Supplements
Butterbur: extract-specific findings, liver concerns and product uncertainty
On this page
Butterbur usually means Petasites hybridus. Migraine studies used root extracts; hay-fever studies used leaf extracts. A plant name does not identify a tested finished product.[1]
Separate the plant, intended constituents and unwanted constituents
| Term | Meaning |
|---|---|
| Root / leaf extract | Different plant parts and preparations used in migraine and hay-fever research.[1] |
| Petasins | Constituents measured in butterbur products; their presence does not establish absence of unwanted alkaloids.[5] |
| Pyrrolizidine alkaloids (PAs) | Naturally present compounds that can damage the liver and lungs and may cause cancer.[1] |
| PA-free claim | A processing/quality claim requiring attention; it is not proof of clinical benefit.[5] |
A PA-free label does not settle every safety question
NCCIH reports rare liver injury involving products described as PA-free. It advises considering only products processed to remove PAs and labeled or certified accordingly. The American Academy of Neurology stopped recommending butterbur in 2015 because of safety concerns. A raw plant or tea does not substitute for a researched extract.[1]
Check the warning that applies where you live
MHRA’s 2021 reminder says its advice not to take unlicensed butterbur remedies remains unchanged. That is UK regulatory advice, not a statement that a foreign label establishes UK authorization. Read the relevant patient information and discuss the actual product with a clinician or pharmacist.[6]
Keep enough detail to assess a suspected problem
MHRA advises recording the product name, ingredients, manufacturer, intended use, other medicines and a photograph of labeling when reporting a suspected reaction. A report is a safety signal rather than proof of causation. Clinical assessment and reporting serve different purposes; do not wait for certainty before seeking advice about a suspected reaction.[6]
Evidence & uncertainty
The migraine trial evaluated a standardized root extract in adults over four months. In the per-protocol analysis, its higher study arm improved attack frequency versus placebo; the lower arm did not significantly improve the primary endpoints. This is evidence about that preparation and preventive outcome, not a class-wide guarantee or an acute headache treatment.[2]
Selected-source overview · Independent review pendingPublished questions
Can a migraine-prevention result show that butterbur stops a headache already underway?
Selected-source overview · Independent review pending. Read the evidence context.
No. The trial assessed attack frequency over months. It did not test an acute rescue treatment, so that is a different clinical question.[2]
Evidence behind this answer
- Petasites hybridus root (butterbur) is an effective preventive treatment for migraine
randomized placebo controlled trial · 2004
Complete original abstract freshly reread on author institution record; full paper not independently appraised. · Complete author-institution original abstract: 245 adults, four months, standardized root extract, preventive frequency endpoint and differing trial arms
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Can I substitute any butterbur leaf, root or tea for the studied product?
Selected-source overview · Independent review pending. Read the evidence context.
No. The migraine evidence concerns a standardized root extract. A differently prepared plant product does not inherit that trial’s benefit or short-term tolerability findings.[2]
Evidence behind this answer
- Petasites hybridus root (butterbur) is an effective preventive treatment for migraine
randomized placebo controlled trial · 2004
Complete original abstract freshly reread on author institution record; full paper not independently appraised. · Complete author-institution original abstract: 245 adults, four months, standardized root extract, preventive frequency endpoint and differing trial arms
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Does one allergy-pill comparison prove all butterbur products work equally well?
Selected-source overview · Independent review pending. Read the evidence context.
No. That two-week study tested ZE 339 and selected rating scales. Its active-comparator design cannot validate every product or establish long-term equivalence.[3]
Evidence behind this answer
- Randomised controlled trial of butterbur and cetirizine for treating seasonal allergic rhinitis
randomized active comparator trial · 2002
Original full text recovered through Firecrawl after publisher 403; complete relevant methods, efficacy/safety, methodological discussion and funding sections read. · 125 adults, two weeks, ZE 339, SF-36/global ratings, noninferiority design, harms and manufacturer sponsorship
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Were all placebo-controlled hay-fever findings favorable?
Selected-source overview · Independent review pending. Read the evidence context.
No. The 35-person crossover study found no significant improvement in its airflow, symptom or quality-of-life outcomes versus placebo. Product and endpoint details matter.[4]
Evidence behind this answer
- Effects of butterbur treatment in intermittent allergic rhinitis: a placebo-controlled evaluation
randomized placebo controlled crossover trial · 2004
Complete original publisher abstract read through Firecrawl after web reader 403; full subscription paper not read. · Complete original publisher abstract: 35 people, two weeks each, nasal airflow/symptoms/quality of life null versus placebo
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Does a petasin content claim also prove a product contains no PAs?
Selected-source overview · Independent review pending. Read the evidence context.
