Migraine is not just a bad headache. It is a neurological condition with its own biology — one that involves how brain cells handle energy, how blood vessels and nerves talk to each other, and how sensitive the nervous system is to ordinary things like light, noise and hunger. That biology is exactly why nutrition keeps coming up in migraine research. A handful of nutrients feed directly into the energy machinery that appears to run short in migraine-prone brains, and several of them have been tested in real randomised trials. This article walks through what those trials found, at what doses, how long they took to work, and where the evidence runs out.
Quick Facts
- Four nutrient approaches have genuine randomised trial data behind them for migraine prevention: riboflavin (vitamin B2), magnesium, coenzyme Q10, and shifting the balance of dietary fats toward omega-3.
- The doses used in trials are much higher than what a multivitamin provides — 400 mg riboflavin, roughly 600 mg magnesium and 300 mg CoQ10 per day.
- None of these work overnight. Most trials only saw a clear separation from placebo in the second or third month of use.
- The realistic effect is fewer attacks per month, not zero attacks. In the riboflavin trial, 59% of participants halved their attack frequency versus 15% on placebo.
- Nutrients are a preventive strategy, not a rescue treatment. They do nothing for an attack already underway.
Why Nutrients Matter in Migraine at All
The leading working model of migraine involves a brain that is unusually excitable and unusually expensive to run. Studies using magnetic resonance spectroscopy have repeatedly found signs of reduced energy reserve in the brains of people with migraine — the cellular equivalent of running a phone permanently at 12% battery. When demand spikes (a missed meal, a bad night, a stressful week, a hormonal shift), the system tips over into an attack.
That model makes specific, testable predictions. If migraine brains struggle with mitochondrial energy production, then nutrients that sit inside the mitochondrial energy pathway should help some people. Riboflavin and coenzyme Q10 both do exactly that. And if migraine involves excessive neuronal excitability, then magnesium — which acts as a natural brake on the NMDA receptor and on calcium entry into nerve cells — becomes an obvious candidate. This is one of the rarer cases in supplement science where the mechanism came first and the trials followed, rather than the other way around.
What that means practically
It means the nutrients worth considering are not a random assortment of "brain health" ingredients. They are a short, specific list, and the ones that work appear to work by topping up an energy or excitability threshold rather than by blocking pain. That also explains the slow onset — you are refilling a reserve, not switching something off.
The Three Nutrients With Real Trial Data
Riboflavin (vitamin B2)
Riboflavin is the raw material for FAD and FMN, two molecules that carry electrons through the mitochondrial respiratory chain. Give the chain more of its cofactor, the theory goes, and cellular energy production becomes more resilient.
The landmark trial, published in Neurology in 1998, gave 55 adults with migraine either 400 mg of riboflavin daily or placebo for three months. Riboflavin beat placebo on both attack frequency and headache days, with 59% of the riboflavin group achieving at least a 50% reduction in attacks compared with 15% on placebo. Side effects were minor and uncommon. A 2017 systematic review found the adult evidence broadly supportive while noting that paediatric trials have been much less consistent.
Two practical points. First, 400 mg is roughly 300 times the daily reference intake, so this is a therapeutic dose, not a nutritional one. Second, it turns urine a vivid fluorescent yellow within hours — harmless, and a useful sign you have actually absorbed it. Our full breakdown of forms and dosing lives on the vitamin B2 (riboflavin) page.
Magnesium
Magnesium sits at the intersection of several migraine mechanisms: it dampens NMDA receptor activity, stabilises blood vessel tone, and appears to raise the threshold for cortical spreading depression, the slow wave of neuronal activity thought to underlie migraine aura.
The best-known trial gave 600 mg of magnesium dicitrate daily to adults with migraine over 12 weeks. Attack frequency fell by 41.6% in the magnesium group versus 15.8% on placebo. Intensity and duration trended in the right direction but did not reach statistical significance. A 2018 systematic review in Headache concluded that magnesium has a reasonable evidence base for prophylaxis while pointing out that the individual trials are small and methodologically uneven.
