Coltsfoot: Regulation, Bans and 'PA-Free' Products

Most herbs are unregulated in the sense that matters to a shopper: nobody has assessed them, nobody has restricted them, and their safety status is a matter of opinion. Coltsfoot is different, and that difference is the most useful practical information on this whole site about the plant. Health authorities in several countries have looked directly at coltsfoot and acted — restricting it, refusing to license it for internal use, or capping tolerable pyrrolizidine-alkaloid intake so tightly that ordinary coltsfoot tea cannot comply.

That is rare. Regulators have limited resources and a very long list of plants; a herb reaches the top of the list only when the toxicological case is strong enough to survive scrutiny and pushback. So when you find that Germany restricted coltsfoot decades ago, that European herbal-medicine assessors declined to grant it a traditional-use monograph for internal use, and that the EU has since written pyrrolizidine-alkaloid limits into food law, you are looking at independent expert bodies reaching the same conclusion as the mechanism predicts. This page lays out that position country by country, then deals carefully and fairly with the "PA-free" question — which is a real idea, not marketing nonsense, and still not something a shopper can verify.

Table of Contents

1. Why Regulators Acted At All
2. Germany and Austria: The Restrictions
3. The European Union Position
4. The United Kingdom
5. The United States: Legal Is Not Endorsed
6. Canada, Australia and China
7. Why the Limits Are So Low
8. The Low-PA Cultivar: What Was Bred
9. What "PA-Free" Can and Cannot Mean
10. The Butterbur Precedent
11. How to Read a Product, If You Insist
12. Wild-Collected and Home-Grown Material
13. The Practical Bottom Line
14. Key Research Papers
15. Connections

Why Regulators Acted At All

Regulatory attention to pyrrolizidine alkaloids did not begin with coltsfoot or with herbal tea. It began with livestock and with famine.

Veterinary science understood PA poisoning early, because ragwort-contaminated pasture killed horses and cattle in identifiable outbreaks. Then came human episodes on a large scale: harvests contaminated with the seeds of PA-bearing weeds produced mass poisonings, with hundreds of cases of veno-occlusive liver disease and many deaths, in several parts of the world. Those events established beyond argument that these compounds cause a specific, severe human liver disease.

Attention turned to herbal medicine next, for an obvious reason: a person deliberately drinking a daily infusion of a PA-bearing plant is doing on purpose what the famine victims did by accident. By the 1980s, European authorities were reviewing PA-containing herbs as a class — comfrey, coltsfoot, borage, butterbur, the Senecio species — and the 1988 report of a fatal newborn case of hepatic veno-occlusive disease after maternal herbal tea consumption gave the review a human face.

The three findings that drove the regulatory conclusion are the ones set out on the liver-risk page: 1,2-unsaturated PAs are hepatotoxic, they are genotoxic, and several are carcinogenic in animals. The second of those is what makes the regulatory arithmetic unforgiving, and we explain why in Why the Limits Are So Low. (Evidence tier: established human outbreak epidemiology plus animal and mechanistic data.)

Germany and Austria: The Restrictions

Germany moved first and hardest, and German rules have shaped the European market ever since because Germany is the continent's largest herbal-medicine market.

In the early 1990s the German federal health authority issued a general restriction on medicinal products containing 1,2-unsaturated pyrrolizidine alkaloids. The structure of that decision is worth understanding, because it is more sophisticated than a simple ban:

• A very low daily ceiling on total PA intake for oral products — measured in the low single-digit micrograms per day, and permitted only for a short course of a few weeks per year rather than continuously.
• A tighter ceiling still for longer-term use.
• A considerably higher allowance for topical products on intact skin, since absorption is far lower — which is why comfrey ointment survives in European pharmacy while comfrey tea does not. See Comfrey: Never Take It Internally.
• An effective prohibition in pregnancy and lactation, on the basis that placental transfer and transfer into milk are established and no exposure is justifiable.

