Menstrual Migraine Tracking: Why Three Cycles Is the Minimum

Menstrual migraine is defined by timing rather than by how the attack feels. The criteria sit in the appendix of the International Classification of Headache Disorders, third edition, and they ask one question: do attacks land in a five day window around the first day of menstruation, in at least two out of three cycles? A single bad month cannot answer that. Three cycles of records can, which makes this one of the few headache questions where tracking is the instrument rather than a nice extra.

What the classification calls it

ICHD-3 recognises two menstrual patterns and keeps both in its appendix. Pure menstrual migraine without aura describes attacks in a menstruating woman that fulfil the criteria for migraine without aura and occur exclusively on day 1 plus or minus 2 of menstruation, in at least two out of three menstrual cycles and at no other times of the cycle. Menstrually related migraine without aura uses the same window and the same two-in-three rule, but the attacks also happen at other times of the cycle.

That appendix placement is deliberate rather than an oversight. The classification says that ICHD-3 beta offers criteria for pure menstrual migraine and menstrually-related migraine, but places them in the appendix because of uncertainty over whether they should be regarded as separate entities. These are working definitions under review, not settled diagnoses.

The difference between the two is not academic. ICHD-3 states that the importance of distinguishing between pure menstrual migraine without aura and menstrually related migraine without aura is that hormone prophylaxis is more likely to be effective for the former. Which of the two you have is a treatment-relevant fact, and only a record can establish it.

One detail before the counting starts. ICHD-3 notes that menstrual migraine attacks are mostly without aura, which is why the appendix criteria are written for migraine without aura.

The window, and how the days are numbered

Day 1 plus or minus 2 sounds vague and is not. ICHD-3 spells the numbering out: the first day of menstruation is day 1, the preceding day is day minus 1, and there is no day 0. The window therefore runs from two days before bleeding starts through the third day of bleeding, a span of five days.

In plain language, the American Migraine Foundation puts the same window this way, describing pure menstrual migraine as attacks that occur only during the menstrual window, typically beginning two days before menstruation through the first three days of bleeding.

What counts as menstruation is defined more broadly than people expect. For the purposes of ICHD-3, menstruation is endometrial bleeding resulting either from the normal menstrual cycle or from the withdrawal of exogenous progestogens, as in the use of combined oral contraceptives or cyclical hormone replacement therapy. Bleeding during a pill-free week counts.

Why estrogen withdrawal is the leading explanation

On mechanism, ICHD-3 is careful. There is some evidence that menstrual migraine attacks, at least in some women, result from estrogen withdrawal, the classification states, although other hormonal and biochemical changes at this time of the cycle may also be relevant. A 2021 review in Lancet Neurology by Vetvik and MacGregor names two identified mechanisms, estrogen withdrawal and prostaglandin release.

The most direct evidence comes from a study that did the tracking properly. MacGregor and colleagues, in Neurology in 2006, recruited women with regular cycles who were not using hormonal contraception or treatments and who experienced between one and four attacks a month, one of which regularly fell on or between days 1 plus or minus 2 of menstruation, and followed them for three cycles. Urine samples were collected daily for hormone assay and all the women kept a daily migraine diary. Of the 40 women recruited, data from 38 were available for analysis.

What they found points at the direction of change rather than at the hormone itself. Compared with the expected number of attacks, there was a significantly higher number during the late luteal and early follicular phase of falling estrogen, and a lower number during rising phases of estrogen. Their conclusion added that rising levels of estrogen appear to offer some protection against migraine.

For a general audience, the American Migraine Foundation describes the same idea: estrogen levels naturally rise and fall through the cycle, drop rapidly just before menstruation begins, and for some people this sudden decline appears to make the brain more vulnerable to a migraine attack.

ICHD-3 also carries a separate code for the version driven by medication. Estrogen-withdrawal headache is headache or migraine developing within 5 days after daily consumption of exogenous estrogen for 3 weeks or longer that has been interrupted, usually during the pill-free interval of combined oral contraception, and resolving spontaneously within 3 days in the absence of further consumption. If your worst headaches land in that week, that timing is worth describing to whoever prescribes the pill, along with the aura question further down this page.

Why a single cycle proves nothing

One sentence in ICHD-3 should govern how anyone approaches this. In the comments on non-menstrual migraine, the classification states that many women over-report an association between attacks and menstruation, and that for research purposes, diary-documented, prospectively-recorded evidence over a minimum of three cycles is necessary to confirm the diagnosis.

This is arithmetic rather than doubt about anyone's account. The window is five days wide, and attacks land inside it sometimes for no reason at all. Vetvik and MacGregor define the term on exactly that basis: menstrual migraine refers to migraine that is associated with menstruation by more than chance, but it does not define pathophysiology. In a single month, one coincidence and one pattern look identical.

Statisticians have worked on where the floor sits. Barra, Dahl and Vetvik, writing in Headache in 2015, set out a method for testing the association and noted that methods aiming to exclude spurious associations are wanted, so that further research into these mechanisms can be performed on a population with a true association.

They concluded that the proposed standard of three cycle observations before setting a menstrually related migraine diagnosis should be extended with at least one perimenstrual window to obtain sufficient information for statistical processing. Three cycles is a floor, not a comfortable margin.

The clinical bar is lower than the research bar, which is where a personal record earns its place. ICHD-3 notes that a prospective diary is recommended for research purposes but is not mandatory for clinical diagnosis. Nothing stops you from arriving with the diary anyway, and it moves the conversation from a story about bad weeks to a documented pattern.

