Update: The Intersection of Migraine and Stroke. New considerations regarding migraine-associated stroke risk, migraine aura and oral contraceptive use
/
Stroke and migraine are the most prevalent neurologic disorders. About 800,000 strokes occur each year in the United States, and migraine actively afflicts at least 40 million Americans. Stroke is the most common neurologic cause of hospitalization, and migraine is the most common neurologic cause for an outpatient visit to a healthcare provider. Do these two common neurologic disorders intersect? If so, how and why?
Stroke implies the acute onset of neurologic deficit caused by alteration in blood flow to all or a portion of the brain; if the signs and symptoms of a stroke resolve within 24 hours of onset, the clinical event is said to be a “transient ischemic attack (TIA)”. There are two basic types of stroke: hemorrhagic and ischemic. “Hemorrhagic” implies that a brain blood vessel – usually an artery – has ruptured and is spilling blood into the space it occupies or into adjacent brain tissue.
“Ischemic” implies that a vessel – usually an artery – has become occluded and that there is immediate loss of neurologic function and progressive death in the brain tissue supplied by that vessel due to impaired delivery of glucose and oxygen, nutrients which brain neurons require for function and survival on literally a second to second basis.
Especially in migraineurs with aura, the symptoms of migraine at times can mimic those of stroke, and the reverse can occur as well (ie, stroke or TIA misdiagnosed as migraine), but is there a connection between the two disorders that goes beyond one simply mimicking the other? We addressed the potential intersection of stroke and migraine in our Spring 2021 issue. Given the importance of the topic and the research data which have emerged since that issue was published, an update appears to be timely.
Is there a relationship between migraine and stroke?
This very large question can be broken down into four smaller and interrelated questions:
Is migraine associated with an increased risk of stroke?
If yes, are there subpopulations of individuals with migraine who are at a particularly increased risk of stroke?
Specifically, is migraine aura and/or use of an oral contraceptive (OCP) associated with an increased risk of stroke in the female migraine population?
If the answer to the first question is yes, can migraine directly cause stroke?
The answers are, respectively: minimally (at most), uncertain, probably not, and probably (but rarely).
Surprising, but perhaps not entirely satisfying, eh?
For quite some time now it has been assumed that migraine is associated with an increased risk of stroke in both males and females, with that risk running about 2 to 3 times that of age-matched population in females and roughly 2 times that of age-matched population in males. Along with this, it has been assumed that the increased stroke risk associated with migraine is higher in those who have migraine with aura and is especially high in females who have migraine with aura and are taking an estrogen-based oral contraceptive…with that combination - female gender, history of aura, use of OCP - conveying as much as a 9-fold increase in the relative risk of stroke.
These long held assumptions are not holding up very well to careful scrutiny. For example, more recent studies have suggested that stroke risk may be no higher in the migraine population than in non-migraineurs. As for aura’s potential contribution to stroke risk, one recent investigation found that a history of aura does not increase the risk of migraine-associated stroke and, in fact, is correlated with a lower risk than exists for those who have migraine without aura. Regarding OCP use, there is recent evidence suggesting that use of an OCP containing less than 30 µg of estrogen does not convey any increase in stroke risk for females with migraine.
Somewhat at odds with the findings regarding aura, one recent study reported that females under age 25 with migraine and a history of aura have a 56% increase in stroke risk. Sounds pretty impressive: 56%!!!, but relative risk needs to be put into context. The annual incidence of stroke in this population of individuals - females < age 25 - is 2.5 cases per 100,000 individuals. A 56% increase in that risk translates to an increase in stroke rate to 3.9 cases per 100,000 individuals annually. In other words, even if aura does increase the relative risk of stroke, the absolute increase in stroke risk is very, very low (an additional 1.4 cases per 100,000 individuals per year).
This migraine/OCP use/stroke risk issue has much clinical relevance. For one thing, the question of OCP safety invariably arises at high frequency in the clinics of primary care providers, gynecologists and neurologists. For another, even if OCP use by migraineurs does predispose to stroke (a very open question), a risk:benefit analysis does little to support any wholesale contraindication to prescribing OCPs in the migraine population. Many investigators have emphasized that the risk of unintended pregnancy is much greater than any purported OCP-related risk of stroke. Some have calculated that restricting OCP use in female migraineurs in the U.S. would avoid ~425 cases of stroke annually at a cost of ~700,000 unintended pregnancies…and, again, that’s assuming there is in fact any increased risk conveyed by the lower estrogen content OCPs.