We are deliberately not quoting a precise figure for those ceilings, because the numbers have been revised over time and differ between the medicinal and food frameworks. The qualitative point is the one that matters and it is not in doubt: the permitted daily intake was set so low that ordinary, unpurified coltsfoot leaf or flower — whose alkaloid content is variable and not routinely assayed — cannot reliably be shown to comply. A limit you cannot demonstrate compliance with is, in practice, a prohibition on the traditional preparation.

Austria took a comparably restrictive line on coltsfoot products. Between them, the two German-speaking markets removed internal coltsfoot from mainstream pharmacy in the region where it had been most used. (Evidence tier: documented regulatory action.)

The European Union Position

The EU has approached pyrrolizidine alkaloids from two directions at once, and it is worth separating them because they answer different questions.

As a medicines question

The European Medicines Agency's herbal committee assesses whether a herbal substance can be granted a monograph — the document that underpins registration of traditional herbal medicinal products. The committee has published a public statement on herbal medicinal products containing toxic, unsaturated pyrrolizidine alkaloids, setting out limits on PA content in such products and the reasoning behind them. The practical consequence for coltsfoot is that internal coltsfoot has not been supported for traditional-use registration in the EU on safety grounds. Several other cough herbs — thyme, marshmallow root, ivy leaf, primrose root — do have monographs. Coltsfoot's absence from that company is not an oversight.

As a food-contaminant question

Separately, the European Food Safety Authority has issued scientific opinions on pyrrolizidine alkaloids in food and feed, and a later statement addressing PAs in honey, tea, herbal infusions and food supplements — the products where analytical surveys kept finding them. On the back of that work, the European Commission adopted a regulation setting maximum permitted levels of pyrrolizidine alkaloids in specified foods, including herbal infusions and teas, cumin, borage leaves and herbal food supplements, which came into application in the early 2020s.

Two features of that regulation deserve emphasis:

1. It applies to products not intended to contain PA plants at all. The problem being solved is contamination — PA-bearing weed fragments harvested alongside chamomile or peppermint, and honey from bees foraging on PA-bearing flowers. It exists because surveys found measurable PAs in perfectly ordinary supermarket tea.
2. It is expressed as a concentration limit on the product, which is a different instrument from a daily-intake ceiling but pushes in the same direction: total PA exposure down, across the whole diet.

Read together, the two strands say something clear. Europe treats pyrrolizidine alkaloids as contaminants to be minimised wherever they occur — and it declines to make an exception for a plant whose PAs are deliberate rather than accidental. (Evidence tier: documented regulatory action and published scientific opinions.)

The United Kingdom

The UK sits in an interesting middle position, and its history explains why coltsfoot is still culturally familiar there.

Coltsfoot appeared in the British herbal tradition and in the British Herbal Pharmacopoeia, and British Herbal Tobacco — a coltsfoot-based smoking mixture — was sold well into the twentieth century. Coltsfoot rock is still made. So the plant has a living presence in British confectionery and folk memory.

The regulatory reality is more restrictive than that memory suggests. UK medicines regulation follows the same scientific assessments of unsaturated pyrrolizidine alkaloids, no internal coltsfoot product holds a traditional herbal registration, and the pyrrolizidine-alkaloid limits that apply to teas and herbal infusions constrain the market for PA-bearing material regardless of the plant's cultural standing. A sweet flavoured with coltsfoot extract and a daily therapeutic infusion of coltsfoot leaf are different propositions, and it is the second one that the framework rules out. (Evidence tier: documented regulatory position.)

The United States: Legal Is Not Endorsed

The United States is where a reader is most likely to be able to buy coltsfoot easily, and it is important to understand what that availability does and does not mean.

Under the US dietary-supplement framework, a botanical marketed before the mid-1990s can generally remain on the market without pre-market safety approval. The regulator's power is largely reactive: it acts against a product after a problem is documented. So the fact that dried coltsfoot leaf is on a shelf carries no implication whatever that anyone assessed it and cleared it. It means nobody removed it.

