brain Health Inflammation Balance Digestive Support Blood Pressure Support

Editorial transparency

Written by: AlternativeMed editorial team. Research compiled with AI assistance, then checked against PubMed records.
Medical review: None. This page has not been reviewed by a physician or pharmacist.
First published: 2026-10-25
Last substantive update: 2026-10-25
How we research: We searched PubMed, read the NCCIH page and a Medsafe regulatory note through a summary tool, and recorded each study's design, people, product and result as reported. Only abstracts were read for most studies. The MHRA, BfArM and AAN notices, the EMA pyrrolizidine alkaloid statement, LiverTox and LactMed could not be opened and are marked not verified. Certainty ratings are the compiler's own reading unless a review's own rating is quoted. Several safety and efficacy papers are by authors linked to the manufacturers; this is flagged. No study is cited that we could not verify.
How we rate evidence: See "How we rate evidence" in the Evidence tab.
Independence: No product is promoted on this page. No affiliate links are shown. If one is added later it will be labelled and kept apart from the evidence.

Sources 45

Each source was opened and checked. Where we read only the abstract, it says so. The small numbers in the text point here.

Agosti R, Duke RK, Chrubasik JE et al.. Effectiveness of Petasites hybridus preparations in the prophylaxis of migraine: a systematic review. Phytomedicine. 2006. PMID 16987643. DOI 10.1016/j.phymed.2006.02.008. (abstract)
Anderson N, Meier T, Borlak J. Toxicogenomics applied to cultures of human hepatocytes enabled an identification of novel petasites hybridus extracts for the treatment of migraine with improved hepatobiliary safety. Toxicol Sci. 2009. PMID 19770483. DOI 10.1093/toxsci/kfp216. (abstract)
Anderson N, Borlak J. Hepatobiliary Events in Migraine Therapy with Herbs-The Case of Petadolex, A Petasites Hybridus Extract. J Clin Med. 2019. PMID 31083451. DOI 10.3390/jcm8050652. (full-text)
Avula B, Wang YH, Wang M et al.. Simultaneous determination of sesquiterpenes and pyrrolizidine alkaloids from the rhizomes of Petasites hybridus (L.) G.M. et Sch. and dietary supplements using UPLC-UV and HPLC-TOF-MS methods. J Pharm Biomed Anal. 2012. PMID 22809670. DOI 10.1016/j.jpba.2012.05.021. (abstract)
Benemei S, De Logu F, Li Puma S et al.. The anti-migraine component of butterbur extracts, isopetasin, desensitizes peptidergic nociceptors by acting on TRPA1 cation channel. Br J Pharmacol. 2017. PMID 28622417. DOI 10.1111/bph.13917. (abstract)
Borlak J, Diener HC, Kleeberg-Hartmann J et al.. Petasites for Migraine Prevention: New Data on Mode of Action, Pharmacology and Safety. A Narrative Review. Front Neurol. 2022. PMID 35585841. DOI 10.3389/fneur.2022.864689. (abstract)
Braunwalder C, Ertl J, Wullschleger M et al.. Efficacy and Safety of Phytotherapy and Anthroposophic Medicine in Seasonal Allergic Rhinitis: A Systematic Review. Int Arch Allergy Immunol. 2025. PMID 39084196. DOI 10.1159/000539645. (abstract)
Danesch U, Rittinghausen R. Safety of a patented special butterbur root extract for migraine prevention. Headache. 2003. PMID 12864764. DOI 10.1046/j.1526-4610.2003.03015.x. (abstract)
Danesch UC. Petasites hybridus (Butterbur root) extract in the treatment of asthma--an open trial. Altern Med Rev. 2004. PMID 15005644. (abstract)
Diener HC, Rahlfs VW, Danesch U. The first placebo-controlled trial of a special butterbur root extract for the prevention of migraine: reanalysis of efficacy criteria. Eur Neurol. 2004. PMID 14752215. DOI 10.1159/000076535. (abstract)
Disch L, Forsch K, Siewert B et al.. In Vitro and In Situ Absorption and Metabolism of Sesquiterpenes from Petasites hybridus Extracts. Planta Med. 2018. PMID 29341029. DOI 10.1055/s-0044-100401. (abstract)
