Condition deep-dive · 8 min read

Migraine prevention stack — the four supplements with guideline-level evidence

Updated 2026-05-02 · Reviewed by SupplementScore editors · No sponsorships

Migraine prevention is one of the few areas of supplement medicine where major neurology societies have included specific compounds in their formal recommendations. The American Academy of Neurology and the American Headache Society have both, in recent guidelines, listed riboflavin, magnesium, and CoQ10 with explicit evidence grades. Feverfew has slightly weaker but real evidence. Together they form the most credible non-pharmaceutical prevention stack.

Read this first. Migraines that change in pattern (sudden new onset after age 50, "thunderclap" severity, new neurological symptoms, fever or stiff neck) need urgent medical assessment, not supplement experimentation. The protocol below is for established migraine pattern, ideally diagnosed by a clinician, where the goal is reducing frequency and severity. Acute treatment of an active attack is a separate question — these supplements are prevention, not abortive.

The four with guideline-level evidence

AAN evidence · Frequency reduction

Riboflavin (vitamin B2)

400 mg/day, single dose with breakfast, for at least 8 weeks

The most reproducibly positive supplement intervention for migraine prevention. The classic Schoenen trial showed about a 50% reduction in migraine days in roughly 60% of responders at 400 mg/day. Effect takes 8 weeks to develop fully. Mechanism is thought to relate to mitochondrial energy metabolism in the trigeminal-vascular system. Side effects are limited to bright-yellow urine (universal, harmless). Active R5P forms exist but the trial dose is standard riboflavin at 400 mg.

AHS evidence · Frequency, particularly menstrual migraine

Magnesium

600 mg elemental magnesium daily — citrate or glycinate; effect over 8–12 weeks

Multiple RCTs support magnesium prophylaxis. The most positive trials used magnesium citrate at 600 mg elemental magnesium per day, divided across the day. Citrate's main side-effect (loose stools) often becomes the dose-limiting factor; if this happens, switching to glycinate sacrifices a modest amount of evidence-base for substantially better tolerance. Particularly effective for menstrual migraine. Stacks cleanly with riboflavin.

AAN evidence · Frequency

CoQ10

100 mg three times daily (300 mg total), for at least 12 weeks

Consistent evidence from several controlled trials. Effect develops more slowly than riboflavin or magnesium — give it 12 weeks before assessing. Ubiquinol form is somewhat better-absorbed (and worth the modest extra cost in adults over 40) but ubiquinone is what most positive trials used. Generally very well tolerated. Theoretical interaction with warfarin (CoQ10 has structural similarity to vitamin K2 and can reduce warfarin effect) — relevant only for anticoagulated patients.

AAN evidence · Standardised feverfew only

Feverfew (parthenolide-standardised, MIG-99 or equivalent)

6.25 mg three times daily of MIG-99-standardised CO2 extract

The trial-validated form is a specific standardised CO2 extract, not generic powdered feverfew leaf (which has highly variable parthenolide content). Effect size is smaller than riboflavin or magnesium. Generally well tolerated. Mild antiplatelet effect; theoretical interaction with anticoagulants. Avoid in pregnancy.

Combination therapy

The four mechanisms are non-overlapping — mitochondrial (riboflavin, CoQ10), neuromuscular and vascular (magnesium), serotonergic and inflammatory (feverfew). Combination protocols have been studied in several open-label trials with reasonable response rates, particularly the riboflavin + magnesium + CoQ10 combination. It is reasonable to start with this triplet and add feverfew if response is partial after 12 weeks.

What about butterbur?

Butterbur (Petasites hybridus) had AAN Level A evidence for migraine prevention until the safety picture changed. The plant naturally contains pyrrolizidine alkaloids that are hepatotoxic and potentially carcinogenic. Properly processed PA-free extracts (the specific Petadolex preparation) had clean liver-safety data, but distribution problems and ongoing concerns about extract quality have led most current guidelines to remove or downgrade butterbur. Until the supply-chain issues are resolved, the four-supplement stack above is a more defensible choice.

What to skip

Lifestyle context

No supplement protocol outperforms a clean trigger-management plan. The interventions with the largest effect sizes for migraine frequency in the controlled-trial literature are: regular sleep schedule (not just sufficient sleep — consistent timing), aerobic exercise three times weekly, hydration, identification and avoidance of personal dietary triggers (which differ widely), and management of comorbid anxiety and depression where present. Supplements work better when these are in place.

Practical quick-start. Riboflavin 400 mg daily + magnesium glycinate 400 mg elemental daily + CoQ10 100 mg three times daily, for 12 weeks. Track migraine days and severity weekly using a simple paper calendar. If you've reduced frequency by ≥30% at 12 weeks, the protocol is working — continue. If not, add MIG-99 feverfew at 6.25 mg three times daily for an additional 8 weeks before reassessing or escalating to prescription prevention.

What to track

Headache diary apps work well, but a paper calendar with one annotation per migraine day is sufficient: date, severity (1 to 10), duration, and any acute medication used. Reassess monthly. Also track caffeine intake, sleep timing, and major stressors — supplement effects can be obscured by lifestyle drift.