Meanwhile the substantive US position is much closer to the European one than the shelf suggests:

• Federal health-information sources and the standard clinical toxicology references flag PA-containing herbs, coltsfoot included, as hepatotoxic and advise against internal use.
• The trade's own safety classification systems place coltsfoot among botanicals not recommended for internal use, with pregnancy and lactation contraindicated.
• Poison-control and hepatology literature treat herbal PA exposure as a recognised cause of drug-induced liver injury, worth asking about when unexplained liver disease presents.

The gap between "legal to sell" and "considered safe" is the single most misread thing in American herbal retail. Coltsfoot is a clean example of it. (Evidence tier: documented regulatory framework and published expert guidance.)

Canada, Australia and China

• Canada regulates herbal products as natural health products requiring pre-market licensing, and applies restrictions to unsaturated pyrrolizidine alkaloids in oral products. The effect is that a coltsfoot oral product faces a licensing hurdle it is unlikely to clear.
• Australia likewise operates a pre-market listing framework with restrictions on pyrrolizidine alkaloids in medicines, following the same toxicological assessments.
• China is the important contrast. Kuan dong hua — the coltsfoot flower bud — remains in the Chinese Pharmacopoeia and in clinical use within traditional Chinese medicine, generally honey-fried and generally within multi-herb formulas rather than alone. Chinese pharmacological and toxicological researchers have themselves published extensively on the plant's pyrrolizidine alkaloids and on the quality-control problem they create; the 2021 comprehensive review of the species came out of that research community. So this is not a case of a tradition ignorant of the issue. It is a case of two regulatory cultures weighing a long-established medicinal material differently.

Being straightforward about that divergence is better than pretending it away. A reader deciding what to do should know that the plant is a pharmacopoeial medicine in China and effectively off the internal market in German-speaking Europe. Our own reading, stated plainly, sides with the restriction — because the cough benefit is unproven and interchangeable while the alkaloid hazard is established, so there is nothing being lost that cannot be got elsewhere. (Evidence tier: documented regulatory positions.)

Why the Limits Are So Low

People often assume a toxin limit is set just below the dose that causes harm, with a safety margin. For PAs that is not how it works, and the reason is worth understanding because it explains why the ceilings look absurdly small.

Unsaturated pyrrolizidine alkaloids are genotoxic: their reactive metabolites bind DNA. For a genotoxic carcinogen, regulators generally do not assume that a threshold exists below which risk is zero, because a single unrepaired DNA lesion is in principle sufficient to initiate a lineage. Instead of identifying a safe level, assessors estimate the exposure associated with a very small theoretical excess risk and recommend keeping intake as low as reasonably achievable.

That approach has two consequences that matter to a herbal-tea drinker:

1. The resulting figures are tiny — micrograms per day, sometimes fractions of one — and are therefore easily exceeded by a single cup of an infusion made from a plant that manufactures the compound on purpose.
2. "Low PA" is not the same category as "safe." A product with a small but real alkaloid content sits on the same continuum, just further down it. This is the honest reason why a low-PA cultivar, discussed next, is an improvement rather than a solution.

Anyone who wants to argue that the regulatory limits are over-cautious has to argue with the genotoxicity data, not with the arithmetic. (Evidence tier: established risk-assessment methodology; the underlying genotoxicity is established laboratory data.)

The Low-PA Cultivar: What Was Bred

Because coltsfoot has genuine traditional standing, European plant breeders tried to solve the problem at its source rather than abandon the plant. They screened coltsfoot populations for alkaloid content and selected and registered a low-alkaloid cultivar in which pyrrolizidine alkaloids are at or below the limit of detection by the assays used. Cultivated clonally, such a line can be propagated true to type.

This deserves to be taken seriously, and dismissing it would be as wrong as trusting it blindly:

• The concept is sound. If the hazard is one class of compound and a plant line does not make it, removing the compound removes the hazard. That is not a hand-wave; it is the same logic behind low-erucic-acid rapeseed and low-tetrahydrocannabinol hemp, both of which worked.
• It concedes the toxicology. Nobody breeds out a compound they consider harmless. The existence of the cultivar is itself evidence that serious people regard ordinary coltsfoot as unsuitable for internal use.
• It is what makes any tolerated internal product possible in jurisdictions with strict PA ceilings.