EMA Committee on Herbal Medicinal Products (HMPC). Public statement on contamination of herbal medicinal products with pyrrolizidine alkaloids. EMA (ema.europa.eu). 2016. (summary-only)
Fiebich BL, Grozdeva M, Hess S et al.. Petasites hybridus extracts in vitro inhibit COX-2 and PGE2 release by direct interaction with the enzyme and by preventing p42/44 MAP kinase activation in rat primary microglial cells. Planta Med. 2005. PMID 15678367. DOI 10.1055/s-2005-837744. (abstract)
Forsch K, Schöning V, Assmann GM et al.. In vitro hepatotoxicity of Petasites hybridus extract (Ze 339) depends on the concentration, the cytochrome activity of the cell system, and the species used. Phytother Res. 2020. PMID 31631423. DOI 10.1002/ptr.6516. (abstract)
GBIF Secretariat. GBIF backbone: Petasites hybridus (L.) G.Gaertn., B.Mey. & Scherb.. GBIF species API (api.gbif.org), retrieved 2026-10-25. 2026. (full-text)
Gray RD, Haggart K, Lee DK et al.. Effects of butterbur treatment in intermittent allergic rhinitis: a placebo-controlled evaluation. Ann Allergy Asthma Immunol. 2004. PMID 15281472. DOI 10.1016/S1081-1206(10)61447-0. (abstract)
Guo R, Pittler MH, Ernst E. Herbal medicines for the treatment of allergic rhinitis: a systematic review. Ann Allergy Asthma Immunol. 2007. PMID 18219828. DOI 10.1016/S1081-1206(10)60375-4. (abstract)
Hashida H, Kurawaka M, Tatehana H et al.. Application to Butterbur Products of a Suggested Daily Intake-Based Safety Evaluation of Individual Herbal Supplements with Cytochrome P450 Expression as a Major Index. J Nutr Sci Vitaminol (Tokyo). 2023. PMID 37394426. DOI 10.3177/jnsv.69.206. (abstract)
Hirono I, Mori H, Yamada K et al.. Carcinogenic activity of petasitenine, a new pyrrolizidine alkaloid isolated from Petasites japonicus Maxim. J Natl Cancer Inst. 1977. PMID 191625. DOI 10.1093/jnci/58.4.1155. (abstract)
Holland S, Silberstein SD, Freitag F et al.. Evidence-based guideline update: NSAIDs and other complementary treatments for episodic migraine prevention in adults: report of the Quality Standards Subcommittee of the American Academy of Neurology and the American Headache Society. Neurology. 2012. PMID 22529203. DOI 10.1212/WNL.0b013e3182535d0c. (abstract)
Jackson CM, Lee DK, Lipworth BJ. The effects of butterbur on the histamine and allergen cutaneous response. Ann Allergy Asthma Immunol. 2004. PMID 14989395. DOI 10.1016/S1081-1206(10)61556-6. (abstract)
Kleeberg-Hartmann J, Vogler B, Messlinger K. Petasin and isopetasin reduce CGRP release from trigeminal afferents indicating an inhibitory effect on TRPA1 and TRPV1 receptor channels. J Headache Pain. 2021. PMID 33849430. DOI 10.1186/s10194-021-01235-5. (abstract)
Kulinowski Ł, Luca SV, Minceva M et al.. A review on the ethnobotany, phytochemistry, pharmacology and toxicology of butterbur species (Petasites L.). J Ethnopharmacol. 2022. PMID 35427728. DOI 10.1016/j.jep.2022.115263. (abstract)
Käufeler R, Polasek W, Brattström A et al.. Efficacy and safety of butterbur herbal extract Ze 339 in seasonal allergic rhinitis: postmarketing surveillance study. Adv Ther. 2006. PMID 16751170. DOI 10.1007/BF02850143. (abstract)
Lee DK, Carstairs IJ, Haggart K et al.. Butterbur, a herbal remedy, attenuates adenosine monophosphate induced nasal responsiveness in seasonal allergic rhinitis. Clin Exp Allergy. 2003. PMID 12859442. DOI 10.1046/j.1365-2222.2003.01705.x. (abstract)
Lee DK, Gray RD, Robb FM et al.. A placebo-controlled evaluation of butterbur and fexofenadine on objective and subjective outcomes in perennial allergic rhinitis. Clin Exp Allergy. 2004. PMID 15080820. DOI 10.1111/j.1365-2222.2004.1903.x. (abstract)
