Feverfew – the aspirin of the middle ages

by Ben Francis

Feverfew flowers Photo- the authorIn what is now a bygone era, the feverfew was perhaps one of the most important medicinal plants in Britain. A perennial that will thrive just about anywhere, “once [it] has a hold, it persists in spite of weeding and contempt” (Grigson, 1975). So there’ll always be a good opportunity to see it in Mecklenburgh Square Garden. Originally native to Eastern Europe and Western Asia, feverfew has since travelled around the globe. In Britain the plant was once used either fresh or dried for pain relief from ailments such as headaches or arthritis, for reducing a fever (from which it gets its name), as a treatment for psoriasis, for treating several gynaecological problems, as a treatment for intestinal worms, and as an insect repellent (Pareek et al., 2011). These days it’s sold as capsules in pharmacies and health food shops as a treatment for migraine, and far less commonly, as a treatment for rheumatoid arthritis. Additionally, research investigating the primary chemical constituent of feverfew, a molecule called parthenolide, as a treatment for several types of cancer, is ongoing, but is still at early stages.. Currently the data on feverfew as a migraine remedy is mixed, with only some studies finding it to be more effective than a placebo. There’s no information readily available about long term toxicity from using feverfew, but it appears not to have adverse effects at doses recommended by product manufacturers, although the fresh plant has been known to cause mouth ulcers when chewed.

Feverfew flowers (above). Photo: the author.

Feverfew leaf (below). Photo: Steffen Heinz Caronna, distributed under a CC-BY-3.0 licence

FeaverfewLeafFeverfew is easily recognised by its clusters of ‘daisy-like’ white and yellow flowerheads during the summer months, its characteristic pinnate ‘feathery’ leaves arranged alternately along the stem, and its strong musty and bitter fragrance. Go on; get a good whiff when you see it. Its weedy habit and utility as a medicine were probably instrumental in allowing it to spread from its place of origin first to Western Europe and then beyond, following the movements of people. Formally known today as Tanacetum parthenium (L.) Sch.Bip. (Asteraceae), feverfew was historically known by many interesting vernacular names in around Britain, including arsesmart in Yorkshire, bachelor’s buttons in Somerset, nosebleed in Kent, and more widely, featherfew (Grigson, 1975). The name feverfew itself is a corruption of the Latin word febrifugia, meaning something that drives away fevers (Hobbs, 1989). It’s this name that indicates the plant’s medicinal use in what might be considered the pre-modern era of medicine prior to the 20th century. And indeed it was integral to healthcare in Britain for centuries, as “a medieval aspirin” (Heptinstall, 1988).

Historic and modern uses

As mentioned, the main uses for feverfew were once for pain relief and reduction of fevers. As more effective pain medicines were developed around the turn of the twentieth century, particularly opiates and aspirin, feverfew was largely sidelined, as were most traditional British herbal remedies. This is equally true with regard to its fall from favour as a result of developments in treatments for arthritis and psoriasis, and especially the considerable improvements in reproductive medicine that produced much safer and more effective abortions than those that used to be induced using feverfew.

After a period of relative obscurity, feverfew began to attract attention in the 1970s as a possible prophylactic treatment for persistent migraine. A few individual published reports led to wider interest and commercialisation of both extracts of feverfew and of dried herbal material. As one of the more popular herbal remedies today, it is cultivated on a large scale for this purpose and machine harvested (MHRA product assessment report, DiaMigraine/DiaFeverfew hard capsules). Since the introduction of European Union legislation on the regulation of traditional herbal remedies in 2004, products may be purchased that have been manufactured according to stringent codes of practice, and which are standardised to contain known amounts of parthenolide.

Parthenolide itself has generated considerable interest as a possible future treatment for blood and lymph cancers (Ghantous et al., 2010). In a chemically modified form, it has reached the first stage of human trials for these cancers; these so called phase 1 clinical trials establish safe doses of a drug, along with its absorption and distribution through the human body. These trials typically take years to complete because of the small numbers of patients involved.