Form matters more here than in most nutrients, because magnesium oxide is poorly absorbed and reliably causes loose stools at these doses. Chelated forms such as bisglycinate are far better tolerated at the higher intakes migraine prevention calls for — the trade-offs between forms are covered on the magnesium bisglycinate page.
Coenzyme Q10
CoQ10 shuttles electrons between complexes I/II and III of the respiratory chain — again, squarely inside the mitochondrial energy hypothesis.
A 2005 randomised trial in Neurology compared 300 mg of CoQ10 daily (given as 3 × 100 mg) against placebo in 42 people with migraine. By the third treatment month, CoQ10 was superior to placebo on attack frequency, headache days and days with nausea. The 50% responder rate was 47.6% for CoQ10 versus 14.4% for placebo, giving a number-needed-to-treat of three. Tolerability was good.
The catch is that this trial was small and the follow-up literature, while broadly positive in meta-analysis, has not produced a single large, definitive study. CoQ10 is also fat-soluble and absorbed poorly on an empty stomach, so it needs to be taken with a meal containing some fat. See the coenzyme Q10 page for absorption details and the ubiquinone versus ubiquinol question.
Diet Patterns, Fats, and Herbal Options
The omega-3 to omega-6 ratio
The most methodologically impressive nutrition trial in this field is not about a supplement at all. In 2021, researchers published a 16-week randomised controlled trial in the BMJ involving 182 adults who had migraine on 5 to 20 days per month. Participants were assigned to one of three controlled diets: a high omega-3 diet, a high omega-3 plus low omega-6 diet, or a control diet matching typical intakes.
Both omega-3 diets reduced headache days compared with control, and the combined high omega-3 / low omega-6 arm performed best. The proposed mechanism is that omega-3 and omega-6 fatty acids compete for the same enzymes, producing oxidised lipid mediators that either dampen or amplify pain signalling. Shift the raw material and you shift the mediator profile.
Worth being precise about what this study does and does not show: it tested whole controlled diets, not fish oil capsules layered onto an unchanged diet. Reducing omega-6 intake — largely from industrial seed oils and processed foods — was part of the intervention. Background on the fatty acids themselves is on the omega-3 (EPA & DHA) page.
Feverfew
Feverfew (Tanacetum parthenium) has the most trial data of any herb here. A 2005 multicentre randomised trial of a standardised CO2 extract (MIG-99, 6.25 mg three times daily) found migraine frequency fell by 1.9 attacks per month versus 1.3 on placebo, with a 50% responder rate of 30.3% versus 17.3%. That is a real but modest effect, and results across the wider feverfew literature have been inconsistent — largely because extract standardisation varies enormously between products.
Triggers versus prevention
A 2020 systematic review of 43 studies on diet and migraine found that alcohol and caffeine were the most consistently reported dietary predictors of more frequent attacks, and that most tested dietary interventions — including low-fat and elimination diets — reduced attack frequency. The review was candid that overall evidence quality was low, since most trigger studies were cross-sectional or survey-based. Self-reported triggers are notoriously unreliable in migraine: the premonitory phase can begin up to 24 hours before pain and often causes food cravings, meaning the chocolate frequently gets blamed for an attack it was actually an early symptom of.
What the Evidence Actually Shows
High-dose riboflavin reduces migraine attack frequency in adults. Evidence level: Promising. A well-conducted randomised trial showed a clear benefit at 400 mg daily, and a systematic review found the adult data broadly supportive. It falls short of well-established because the pivotal trial was small (55 participants) and paediatric replications have been mixed.
Magnesium supplementation modestly reduces attack frequency. Evidence level: Promising. Multiple placebo-controlled trials point the same direction, with the best-known showing a 41.6% versus 15.8% reduction in attacks. A 2018 systematic review supports a rationale for its use while noting small sample sizes and uneven methodology across the trial base.