No. The analytical study assessed petasin content and PAs separately. Meeting an active-constituent label claim is not the same as meeting an unwanted-constituent specification.[5]
Evidence behind this answer
- Simultaneous determination of sesquiterpenes and pyrrolizidine alkaloids from the rhizomes of Petasites hybridus and dietary supplements using UPLC-UV and LC-TOF methods
original product analytical study · 2012
Complete original author research record, interpretive summary and technical abstract read on USDA ARS; full journal article not independently appraised. · USDA author research record: complete interpretive summary and technical abstract; 21 supplement samples; seven met both petasin-label and no-detected-PA criteria
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Does a PA-free label guarantee no liver injury?
Selected-source overview · Independent review pending. Read the evidence context.
No. NCCIH notes rare liver injury reports involving products described as PA-free. A label cannot establish an individual’s long-term safety.[1]
Evidence behind this answer
- Butterbur: Usefulness and Safety
agency herbal safety guidance · 2024
Original official source; relevant complete sections read, not search snippets. · Root versus leaf research, PAs, reported PA-free liver injury, allergy, medicines, pregnancy/breastfeeding and topical limits
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Is PA-free butterbur established as safe during pregnancy or breastfeeding?
Selected-source overview · Independent review pending. Read the evidence context.
No. NCCIH says safety information is limited for PA-free products; PA-containing products should not be used in pregnancy or breastfeeding.[1]
Evidence behind this answer
- Butterbur: Usefulness and Safety
agency herbal safety guidance · 2024
Original official source; relevant complete sections read, not search snippets. · Root versus leaf research, PAs, reported PA-free liver injury, allergy, medicines, pregnancy/breastfeeding and topical limits
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Can someone allergic to ragweed assume butterbur will be suitable for hay fever?
Selected-source overview · Independent review pending. Read the evidence context.
No. NCCIH warns of possible allergic reactions in people allergic to related plants, including ragweed. An allergy indication does not remove that concern.[1]
Evidence behind this answer
- Butterbur: Usefulness and Safety
agency herbal safety guidance · 2024
Original official source; relevant complete sections read, not search snippets. · Root versus leaf research, PAs, reported PA-free liver injury, allergy, medicines, pregnancy/breastfeeding and topical limits
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
What should I bring to a discussion about butterbur and my medicines?
Selected-source overview · Independent review pending. Read the evidence context.
Bring the actual product, ingredient list and all medicines. MHRA emphasizes a full medicine history when assessing suspected herbal-product reactions, rather than judging compatibility from the plant name.[6]
Evidence behind this answer
- Herbal and homeopathic medicines: reminder to be vigilant for suspected adverse reactions and to report them to the Yellow Card scheme
agency herbal product safety warning · 2021
Original official source; relevant complete sections read, not search snippets. · Butterbur warning remains unchanged; patient information, adverse reactions, full medicine list and report identification details
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Does a four-month trial establish safety for years of use?
Selected-source overview · Independent review pending. Read the evidence context.
No. Its duration and selected participants limit that inference. The migraine study’s tolerability findings cannot rule out uncommon or longer-term harms.[2]
Evidence behind this answer
- Petasites hybridus root (butterbur) is an effective preventive treatment for migraine
randomized placebo controlled trial · 2004
Complete original abstract freshly reread on author institution record; full paper not independently appraised. · Complete author-institution original abstract: 245 adults, four months, standardized root extract, preventive frequency endpoint and differing trial arms
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Can I report a suspected reaction without knowing that butterbur caused it?
Selected-source overview · Independent review pending. Read the evidence context.
Yes. MHRA encourages suspected-reaction reports even when information is incomplete. Include product details and other medicines; assessment can then consider alternative causes.[6]
Evidence behind this answer
- Herbal and homeopathic medicines: reminder to be vigilant for suspected adverse reactions and to report them to the Yellow Card scheme
agency herbal product safety warning · 2021
Original official source; relevant complete sections read, not search snippets. · Butterbur warning remains unchanged; patient information, adverse reactions, full medicine list and report identification details
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Prevention and relief of an attack are different goals
The trial randomized 245 adults. Its four-month per-protocol analysis reported attack-frequency reductions of 48% with the higher arm, 36% with the lower arm and 26% with placebo; only the higher arm differed significantly from placebo. Per-protocol results concern participants meeting protocol criteria, rather than every randomized participant. Reported related harms were mainly mild gastrointestinal symptoms, especially burping.[2]
A short antihistamine comparison has a defined scope
A two-week trial in 125 adults compared ZE 339 with cetirizine using quality-of-life and global ratings. It found similar improvement under a noninferiority design, with no untreated placebo group. Zeller supplied the study medication and sponsored the trial. These findings do not establish that every leaf or root product works like an antihistamine.[3]
Not every rhinitis trial was positive
A separate two-week crossover study in 35 people found no significant benefit versus placebo in nasal airflow, nasal/eye symptoms or quality of life. Different products and endpoints make simple equivalence claims difficult. The published abstract does not establish that its preparation was identical to ZE 339.[4]
Chemical checks answer a different question from symptom trials
In a 2012 analysis of 21 supplements claiming butterbur, only seven both matched labeled petasin amounts and showed no detected PAs. That does not mean every remaining sample contained PAs: those were two criteria. The historical sample cannot rank today’s brands or establish a current contamination rate.[5]
References
- Butterbur: Usefulness and Safety
Agency herbal safety guidance
NCCIH, US National Institutes of Health · Content/publication date: 2024
Accessed: 2026-10-20 UTC · Original official source; relevant complete sections read, not search snippets.
