Migraine is ‘vigorous player’ that underlies ischemic stroke in young adults

Any migraine, as well as migraine with aura, demonstrated a strong association with cryptogenic ischemic stroke among adults aged 18 to 49 years, according to the results of a prospective, multicenter case-control study.

This association remained strong regardless of vascular risk factors and the presence of patent foramen ovale. However, the researchers did not find a similar association between cryptogenic ischemic stroke and migraine without aura.

“Many young patients with cryptogenic ischemic stroke (CIS) are diagnosed with patent foramen ovale (PFO), suggesting paradoxical embolism as one of the main causes based on evidence from observational studies and secondary prevention trials comparing PFO closure to medical treatment,” the researchers wrote. “Migraine is another frequently encountered feature in young stroke patients.”

Nicolas Martinez-Majander, MD, of the department of neurology at Helsinki University Hospital and the University of Helsinki in Finland, and colleagues hypothesized that migraine would exert a “relatively strong association” on CIS in patients at a younger age and that sex, as well as the presence of PFO, could potentially moderate that link. Their study included the use of a screening tool they developed to allow for the diagnosis and subtyping of migraine. They compared that screening tool against the gold standard for migraine diagnosis — an interview with a headache specialist following the diagnostic criteria of the International Classification of Headache Disorders.

The researchers prospectively enrolled 347 consecutive adult patients with a recent CIS and 347 age- and sex-matched (± 5 years) control participants with no history of stroke. A screener identified occurrences of any migraine, migraine with aura (MA) and migraine without aura (MO). A headache neurologist validated those results.

Martinez-Majander and colleagues examined the independent relationship between migraine and CIS using conditional logistic regression that was adjusted for age, education, hypertension, diabetes, waist-to-hip ratio, physical inactivity, current smoking status, heavy drinking and use of oral estrogen. They also looked at the impact of PFO on the correlation between migraine and CIS using logistic regression in a subgroup via a transcranial Doppler bubble screen.

To validate the screener, the researchers included 50 patients (median age, 41 years; women, n = 23) and 50 control participants (median age, 42 years; women, n = 24) who supplied a full set of responses to the screener. They found that the agreement was excellent for any migraine, MA and MO in patients and controls, though screener sensitivity for any migraine and MA was lower for controls than for patients. Further, screener sensitivity for MO was lower than for MA in both patients and controls.

The findings revealed an association between CIS and any migraine (OR = 2.48; 95% CI, 1.63-3.76) and CIS and MA (OR = 3.5; 95% CI, 2.19-5.61), but no relationship between CIS and MO, compared with patients who did not experience migraine. The researchers found this association in both women (OR = 2.97 for any migraine [95% CI, 1.61-5.47]; OR = 4.32 for MA [95% CI, 2.16-8.65]) and men (OR = 2.47 for any migraine [95% CI, 1.32-4.61]; OR = 3.61 for MA [95% CI, 1.75-7.45]).

An analysis that adjusted for demographic factors also showed a link between any migraine and CIS as well as MA and CIS both in men and women. These results remained robust in further analyses adjusted for vascular risk factors. In an exploratory subgroup analysis among women, the researchers observed no interaction between MA and estrogen use, current smoking status, obesity and heavy alcohol use. Subgroup analyses suggested a lower stroke risk with MA in men who were current smokers and engaged in heavy alcohol use.

Martinez-Majander and colleagues also found that the relationship between CIS and MA specifically remained significant regardless of PFO. The rate of MA increased as the magnitude of the right-to-left shunt increased in patients with PFO.

“One of the factors explaining the strong association between MA with CIS particularly in young adults could be their high frequency of PFO,” the researchers wrote. “Paradoxical embolism, serotonin-induced platelet activation and transient hypoxemia leading to increased expression of plasminogen activator-1 and suppression of fibrinolysis have been suggested to be specific factors linking ischemic stroke, PFO and migraine, with substantial uncertainty, however. Our study, with the increasing prevalence of MA with increasing RLS magnitude, supports the theory that PFO and MA are interconnected in the pathogenesis of CIS.”

This study is the first to conclusively show that MA “is a vigorous player underlying CIS among young adults, regardless of sex and independent of vascular risk factors,” according to Martinez-Majander and colleagues, who noted that the association also remained robust regardless of PFO.

“Future studies should explore in depth the mechanisms and associated features increasing the risk of CIS in patients with MA, such as activity, recent onset and lifetime burden of migraine attacks, specific cardiovascular structural and functional properties, coagulation abnormalities and shared genetic underpinnings,” the researchers wrote.