CoQ10 at 300 mg daily reduces attack frequency and headache days. Evidence level: Promising. The 2005 randomised trial produced a number-needed-to-treat of three, which is a strong result — but it involved only 42 people and has not been replicated at scale.
Raising omega-3 intake while lowering omega-6 reduces headache days. Evidence level: Promising. The 2021 BMJ trial is the largest and best-controlled nutrition study in migraine, with 182 participants and objective dietary control. The caveat is that it tested a whole dietary pattern, so it does not tell you what a fish oil capsule alone would do.
Alcohol and caffeine are among the most consistently reported dietary triggers. Evidence level: Well-established as an association, weaker as causation. They show up across dozens of studies, but the evidence is dominated by cross-sectional surveys, and the premonitory phase of migraine can generate cravings that make ordinary foods look like culprits.
How to Put This Into Practice
Start with a headache diary, not a supplement
You cannot tell whether anything is working without a baseline. Track attack days per month for at least four weeks before changing anything: date, duration, severity out of ten, and what medication you took. Migraine frequency fluctuates a lot on its own, and without a baseline you will end up crediting or blaming interventions more or less at random.
Change one thing at a time, and give it three months
Every trial described above needed two to three months before separating from placebo. Adding riboflavin, magnesium and CoQ10 simultaneously in week one tells you nothing about which — if any — is doing the work, and leaves you paying for two things you may not need. Pick one, run it for 12 weeks against your baseline, then decide.
Match the trial doses, and know the practicalities
Riboflavin was tested at 400 mg once daily; expect bright yellow urine. Magnesium was tested at around 600 mg of elemental magnesium daily, best split across the day and taken in a chelated form to avoid digestive upset. CoQ10 was tested at 300 mg daily, split into three 100 mg doses and taken with food containing fat.
Do not neglect the boring inputs
Regular meal timing, consistent sleep and wake times including weekends, steady hydration, and a stable rather than escalating caffeine intake all reduce the volatility that appears to precipitate attacks. These are unglamorous and free, and in most people's experience they move the needle at least as much as anything in a capsule.
Know when this is not a supplement question
If you are having attacks on more than four days a month, if attacks are disabling, or if you are using acute pain medication on ten or more days a month, you should be seeing a doctor. That last point matters especially: frequent use of acute painkillers can itself cause medication-overuse headache, a self-perpetuating cycle that no nutrient will fix.
Safety and Who Should Be Cautious
The three main nutrients here have good tolerability records at trial doses, but they are not without considerations. High-dose magnesium commonly causes loose stools and should be used cautiously — and only under medical supervision — by anyone with impaired kidney function, since the kidneys are responsible for clearing excess magnesium. Magnesium can also reduce absorption of certain antibiotics and thyroid medication, so separate doses by several hours. CoQ10 may interact with warfarin, and its structural similarity to vitamin K means anyone on anticoagulants should check with their prescriber first. Riboflavin at 400 mg is generally well tolerated, with occasional diarrhoea or increased urination reported.
Feverfew should be avoided in pregnancy, and people allergic to the daisy family (ragweed, chrysanthemums, marigolds) may react to it. Stopping it abruptly after long-term use has been associated with rebound headaches.
ⓘ A note on butterbur (Petasites hybridus): you will still find it recommended for migraine in older articles, because a 2012 American Academy of Neurology and American Headache Society guideline gave it a Level A rating. That guideline was formally retired in 2015 over safety concerns. Raw butterbur contains pyrrolizidine alkaloids, which are hepatotoxic, and testing of commercial products has repeatedly found them present despite "PA-free" labelling — one review of 21 products found alkaloids in seven of them. We do not recommend butterbur, and neither do the bodies that once did.
Frequently Asked Questions
How long before I know whether a nutrient is working?
Give it a full three months. The riboflavin, magnesium and CoQ10 trials all showed their clearest separation from placebo in month two or three. Judging after two or three weeks will produce a false negative, and stopping early is the single most common reason people conclude these do not work.