Feverfew. Drawing: Franz Eugen Köhler, 1897; public domain.

Scientific evidence

Parthenolide is the most abundant of more than thirty compounds present in feverfew leaves that belong to a chemical class called sesquiterpene lactones. These compounds are often potently biologically and pharmacologically active, and they most likely exist to protect plants from herbivores. Examples of sesquiterpene lactones from other plant species have also attracted research interest as cancer treatments. Parthenolide itself however, is especially effective at targeting and killing tumours, having been shown in laboratory studies to induce programmed cell death (apoptosis) in both leukaemia (Guzman & Jordan, 2005) and melanoma (Lesiak et al., 2010) cells.

It seems parthenolide might work in a number of ways to selectively target cancer cells whilst doing minimal damage to healthy tissues, something all cancer drugs must do. Primarily it appears to act by inhibition of a protein called Nuclear Factor – kappa-B (NF – κB), which is present in almost all cells in the human body, is involved in cellular stress responses (Brasier, 2006; Hehner et al., 1998), and which is also hyperactivated in cancer cells as compared to normal cells. Also important is its ability to maintain production of another protein called p53, which is involved in causing apoptosis, and which is inactivated in many cancers (Ghantous et al., 2010; 2013). It has also been demonstrated that parthenolide has anti-angiogenic activity, that is it prevents the formation of blood vessels, and so it might prevent tumours from recruiting their own blood supply, thus reducing their ability to grow (Ghantous et al., 2013). Interestingly, parthenolide appears to also target what are known as cancer stem cells, which are the cells considered to be involved in the initiation and subsequent spread of tumours to other locations in the body (Ghantous et al., 2013, and references therein).

Since it was once a treatment for pain and other conditions such as fever and psoriasis, it has been hypothesised that feverfew acts in a similar manner to modern pain medicines and antipyretics (fever reducers) such as ibuprofen and aspirin. These medicines are known as non-specific anti-inflammatory drugs (NSAIDs). It has been shown that feverfew has a similar anti-inflammatory action, with parthenolide demonstrating inhibition of the cellular processes that lead to inflammation. Specifically it has been shown that this is again a result of its regulation of NF – κB. Parthenolide acts by preventing NF-κB from being activated and thus ‘switching on’ genes whose expression is involved in causing the inflammatory response (Hehner et al., 1998; Baker et al., 2011).

Feverfew extracts have also been shown to inhibit the synthesis of another group of molecules called prostaglandins, which are also important in the inflammatory process. Other NSAIDs are known to inhibit prostaglandin synthesis as part of their therapeutic effect (Heptinstall, 1988). Meanwhile it has been suggested that a third mechanism contributes to feverfew’s use as a treatment for arthritis and psoriasis, specifically by preventing the release of certain enzymes from white blood cells either in joints or the skin respectively (Pareek et al., 2011).

As for its use as a treatment for migraine, it has been demonstrated that feverfew inhibits release of the neurochemical serotonin into the bloodstream, and also that changes in blood serotonin levels are associated with the incidence of migraine (Pareek et al., 2011; Humphrey et al., 1990).

While parthenolide is generally considered to be the most important active compound in feverfew leaves, also present are a number of other molecules that may contribute to the plant’s medicinal properties. For example, feverfew leaves contain several members of a large class of plant metabolites called flavonoids. Some flavonoids isolated from feverfew have shown anti-inflammatory activity, and so may be partly responsible for its this effect (Williams et al., 1999). Similarly the essential oils of feverfew have been shown to include well known molecules that have analgesic action, as well as being the most likely source of the plant’s insect repellant properties (Hendriks et al., 1996).