Given these new data and the uncertainty as to whether stroke risk is indeed increased in any individual with migraine, the persisting inclusion of “migrainous infarction [stroke due to migraine]” in the International Classification of Headache Disorders (ICHD) - which is, as is the DSM for psychiatric disorders, the “Bible” of headache diagnosis - seems paradoxical and even contradictory.
But…let’s go with the ICHD diagnostic flow. How might migraine cause stroke? Well, it could do so indirectly, serving as a marker to indicate the presence of other co-existing disorders that would predispose to stroke. For example, there appears to be an increased prevalence of mitral valve prolapse (MVP) in individuals with migraine, and for a time cardioembolism from MVP was a leading candidate for the cause of stroke in cases of “stroke of unknown cause”. Subsequent research has pretty much put that notion to rest.
A patent foramen ovale (PFO) is a congenital opening between the right and left atria, the two upper chambers of the heart, that can permit clots travelling in the venous blood to pass through the heart and on to the brain without being filtered out by the lungs. The presence of a PFO is associated with increased risk of recurrent stroke following an initial “cryptogenic stroke” (a.k.a. stroke of unknown cause), especially if that initial stroke occurs in a young person, and it has become routine to employ catheter-based PFO closure as a stroke prevention therapy. PFOs may be detected in as many as 50% or more of individuals who have migraine with aura, and – more specifically – there appears to be a correlation between anatomically significant PFOs and migraine with aura. Although there are those who consider PFO closure to be an acceptable treatment for migraine prophylaxis as well as for secondary stroke prevention, the utility of PFO closure for migraine prevention remains highly controversial.
Two other candidates for the “migraine serving as a marker for pathology predisposing to stroke” hypothesis include a migraine-associated prothrombotic (“sticky blood”) state and migraine-associated abnormalities of the arterial wall (endothelium) predisposing to clot formation or a predisposition for the wall to tear (dissection). There does not exist a great deal of evidence to support either process as being a major player in migraine-associated stroke.
So, how could migraine directly cause stroke? Some have pointed to cortical spreading depression (CSD), the electrochemical phenomenon that appears to be the source of migraine aura and is considered by some to be the potential “headwaters” of migraine headache. Along with the electrical/neuronal silence that occurs with CSD is a directly associated decrease in blood flow to the involved brain tissue. Could that CSD-related decrease in blood flow on occasion become so severe and prolonged as to produce stroke? Maybe, but any existing evidence base to support this hypothesis is paper-thin.
One intriguing and plausible mechanism by which migraine directly could cause stroke would involve spasm of brain arteries induced by the acute migraine attack itself. This has been reported in the medical literature and documented by acutely performed and serial angiography. Even so, if migrainous vasospasm causing stroke does occur, it does so quite infrequently… perhaps so infrequently that in population studies it does not make enough of a contribution to demonstrate an increased risk of stroke in migraineurs relative to age-matched population.
Summary
It remains unclear whether there is in fact any correlation between migraine and an increased risk of stroke.
Recall that with any medical disorder – including migraine – correlation is not synonymous with causation (in this case, even if migraine does correlate with an increase in stroke risk - now an open question – it may not be migraine itself that is causing the excess of strokes).
Aura may make no meaningful contribution to an increase in migraine-associated stroke risk.
Estrogen can be prothrombotic, but the <30 µg dose recommendation for OCPs appears to be sufficiently low as to eliminate stroke risk.
Even if stroke risk is increased in the migraine population, the absolute risk and the consequent number of patients affected is so small as to be undetectable or barely detectable in population studies.
Given the high prevalence of migraine and the tremendous and adverse effect migraine has on the public health, it seems backwards that we worry so much about migraine causing stroke. Of the roughly 800,000 strokes which occur each year in the United States, well less than 1000 of those strokes may be a direct consequence of migraine.
