Migraine Prevention and Anti-Inflammatory Effects

Evidence Review

Early Clinical Evidence: Murphy et al. (1988)

The landmark early trial by Murphy, Heptinstall, and Mitchell published in the Lancet established feverfew's clinical credibility [1]. In this randomized, double-blind, crossover design, 72 volunteers — most of whom had been self-medicating with fresh feverfew leaves — were randomized to either dried feverfew leaf capsules or matching placebo for consecutive 4-month periods.

Key findings from the 59 participants who completed both crossover periods:

• Mean migraine frequency was significantly lower during feverfew treatment compared to placebo
• Severity of individual attacks and degree of vomiting were also reduced with feverfew
• Patients crossing from feverfew to placebo experienced an increase in headaches above their pre-treatment baseline — a pattern consistent with loss of an active preventive rather than regression to the mean
• Visual analogue pain scores were significantly improved in the feverfew period
• No serious adverse effects were reported

The crossover design is particularly robust for migraine prevention studies: patients serve as their own controls, removing inter-individual variation in baseline attack frequency that confounds parallel-group designs. The rebound on crossover to placebo strengthens the case for genuine pharmacological activity.

Dose-Response Trial: Pfaffenrath et al. (2002)

Pfaffenrath et al. conducted a multicentre, randomized, double-blind, placebo-controlled dose-response study in 147 migraine patients across Germany [2]. Participants were randomized to three doses of MIG-99 feverfew extract (6.25 mg once, twice, or three times daily) or placebo for 12 weeks.

In the overall intention-to-treat analysis, feverfew did not reach statistical significance versus placebo — a result influenced by the inclusion of patients with lower baseline attack frequencies where any treatment is harder to distinguish from placebo variability.

In the pre-specified subgroup of patients with at least four migraines per 28 days during baseline (n = 89), results were markedly different:

• Attack frequency decreased in a statistically significant, dose-dependent manner across all three feverfew doses vs. placebo (p = 0.001)
• The highest dose (6.25 mg three times daily) produced the greatest absolute change: −1.8 attacks per 28 days compared to placebo
• Adverse event rates were similar between feverfew and placebo groups (approximately 35% in both), with predominantly mild and transient gastrointestinal complaints
• No serious adverse events were attributed to the study drug

The dose-response relationship — absent from random noise — provides strong evidence that the effect is pharmacologically mediated. The finding also highlights an important clinical point: feverfew may be most effective for patients with higher baseline migraine frequency, consistent with how most preventive agents perform.

Systematic Review: Ernst & Pittler (2000)

Ernst and Pittler systematically reviewed all randomized controlled trials of feverfew for migraine prevention available up to that point [5]. Their analysis of the five highest-quality double-blind RCTs found:

• The majority of trials showed reduction in attack frequency with feverfew versus placebo
• Effect sizes were modest but clinically meaningful for a preventive agent with a clean safety profile
• The primary methodological limitation across trials was heterogeneity in feverfew preparations and parthenolide content — dried leaf products showed variable results while more standardized preparations showed more consistent effects
• The review identified feverfew as likely effective for migraine prophylaxis and recommended standardized extract preparations for both clinical use and future research

This methodological point — that preparation standardization determines trial outcome consistency — has been borne out in subsequent research. The MIG-99 extract used in later trials provided more reproducible results than earlier studies using variable dried leaf preparations.

Mechanistic Research

Two key laboratory studies established the pharmacological basis for feverfew's effects:

Kwok et al. (2001) identified parthenolide as a specific inhibitor of IKKβ through covalent modification [3]. Using a combination of biochemical assays and structural analysis, they showed that parthenolide reacts with cysteine 179 in the IKKβ activation loop — a residue critical for kinase activation. This covalent mechanism means that parthenolide's anti-inflammatory effect is not competitive (cannot be simply displaced by substrate) and persists until new IKKβ protein is synthesized. IKKβ phosphorylates IκBα, releasing NF-κB to translocate to the nucleus and activate inflammatory gene expression. By blocking this step, parthenolide reduces transcription of TNF-α, IL-6, COX-2, and other inflammatory mediators at their source. The precision of this mechanism — acting on a defined molecular target rather than broadly inhibiting all cellular processes — distinguishes parthenolide from many phytochemicals whose mechanisms are loosely defined.

Heptinstall et al. (1985) provided the vascular and platelet mechanism in a series of experiments showing that feverfew extracts inhibit granule secretion from both platelets and polymorphonuclear leucocytes in a concentration-dependent manner [4]. Dense granule secretion — which releases serotonin — was particularly potently inhibited. The inhibition was shown to involve sulfhydryl group modification rather than receptor blockade, explaining why feverfew's platelet effects differ from those of aspirin or ADP receptor antagonists. Serotonin released from activated platelets causes initial cerebrovascular constriction; the subsequent vasodilation is one model for the throbbing quality of migraine pain. Reducing platelet serotonin secretion may interrupt this vascular cascade before it amplifies.

Strength of Evidence

Feverfew holds one of the better-evidenced positions among herbal migraine preventives. Multiple independent RCTs across several decades show consistent directional effects. The underlying mechanism — inhibition of platelet serotonin release, prostaglandin synthesis, and NF-κB signaling via parthenolide — is pharmacologically coherent, experimentally verified at the molecular level, and provides a plausible biological explanation for the clinical observations.

Limitations: Most trials are short (12–16 weeks), limiting conclusions about long-term efficacy and safety. Individual studies are underpowered by modern standards. Preparation heterogeneity across studies complicates direct comparison. The Cochrane systematic review rates the overall evidence as moderate quality, noting that more rigorous trials with standardized extracts and clearly defined patient populations are needed to establish precise effect size estimates.

For patients with frequent episodic migraine seeking a non-pharmaceutical preventive option with a well-characterized safety profile and no known dependence liability, the evidence supports feverfew as a reasonable addition to a prevention strategy — taken consistently at a standardized dose for at least 3–4 months before drawing conclusions about individual response.