Clinical evidence

The renaissance of feverfew in the 1970s resulted in a considerable body of published clinical studies that tested its use as a treatment for migraine. In their review, Pareek et al. (2011) provide a more detailed history of the data from clinical trials, from the early work on small cohorts, through to larger studies more recently. What is most apparent is that the results reported are generally contradictory, with some studies suggesting feverfew works better than placebo, and others not. This is summed up in the conclusion of researchers working under the auspices of the Cochrane Collaboration, perhaps the most authoritative reviewer of published medical literature. The Cochrane authors found just five studies that were well designed enough to be included in their review – i.e. trials that were fully randomised, placebo-controlled, and double-blind studies – with a total of just 343 patients (Pittler & Ernst, 2004). Perhaps unsurprisingly they concluded that there was insufficient evidence to make firm statements about the efficacy of feverfew. Unfortunately this is a common conclusion from Cochrane reviewers studying herbal medicines because they typically cannot account for differences in preparation of the herbal drug nor for differences in dosage.

One interesting feature Pareek et al. (2011) note is that, of the studies included in the Cochrane review, the two that used standardised extracts of feverfew found no apparent benefit, whereas the two that used dried, ground leaf powder saw greater improvement in their patients. The earlier, successful, trials Pareek et al. cite also used either fresh or dried leaves, rather than extracts. This may suggest that extraction with alcohol or other solvents could reduce the ability of feverfew to prevent migraine. A short paper titled ‘Parthenocide: The demise of a facile theory of feverfew activity’ (Awang, 1998), cites a similar failure of feverfew extract as compared to dried leaves in terms of reducing migraine symptoms. There doesn’t appear to have been formal investigation into the most effective method of preparation for feverfew, and the possibility of a greater efficacy from crude plant material rather than an extract is no more than speculation. What is clear though, is that it is hard to say currently whether or not feverfew is genuinely useful in preventing persistent migraine in a clinical setting.

Given that it is equivocal whether feverfew is effective in treating migraine it may be that the observed biological mechanism involving serotonin release is simply not relevant in reducing migraine. Alternatively it could be that feverfew plants, being living organisms, are inherently variable, and therefore do not always have the chemical profile needed to produce the desired effects in clinical trials. Furthermore, migraine itself is a complex condition with diverse possible causes (Rodriguez-Sainz et al., 2013), and as such it may also be the case that feverfew is only effective in a subset of migraine sufferers. These layers of complexity make treatment of migraine inherently difficult.

Toxicity and side effects

There are several important safety and toxicity issues regarding feverfew that must be borne in mind, despite its generally good safety profile (Pittler & Ernst, 2004). First among them is the importance of the advice for pregnant women not to take it, due to its well-known and long documented popular use as an abortifacient (Pareek et al., 2011). Despite frequent mentions in the modern scientific literature of this historic use, it does not appear that any specific compound has been proposed to be responsible for this effect. Feverfew should also be avoided by lactating mothers as it may be expressed in breast milk, and toxicity in children has not been established (Pareek et al., 2011).

Another safety concern is ‘post-feverfew syndrome’, which was reported to affect patients switched from long term feverfew use to placebo in a clinical trial. This syndrome presented as a return of migraine, as well as anxiety, insomnia, and muscle and joint stiffness (Pittler & Ernst., 2004). The results from clinical trials show no other significant problems, and any side-effects associated with feverfew were mild and reversible (Pittler & Ernst., 2004). There does not appear to be hard data on long-term use of feverfew, however, and as such long-term toxicity has yet to be established.

Closing thoughts

It is worth remembering that for all of its historic use, feverfew was much less commonly used during the 20th century until its return to prominence in the 1970s, and therefore the last 40 years or so of study represent a quite distinct era, separate from the earlier, historical use. Indeed the methods of production and delivery of the drug have become much more varied compared to historical preparations of either fresh or dried leaves / flower heads. Modern cultivation of the plant for herbal remedies and industrial processing of plant material into the capsular form most commonly used today might well be expected to impact the pharmacological properties of the plant. This presents us with the problems outlined here, namely that it has become difficult to establish the efficacy of feverfew as a preventative treatment for migraine in a modern context – i.e. using randomised, placebo controlled clinical trials – despite what one might consider to be a strong and consistent use in previous centuries for pain relief. The future for feverfew may lie in further investigation of what methods of preparation are the most effective, or in continued developments in the understanding of migraine. Whether these developments occur, however, remains to be seen, and it may be that more centuries of popular rather than clinical use are required before feverfew becomes established definitively as either an effective treatment for migraine, or as a relic of traditional medicine either convincingly demonstrated to be ineffective, or superseded by other, better forms of treatment. So as you pass by the Medical Herb Spiral in the garden, pause for a moment and regard this weed, and think of the ancestors with sore heads and joints who would have had to rely on it as a primary source of relief.