Lee DK, Haggart K, Robb FM et al.. Butterbur, a herbal remedy, confers complementary anti-inflammatory activity in asthmatic patients receiving inhaled corticosteroids. Clin Exp Allergy. 2004. PMID 14720270. DOI 10.1111/j.1365-2222.2004.01838.x. (abstract)
Lim XY, Lau MS, Zolkifli NA et al.. Medicinal plants for allergic rhinitis: A systematic review and meta-analysis. PLoS One. 2024. PMID 38603736. DOI 10.1371/journal.pone.0297839. (abstract)
Lipton RB, Göbel H, Einhäupl KM et al.. Petasites hybridus root (butterbur) is an effective preventive treatment for migraine. Neurology. 2004. PMID 15623680. DOI 10.1212/01.wnl.0000147290.68260.11. (abstract)
Lopresti AL, Smith SJ, Drummond PD. Herbal treatments for migraine: A systematic review of randomised-controlled studies. Phytother Res. 2020. PMID 32310327. DOI 10.1002/ptr.6701. (abstract)
Medsafe, New Zealand. Butterbur and liver toxicity (Medsafe Prescriber Update). Medsafe (medsafe.govt.nz). 2012. (summary-only)
Merk VM, Boonen G, Butterweck V et al.. Efficacy and Safety of P. hybridus Leaf Extract Ze 339 for the Treatment of Allergic Rhinitis. Adv Respir Med. 2025. PMID 40558112. DOI 10.3390/arm93030013. (abstract)
NCBI. NCBI Taxonomy: Petasites hybridus. NCBI Taxonomy (eutils), retrieved 2026-10-25. 2026. (full-text)
NCBI PubChem. PubChem compound records. PubChem PUG REST, retrieved 2026-10-25. 2026. (full-text)
National Center for Complementary and Integrative Health. Butterbur: Usefulness and Safety. NCCIH (nccih.nih.gov). 2024. (summary-only)
Oelkers-Ax R, Leins A, Parzer P et al.. Butterbur root extract and music therapy in the prevention of childhood migraine: an explorative study. Eur J Pain. 2008. PMID 17659990. DOI 10.1016/j.ejpain.2007.06.003. (abstract)
Orr SL, Venkateswaran S. Nutraceuticals in the prophylaxis of pediatric migraine: Evidence-based review and recommendations. Cephalalgia. 2014. PMID 24443395. DOI 10.1177/0333102413519512. (abstract)
Pothmann R, Danesch U. Migraine prevention in children and adolescents: results of an open study with a special butterbur root extract. Headache. 2005. PMID 15836592. DOI 10.1111/j.1526-4610.2005.05044.x. (abstract)
Schapowal A, Petasites Study Group. Randomised controlled trial of butterbur and cetirizine for treating seasonal allergic rhinitis. BMJ. 2002. PMID 11799030. DOI 10.1136/bmj.324.7330.144. (abstract)
Schapowal A, Petasites Study Group. Butterbur Ze339 for the treatment of intermittent allergic rhinitis: dose-dependent efficacy in a prospective, randomized, double-blind, placebo-controlled study. Arch Otolaryngol Head Neck Surg. 2004. PMID 15611396. DOI 10.1001/archotol.130.12.1381. (abstract)
Schapowal A, Study Group. Treating intermittent allergic rhinitis: a prospective, randomized, placebo and antihistamine-controlled study of Butterbur extract Ze 339. Phytother Res. 2005. PMID 16114089. DOI 10.1002/ptr.1705. (abstract)
Schenk A, Siewert B, Toff S et al.. UPLC TOF MS for sensitive quantification of naturally occurring pyrrolizidine alkaloids in Petasites hybridus extract (Ze 339). J Chromatogr B Analyt Technol Biomed Life Sci. 2015. PMID 26094206. DOI 10.1016/j.jchromb.2015.05.027. (abstract)
Silva-Néto RP. Efficacy of Petasites hybridus in migraine prophylaxis: the first real-world study. Front Neurol. 2026. PMID 41908273. DOI 10.3389/fneur.2026.1784624. (abstract)
UK Medicines and Healthcare products Regulatory Agency. MHRA advice against unlicensed butterbur products (as reported by secondary sources). Reported by Medsafe, ConsumerLab and trade press. 2012. (summary-only)
Wildi E, Langer T, Schaffner W et al.. Quantitative analysis of petasin and pyrrolizidine alkaloids in leaves and rhizomes of in situ grown Petasites hybridus plants. Planta Med. 1998. PMID 17253240. DOI 10.1055/s-2006-957422. (abstract)

Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.