And here are the honest limits:

• "Below the limit of detection" depends on the method. Analytical sensitivity has improved substantially; material that was undetectable on an older method may not be on a modern one.
• A cultivar is a claim about genetics, not about the bag in your hand. The bag's contents depend on what was actually planted, harvested, stored and shipped — and on whether anything else got in.
• Cultivar identity is unverifiable by the buyer. You cannot distinguish low-PA coltsfoot from ordinary coltsfoot by look, smell or taste. Nobody can, without a laboratory.
• Very little retail coltsfoot is the cultivar. Cheap bulk dried herb, foraged material and imported flower bud are ordinary, alkaloid-containing coltsfoot.

(Evidence tier: documented plant-breeding work; the practical availability caveats are commercial observation, not laboratory data.)

What "PA-Free" Can and Cannot Mean

Precision matters here, because the same two words are used for three quite different things.

1. PA-free by cultivar — the plant line does not produce detectable alkaloid. Depends entirely on true-to-type propagation and on no admixture in the supply chain.
2. PA-depleted by manufacturing — an extract from ordinary plant material, processed to remove alkaloids, with each batch assayed to confirm they are below a specified limit. This is the strongest version, and it requires an industrial process plus batch-release testing.
3. "PA-free" as a marketing phrase — printed on a package with nothing behind it. Nothing prevents this in a market where such claims are not pre-approved.

The trouble is that all three look identical on a label. A shopper cannot tell them apart, which means the phrase carries no reliable information at the point of purchase. So:

• Do not treat a label as a test result. A certificate of analysis for your batch, from an independent laboratory, using a current method, naming the alkaloids measured and their limits of quantification, is a test result. Everything else is a statement of intent.
• Loose-leaf herb is the weakest case of all. Bulk dried leaf or flower has no batch identity a buyer can trace and no realistic prospect of assay.
• Admixture and misidentification remain live risks even for a genuine low-PA line, and the 1988 infant case is exactly what that failure mode looks like when it goes wrong.

To be fair to the concept: a properly manufactured, batch-assayed PA-depleted extract does remove the principal hazard, and it would be dishonest to pretend otherwise. The problem is not that PA removal is impossible. It is that for coltsfoot no such product is meaningfully available to consumers, and the thing that is available — dried leaf, dried flower bud, tincture, home-made tea — is not it.

The Butterbur Precedent

The closest real-world test of the PA-depletion idea is not coltsfoot but butterbur (Petasites hybridus), and its history is instructive in both directions.

Butterbur has a genuine evidence base for migraine prevention and for allergic rhinitis — better than coltsfoot's for cough, with randomised trials behind it. It also contains pyrrolizidine alkaloids. The response was to develop standardised PA-depleted extracts, manufactured to remove the alkaloids and assayed batch by batch, and those depleted extracts are what the clinical trials used.

Two lessons follow, and both matter for coltsfoot:

• PA depletion can be done properly, and when it is, a PA-bearing plant can support a legitimate medicinal product. This is the honest case in favour of not dismissing "PA-free" out of hand.
• Even so, it did not end the safety story. Butterbur products subsequently attracted regulatory attention over reports of liver injury, and some were suspended or withdrawn in European markets. Whether every case reflected residual alkaloids, another constituent, or an unrelated cause has been debated — which is precisely the point. Manufacturing control reduced a known hazard; it did not deliver certainty.

Our own pages work through this in detail: Butterbur: Pyrrolizidine Alkaloids and Liver Safety and Choosing a PA-Free Product. If you read one thing alongside this page, read the second of those, because the product-evaluation logic transfers directly.