Cited sections: Root versus leaf research, PAs, reported PA-free liver injury, allergy, medicines, pregnancy/breastfeeding and topical limits - Petasites hybridus root (butterbur) is an effective preventive treatment for migraine
Randomized placebo controlled trial
Lipton and colleagues · Content/publication date: 2004
Accessed: 2026-10-20 UTC · Complete original abstract freshly reread on author institution record; full paper not independently appraised.
Cited sections: Complete author-institution original abstract: 245 adults, four months, standardized root extract, preventive frequency endpoint and differing trial arms - Randomised controlled trial of butterbur and cetirizine for treating seasonal allergic rhinitis
Randomized active comparator trial
Schapowal and Petasites Study Group · Content/publication date: 2002
Accessed: 2026-10-20 UTC · Original full text recovered through Firecrawl after publisher 403; complete relevant methods, efficacy/safety, methodological discussion and funding sections read.
Cited sections: 125 adults, two weeks, ZE 339, SF-36/global ratings, noninferiority design, harms and manufacturer sponsorship - Effects of butterbur treatment in intermittent allergic rhinitis: a placebo-controlled evaluation
Randomized placebo controlled crossover trial
Gray and colleagues · Content/publication date: 2004
Accessed: 2026-10-20 UTC · Complete original publisher abstract read through Firecrawl after web reader 403; full subscription paper not read.
Cited sections: Complete original publisher abstract: 35 people, two weeks each, nasal airflow/symptoms/quality of life null versus placebo - Simultaneous determination of sesquiterpenes and pyrrolizidine alkaloids from the rhizomes of Petasites hybridus and dietary supplements using UPLC-UV and LC-TOF methods
Original product analytical study
Avula and colleagues · Content/publication date: 2012
Accessed: 2026-10-20 UTC · Complete original author research record, interpretive summary and technical abstract read on USDA ARS; full journal article not independently appraised.
Cited sections: USDA author research record: complete interpretive summary and technical abstract; 21 supplement samples; seven met both petasin-label and no-detected-PA criteria - Herbal and homeopathic medicines: reminder to be vigilant for suspected adverse reactions and to report them to the Yellow Card scheme
Agency herbal product safety warning
UK Medicines and Healthcare products Regulatory Agency · Content/publication date: 2021
Accessed: 2026-10-20 UTC · Original official source; relevant complete sections read, not search snippets.
Cited sections: Butterbur warning remains unchanged; patient information, adverse reactions, full medicine list and report identification details
Scope & review
- Population
- Adults considering oral butterbur products for migraine prevention or allergic rhinitis; pregnancy and allergy safety limits noted
- Exposure or intervention
- Specified Petasites hybridus root/leaf preparations, petasins and PA-removal claims
- Comparison
- Extract-specific placebo or active-comparator research versus generic plant products and unverified market labels
- Outcomes
- Migraine frequency, short-term hay-fever measures, product identity and reported harms
- Timeframe
- Two-week rhinitis studies and four-month migraine study versus uncertain long-term use
- Jurisdiction
- International trials, US NCCIH and UK MHRA guidance; no individualized regimen, treatment guarantee or universal licensing claim
Bounded source-based educational explanation. Original trial or analytical-study abstracts, selected complete full-text sections and authoritative guidance were read with access limits recorded. No exhaustive search, independent synthesis, formal evidence grade or qualified medical/pharmacology review has been completed.
Sources checked on their individual access dates. Selected sources were checked; ongoing surveillance and qualified review have separate statuses on the coverage page.
Revision history
- 2026-10-20 UTC: Prepared definition-first draft with claim-linked sources for the authorized fifth batch.
- 2026-10-20 UTC: After separate bounded source-fit review, specified the migraine result’s per-protocol analysis and added the safety-related 2015 withdrawal of the AAN recommendation, as reported by current NCCIH guidance.
- 2026-10-20 UTC: Restored actual migraine arm results in evidence section with explicit per-protocol rather than all-randomized scope; no additional FAQs or sources.
What this article still leaves open
- How reliably do current products document identity and PA removal across batches?
- What independent, longer-term evidence can separate extract effects, contamination and other causes of reported liver injury?
Suggest a correction
Submit a content correction to the editorial queue, or download a local note. Leave out personal medical details. Submitted text is stored privately with a browser receipt session and is visible to this browser and the editor. Clearing cookies removes access to your submission history. This is not an account or a medical service.
Your submitted corrections for this article
Explore the topic map
Browse the health domains connected to this article. The map also identifies topics whose articles remain unwritten.