Skip navigation

Feverfew for Migraine Prophylaxis

Clinical bottom line: Overall, these studies suggest that feverfew may be beneficial for the prevention of migraine attacks. However, the effectiveness has not been established beyond reasonable doubt. More data is needed to determine which dose and formulation should be prescribed, and how effective it is. Future trials should measure clinically relevant outcomes in accordance with International Headache Society (IHS) criteria.

Feverfew (Tanacetum parthenemium L) is a popular herbal remedy recommended for the prevention of migraine. The pharmacological properties of feverfew have been extensively investigated but remain unclear.

Systematic review

Vogler BK, Pittler MH and Ernst E. Feverfew as a preventive treatment for migraine: a systematic review. Cephalalgia. 1998; 18:704-708

Date review completed: April 1998

Number of trials included: 5

Number of patients: 196

Control group: placebo

Main outcomes: headache severity and frequency, nausea and vomiting, photophobia

Inclusion criteria were randomised, double-blind, placebo-controlled trials of feverfew mono-preparations taken for migraine prophylaxis. No restriction to language of publication.

Reviewers conducted extensive searches of all main databases and reference lists of retrieved reports, and contacted manufacturers of feverfew preparations. Trials were scored for methodological quality using the Oxford scale (Jadad et al., 1996). Reviewers provided a descriptive summary of included trials, as pooling of data for meta-analysis was not possible.


Five trials were found, one in abstract form that measured serotonin uptake and platelet activity and is not discussed further here. Four randomised, double-blind, placebo-controlled trials were found reporting on clinical outcomes. All trials scored at least three of five points on the quality scale. Trials varied with respect to dosing schedules and formulations of feverfew, and outcome measures. Only one trial used IHS criteria for migraine definition, diagnostic criteria were unclear in two of the other trials. Three trials were of crossover and one of parallel design. Reviewers did not state if there were any carry-over effects in the crossover studies. One trial had less than ten patients per treatment group. Overall three trials reported results in favour of feverfew over placebo and one found no difference.

One trial randomised 17 patients to 100 mg feverfew daily (capsules containing freeze-dried leaves) or placebo for 24 weeks. All patients had taken raw feverfew leaves every day for three to four years. There was a significant increase in attack frequency in the placebo group (p<0.02), attack frequency remained constant in the feverfew group. Five of eight patients in the feverfew group reported good to excellent effectiveness based on a global assessment of treatment, compared with one of nine in placebo group. There are too few patients to draw meaningful conclusions from these results.

One trial randomised 72 patients to either one capsule dried feverfew leaves daily (dose not stated) or placebo for eight months (crossing over to other treatment after four months). A significant reduction in headache frequency (p<0.005) was reported by the feverfew group compared with placebo. Global assessment of treatment showed feverfew significantly better than placebo (p<0.0001).

One trial randomised 50 patients with IHS definition of migraine to 143 mg feverfew daily (capsule containing an alcoholic extract of feverfew) or placebo for four months (cross over at two months). There was no significant difference in headache frequency or number of workdays lost between the groups.

One trial randomised 57 patients with medical diagnosis of migraine to 100 mg feverfew daily or placebo for two months (cross over after one month). All patients had received 100 mg feverfew daily for two month run-in period. There was a significant difference in all outcomes including migraine severity between feverfew and placebo (p<0.01).

Adverse effects were reported as being mild and reversible. In total three patients withdrew on feverfew and five on placebo.

Further reading

Pittler MH, Vogler BK and Ernst E. Feverfew for preventing migraine. Cochrane review 2000

Related topics