The asymmetry with coltsfoot is the decisive part. Butterbur's depleted extracts were developed because butterbur had a demonstrated benefit worth preserving. Coltsfoot has no demonstrated benefit to preserve — see Coltsfoot for Cough: The Tradition and the Evidence — which is very likely why no comparable industry effort exists for it. (Evidence tier: documented regulatory and clinical history.)

How to Read a Product, If You Insist

This site's position is that internal coltsfoot is not worth taking. But readers make their own decisions, and a reader who has decided otherwise is better served by knowing how to evaluate a product than by being refused the information. So, in descending order of what actually protects you:

1. A batch-specific certificate of analysis. Independent laboratory, current method, named alkaloids including senkirkine and senecionine, stated limits of quantification, matched to the lot number on your package. Nothing else in this list is a substitute.
2. A licensed or registered medicinal product in a jurisdiction with PA limits, rather than a food supplement or bulk herb. Licensing implies someone reviewed the specification.
3. Named cultivar and documented provenance, with the supplier able to say where it was grown and by whom.
4. Botanical authentication — evidence that the material is Tussilago farfara and not a look-alike. This is the failure mode that killed an infant.
5. Short duration. Days, not weeks, and not repeated seasonally. Cumulative dose is the whole risk.
6. No stacking. Not alongside comfrey, borage, butterbur that is not certified depleted, or unidentified herbal blends.
7. Absolute exclusions, regardless of any certificate: pregnancy, breastfeeding, children, any liver condition, heavy alcohol use, and multiple concurrent medications.

And the thing that is not reassurance, however much it feels like it: "organic," "wildcrafted," "traditional," "grandmother drank it," a pretty package, a herbalist's recommendation, or the word "natural." None of these speaks to alkaloid content. "Wildcrafted" is actively worse than cultivated, because wild material is unselected and unverified.

Wild-Collected and Home-Grown Material

Coltsfoot is easy to find and easy to grow, which makes foraging tempting and makes this the highest-risk route of all.

• Wild plants are ordinary coltsfoot. They are not the low-PA cultivar, and their alkaloid content is unknown and variable with part, season, stage and provenance.
• Misidentification is a real hazard. Coltsfoot can be confused with, or grow alongside, other composites — butterbur, Adenostyles, various Senecio weeds — several of which carry higher alkaloid loads. Early in the season, when only the flower stalks are up, the diagnostic hoof-shaped leaves are not even present.
• Roadside and industrial-margin collection adds unrelated contamination — vehicle-derived metals and hydrocarbons, herbicide residues — on top of the alkaloids. Coltsfoot's favourite habitats are exactly these disturbed margins.
• Home-grown material from ordinary seed or wild-dug rhizome is ordinary coltsfoot too. Growing it yourself controls the soil, not the genetics.

Coltsfoot is a genuinely lovely early-spring plant and a useful early nectar source. Enjoying it in the ground is entirely safe. It is putting it in the teapot that is the problem.

The Practical Bottom Line

• Germany and Austria restricted coltsfoot for internal use decades ago; its permitted alkaloid intake is set so low that ordinary coltsfoot cannot demonstrate compliance. (Documented.)
• EU herbal-medicine assessment has not supported internal coltsfoot, while several other cough herbs did receive monographs. (Documented.)
• The EU has written maximum pyrrolizidine-alkaloid levels into food law for teas, herbal infusions and supplements. (Documented.)
• US availability reflects a reactive regulatory framework, not a safety assessment. (Documented.)
• A low-PA cultivar exists and the concept is legitimate — but cultivar identity is unverifiable by a buyer and most retail coltsfoot is not it. (Documented breeding; commercial caveat.)
• Batch-assayed PA depletion genuinely works, as butterbur shows — and no such coltsfoot product is meaningfully available. (Documented.)
• Nothing about the regulatory picture argues in coltsfoot's favour. Every jurisdiction that examined it closely restricted it.

If you came here wanting a cough herb rather than an argument, Safer Alternatives to Coltsfoot for Cough is the page to read next. Several of the options there hold the European monographs coltsfoot could not get.