Can I take riboflavin, magnesium and CoQ10 together?
There is no known interaction between them, and some headache clinics do combine them. But if you start all three at once you will never know which one earned its place. Sequential trials are slower and considerably more informative.
Will these replace my migraine medication?
No, and you should not stop prescribed medication to try them. These are preventive strategies with modest effect sizes — fewer attacks per month, not none — and they do nothing for an attack in progress. Think of them as something that may sit alongside medical treatment, discussed with your doctor, rather than instead of it.
Do I need to eliminate chocolate, cheese and red wine?
Not automatically. Blanket elimination diets are hard to sustain and often unnecessary. The premonitory phase of a migraine can start up to a day before the pain and frequently causes food cravings, so a food eaten shortly before an attack may be a symptom rather than a cause. Alcohol is the exception — it shows up as a trigger consistently enough across the literature to be worth testing deliberately.
Is magnesium useful if my blood levels are normal?
Possibly. Serum magnesium reflects less than 1% of total body magnesium and is a poor indicator of tissue status, which is why a normal blood result does not rule out a benefit. Trials of magnesium in migraine have generally not selected participants on the basis of blood levels, and still found effects.
Scientific References
- Schoenen J, Jacquy J, Lenaerts M. Effectiveness of high-dose riboflavin in migraine prophylaxis: a randomized controlled trial. Neurology. 1998;50(2):466–470.
- Thompson DF, Saluja HS. Prophylaxis of migraine headaches with riboflavin: a systematic review. Journal of Clinical Pharmacy and Therapeutics. 2017;42(4):394–403.
- Peikert A, Wilimzig C, Köhne-Volland R. Prophylaxis of migraine with oral magnesium: results from a prospective, multi-center, placebo-controlled and double-blind randomized study. Cephalalgia. 1996;16(4):257–263.
- von Luckner A, Riederer F. Magnesium in migraine prophylaxis — is there an evidence-based rationale? A systematic review. Headache: The Journal of Head and Face Pain. 2018;58(2):199–209.
- Sándor PS, Di Clemente L, Coppola G, et al. Efficacy of coenzyme Q10 in migraine prophylaxis: a randomized controlled trial. Neurology. 2005;64(4):713–715.
- Ramsden CE, Zamora D, Faurot KR, et al. Dietary alteration of n-3 and n-6 fatty acids for headache reduction in adults with migraine: randomized controlled trial. BMJ. 2021;374:n1448.
- Diener HC, Pfaffenrath V, Schnitker J, Friede M, Henneicke-von Zepelin HH. Efficacy and safety of 6.25 mg t.i.d. feverfew CO2-extract (MIG-99) in migraine prevention: a randomized, double-blind, multicentre, placebo-controlled study. Cephalalgia. 2005;25(11):1031–1041.
- Hindiyeh NA, Zhang N, Farrar M, Banerjee P, Lombard L, Aurora SK. The role of diet and nutrition in migraine triggers and treatment: a systematic literature review. Headache: The Journal of Head and Face Pain. 2020;60(7):1300–1316.
- Holland S, Silberstein SD, Freitag F, Dodick DW, Argoff C, Ashman E. Evidence-based guideline update: NSAIDs and other complementary treatments for episodic migraine prevention in adults. Neurology. 2012;78(17):1346–1353. (Retired by the American Academy of Neurology in 2015 over butterbur safety concerns.)
Disclaimer
This article is provided for educational and informational purposes only. It is not medical advice, and it is not intended to diagnose, treat, cure or prevent any disease. Migraine is a medical condition that warrants proper diagnosis — new, severe, sudden-onset or changing headaches should always be assessed by a healthcare professional. Always consult a qualified doctor or pharmacist before starting any supplement, particularly at the therapeutic doses discussed here, and especially if you are pregnant or breastfeeding, have kidney or liver disease, or take prescription medication including anticoagulants. Do not discontinue prescribed treatment without medical guidance. Individual responses to nutritional interventions vary considerably.