What perimenstrual attacks are like

Women with a diagnosed menstrual pattern tend to lose more to those attacks. Vetvik and MacGregor report that in women diagnosed with menstrual migraine, perimenstrual attacks are associated with substantially greater disability than their non-menstrual attacks. The same review puts the prevalence at about 20% to 25% of female migraineurs in the general population and 22% to 70% of patients presenting to headache clinics.

What patients report is broader than what gets diagnosed. The American Migraine Foundation states that more than half of women living with migraine report that their attacks are associated with menstruation, and that these attacks are often more severe, last longer, and can be more difficult to treat than migraine attacks at other times of the month. That figure and the 20% to 25% above are not measuring the same thing. One counts women reporting an association, the other counts criteria-diagnosed menstrual migraine among female migraineurs in the general population, and no source compares the two directly, so neither is evidence about the other. What sits alongside both is the ICHD-3 comment that many women over-report an association between attacks and menstruation, which is a statement about recollection and the reason a prospective diary is asked for in the first place.

Whether perimenstrual attacks differ in kind is not settled. Vetvik and Russell reviewed the question in Current Pain and Headache Reports in 2011 and concluded that the different studies provide conflicting results, so it is not possible to answer the question firmly. Vetvik and MacGregor blame loose interpretation of the diagnostic criteria for conflicting results on prevalence figures, clinical characteristics and response to treatment. More prospectively recorded cycles, from more people, is the fix the field is asking for.

What to record, and for how long

Three cycles is the target, and the daily entry is short. The American Migraine Foundation lists what a menstrual migraine diary should hold: the first day of your menstrual period, when migraine attacks occur, symptoms experienced, medication used, possible triggers, and attack duration. It notes that this information also helps healthcare providers determine whether migraine attacks are hormonally related.

Day 1 is the entry people skip, and it is the one the whole classification hangs on. Without it, an attack log is a list of bad days with no axis to plot them against.

Some tracking apps keep hormones in the trigger list for this reason. MigrAid, for example, has hormones among its 12 built-in triggers alongside custom ones you can add, and exports any date range as a PDF or CSV. Marking day 1 is still your job, in whatever tool you use.

Log what the attack was like, not only that it happened. Since the open question in the research is whether perimenstrual attacks differ from the rest, your own answer is the part a clinician cannot get anywhere else. Our guide to tracking migraine triggers covers how to keep that record short enough to survive a bad week.

A hormonal rhythm is not the only pattern people look for, and the two are easy to confuse. Weather is the other common suspect, and we built a four-factor risk index covering 100 US cities that includes it, in our US migraine risk index.

When the pattern is worth a conversation, and when it is not one

The American Migraine Foundation suggests raising menstrual migraine with a healthcare provider if attacks occur with most menstrual cycles, if they become more frequent or severe, if migraine interferes with work, school or daily activities, if current treatment is no longer effective, if you are planning a pregnancy, if you are considering hormonal contraception or hormone therapy, or if you notice significant changes in your pattern during perimenopause or menopause.

Patterns are not permanent, which is an argument for keeping the record going after the first question is answered. ICHD-3 notes that the menstrual relation may change over a woman's reproductive lifetime.

Hormonal medication cuts both ways and cannot be predicted from someone else's experience. The American Migraine Foundation puts it neutrally: hormonal medications, including certain forms of birth control or hormone therapy, may also influence migraine patterns, with some individuals experiencing improvement, others noticing worsening symptoms, and many noticing no change. That is a question for your own doctor with your own three cycles in hand.

There is one specific thing worth raising by name, and it turns on aura instead of on timing. A 2017 consensus statement from the European Headache Federation and the European Society of Contraception and Reproductive Health reviewed the evidence and reported that several data indicate that migraine, especially migraine with aura, is associated with an increased risk of ischemic stroke and other vascular events, and that available data suggest that combined hormonal contraceptive may further increase the risk of ischemic stroke in those who have migraine, specifically migraine with aura. Those authors are careful about how firm this is: they write that the overall quality of current evidence on this question is low, and that their statements privilege safety. Menstrual attacks are mostly without aura, so this may well not describe you. But whether you have ever had aura symptoms at any point in the cycle is something the person prescribing your contraception needs to know, and it is not a question a diary answers or that an article can settle for you. Bring it up with them.

One thing here is not a conversation to schedule. A headache that reaches its worst in under a minute, or that comes with new neurological symptoms, needs emergency care the same day. The American Migraine Foundation advises that on thunderclap headache, if you are experiencing this type of headache pain for the first time, you should visit the ED, meaning the emergency department, and that you should also visit the ED right away if you experience any new and sudden-onset neurological symptoms, including weakness on one side of the face or body, numbness, the inability to speak normally, vision changes, unusual dizziness, or trouble walking. Aura can resemble some of those, which is exactly why the first time it happens is not the time to assume it is aura. Go and be told it was.

If your attack count is climbing overall instead of clustering, that is a different question with its own thresholds, and we covered it in how many migraines a month is normal.

The bottom line

Menstrual migraine is a timing diagnosis. It has a five day window running from two days before bleeding through the third day, a rule requiring attacks in at least two out of three cycles, and a research standard of at least three prospectively recorded cycles. No single month can satisfy any of that, however convincing that month felt at the time.

So mark day 1, log the attacks as they happen, and give it three cycles. After that the answer is either on the page or it is not, and either outcome is more useful than a suspicion you have been carrying for years. Two things do not wait for the three cycles: aura symptoms, if you are using or considering combined hormonal contraception, and any headache that peaks in under a minute or brings new neurological symptoms, which is an emergency.