Key Research Papers

Each link runs a live PubMed topic search rather than pointing at a single record. Regulatory documents are linked to their agencies' landing pages under External Resources.

1. Regulatory limits and risk management for pyrrolizidine alkaloids in herbal medicinal products — how the European limits were derived and what they require of manufacturers. Search PubMed
2. Determination of pyrrolizidine alkaloids in teas, herbal drugs and honey — the analytical surveys that put PAs onto food-safety agendas. Search PubMed
3. Detection of pyrrolizidine alkaloids in licensed herbal medicinal teas — findings in products already on regulated markets. Search PubMed
4. Risk assessment of pyrrolizidine alkaloids in food of plant and animal origin — the exposure modelling behind the maximum levels. Search PubMed
5. A review of the ethnobotanical value, phytochemistry, pharmacology, toxicity and quality control of Tussilago farfara L. (coltsfoot), Journal of Ethnopharmacology, 2021 — includes the quality-control problem that regulation is responding to. Search PubMed
6. Breeding and selection of low-pyrrolizidine-alkaloid coltsfoot and comfrey lines — the plant-breeding response to the toxicology. Search PubMed
7. PA-depleted Petasites hybridus (butterbur) extracts in migraine prophylaxis — the trials that used depleted extract, and the model for doing depletion properly. Search PubMed
8. Hepatotoxicity reports associated with butterbur products and the subsequent regulatory response — why depletion reduced but did not end concern. Search PubMed
9. Hepatic veno-occlusive disease in newborn infant of a woman drinking herbal tea, The Journal of Pediatrics, 1988 — the case that gave the European reviews a human face, including the plant-identification problem. Search PubMed
10. Herb-induced liver injury: identification, reporting and adulteration — the clinical-pharmacovigilance context in which PA herbs are assessed. Search PubMed
11. Botanical authentication and adulteration of herbal raw material by DNA barcoding — how misidentification is detected, and how often it is found. Search PubMed
12. Genotoxic carcinogens and the margin-of-exposure approach in food safety — why PA limits are set without assuming a safe threshold. Search PubMed
13. Pyrrolizidine alkaloid content variation by plant part, harvest stage and provenance — the analytical basis for saying content is unpredictable. Search PubMed

External Resources

• European Medicines Agency — herbal monographs and the committee statement on herbal products containing unsaturated pyrrolizidine alkaloids.
• European Food Safety Authority — scientific opinions on pyrrolizidine alkaloids in food, feed, honey, tea and supplements.
• German Federal Institute for Risk Assessment (BfR) — much of the analytical and risk-assessment work behind the European limits.
• BfArM (German Federal Institute for Drugs and Medical Devices) — the German medicines authority.
• UK MHRA — UK herbal-medicine registration and safety.
• US Food and Drug Administration — the US dietary-supplement framework.
• Health Canada — natural health product licensing.
• Australian Therapeutic Goods Administration — listed-medicine requirements.

Connections

• All Herbs
• Coltsfoot
• Coltsfoot Benefits Hub
• PA Liver Risk
• Cough Tradition and Evidence
• Safer Alternatives for Cough
• Butterbur
• Butterbur PA Liver Safety
• Choosing a PA-Free Product
• Comfrey
• Comfrey: Never Internally
• Comfrey Benefits
• Liver Disease
• Liver Function Tests
• Toxins

Safety Note and Disclaimer

This page describes regulatory positions in general terms for education; it is not legal advice, and the details of any framework change over time — check the agency landing pages above for the current text. Nothing here is a recommendation to take coltsfoot. A product being legal to sell where you live is not evidence that it is safe, and a "PA-free" claim on a package is not a test result. Do not take coltsfoot internally if you are pregnant or breastfeeding, do not give it to a child, and do not take it if you have any liver condition, drink heavily, or take prescription medicines. If you have been taking coltsfoot or another pyrrolizidine-alkaloid herb and develop right-upper-abdominal pain, abdominal swelling, jaundice or unexplained fluid weight gain, seek medical attention promptly and say exactly what you have been taking.

Back to Table of Contents