In this essay we do not to advise or recommend herbs for medicinal or health use. This information is intended for educational purposes only and should not be considered as a recommendation or an endorsement of any particular medical or health treatment. The use of any such product should be based on the appropriate advice of a healthcare professional or based on the information available on patient information leaflets of EU regulated Traditional Herbal Remedy (THR) products.

© Ben Francis, 2015. All Rights Reserved

Ben Francis, MSc student in Pharmacognosy at the UCL School of Pharmacy, University of London,

Research Cluster Biodiversity and Medicines/ Centre for Pharmacognosy and Phytotherapy,

29 – 39 Brunswick Square, London.


If you would like to contact the author please use the Contact Form on this website

Literature Cited

Awang DVC. 1998. Parthenocide: The demise of a facile theory of feverfew activity. Journal of Herbs, Spices & Medicinal Plants 5(4) 95-98.

Baker RG, Hayden MS, Ghosh S. 2011. NF-κB, inflammation, and metabolic disease. Cell metabolism 13(1), 11-22.

Brasier AR. 2006. The NF-κB regulatory network. Cardiovascular toxicology 6(2), 111-130.

Ghantous A, Gali-Muhtasib H, Vuorela H, Saliba NA, Darwiche N. 2010. What made sesquiterpene lactones reach cancer clinical trials?. Drug discovery today 15(15), 668-678.

Ghantous A, Sinjab A, Herceg Z, Darwiche N. 2013. Parthenolide: from plant shoots to cancer roots. Drug discovery today 18(17), 894-905.

Grigson G. 1975. The Englishman’s Flora. Paladin. Pp 407-408

Guzman ML, Jordan CT. 2005. Feverfew: weeding out the root of leukaemia. Expert Opinion on Biological Therapy 5(9) 1147-1152.

Hehner SP, Heinrich M, Bork PM, Vogt M, Ratter F, Lehmann V, Schulze-Osthoff K, Dröge W, Schmitz ML. 1998. Sesquiterpene lactones specifically inhibit activation of NF-κB by preventing the degradation of IκB-α and IκB-β. Journal of Biological Chemistry 273(3), 1288-1297.

Hendriks H, Bos R, Woerdenbag HJ. 1996. The Essential Oil of Tanacetum parthenium (L.) Schultz‐Bip. Flavour and Fragrance Journal 11(6), 367-371.

Heptinstall S. 1988. Feverfew – an ancient remedy for modern times? Journal of the Royal Society of Medicine 81(7), 373-374.

Hobbs C. 1989. Feverfew Tanacetum parthenium: A Review. HerbalGram 20(Spring), 26-36.

Humphrey P, Feniuk W, Perren MJ, Beresford IJ, Skingle M, Whalley ET. 1990. Serotonin and migraine. Annals of the New York Academy of Sciences 600(1), 587-598.

Lesiak K, Koprowska K, Zalesna I, Nejc D, Düchler M, Czyz M. 2010. Parthenolide, a sesquiterpene lactone from the medical herb feverfew, shows anticancer activity against human melanoma cells in vitro. Melanoma research 20(1), 21-34.

Medicines and Healthcare products Regulatory Agency (MHRA). 2009. MHRA PAR; DiaMigraine Hard Capsules and DiaFeverfew Hard Capsules, THR 33518/0013-14.