Migraine and vascular disease

In the linked paper (doi:10.1136/bmj.i2610), Kurth and colleagues examine the link between migraine and cardiovascular events by using data from the Nurses’ Health Study II. The study followed a large cohort of female nurses for more than 20 years, 17 531 of whom reported a physician’s diagnosis of migraine at baseline. When compared with participants without migraine, those with migraine had an elevated risk of myocardial infarction, angina, or coronary revascularization and of cardiovascular mortality. Their relative risk for the combined outcome of all those events or stroke was roughly 50% higher than that for women without migraine, an effect that was not modified by age.1

Not all of these findings are new. The elevated risk of stroke in people with migraine is well established, for example. This increased risk of stroke is driven by and probably limited to the roughly 30% of people with migraine who experience aura—a focal neurologic event that usually precedes the headache. Migraine is associated with a roughly twofold increase in the risk of ischemic stroke and a 1.5-fold increased risk of hemorrhagic stroke. The risk is further elevated in the presence of other risk factors such as smoking or the use of estrogen containing contraceptives.2

The association between migraine and other vascular outcomes, including myocardial infarction, is less well understood, with studies showing conflicting results.3 These new data thus strengthen the view that migraine is a risk factor for vascular disorders beyond those that affect the brain. The elevated risk of cardiovascular events is not clearly linked to disease activity, as it persists in older age, when headaches have usually improved or disappeared. Information was not collected about aura, so this study cannot answer the important question of whether, as with stroke, this risk is confined to the subgroup of those who have aura. In addition, as all participants were women, it is uncertain whether cardiovascular risk also is elevated in men who have migraine. Both things seem likely, however.

How can a condition of the brain affect the heart? Some evidence points to endothelial dysfunction or abnormal vascular reactivity as pathophysiological mechanisms that might increase the predisposition to both the neurologic events of aura and vascular disease. Endothelial progenitor cells seem to be reduced in people with migraine in general, and endothelial inflammatory markers are elevated in those with migraine with aura.4 It is uncertain whether phenotypic features such as duration of disease or frequency of aura modify the association between migraine and vascular disease or whether it is altered by commonly used migraine treatments such as β adrenergic blockers.

The study findings have several implications. Firstly, it is time to add migraine to the list of early life medical conditions that are markers for later life cardiovascular risk.5 The magnitude of the risk should not be over-emphasized, however. It is small at the level of the individual patient but still important at a population level because migraine is so prevalent. Furthermore, at least in the United States, prevalence of migraine is especially high in certain minorities and lower socioeconomic groups, who commonly have more cardiovascular risk factors.6 7

Secondly, it is important to understand how the vascular risk attributable to migraine interacts with other risk factors. This is particularly true in the case of commonly used drugs that are known to cause or are suspected of causing adverse vascular outcomes. A special worry is whether the combined effects of two risk factors may result in multiplicative, rather than additive, risks. This question should be borne in mind during the phase III migraine prevention trials now underway for antibodies to calcitonin gene related peptide (CGRP) or its receptor and in post-marketing safety surveillance of these drugs if approved.8 CGRP is a potent systemic vasodilator. On the basis of results from animal studies, one expert has raised concern about a possible increase in risk of vascular events during long term CGRP blockade.9 This is of special importance because migraine is a condition of long duration. Many patients use preventive migraine treatment for decades.

Finally, the results of this study raise many questions about treatment. Do treatments that decrease the frequency or severity of aura or headache reduce later life vascular risks? Should patients with migraine be treated with statins or aspirin? We should be cautious about assuming that these interventions will benefit people with migraine, despite their intuitive appeal. An exploratory subgroup analysis from the Women’s Health Study, for example, showed that women with migraine with aura who took aspirin actually had an increased risk of myocardial infarction. Thus, what little evidence we do have suggests the need for therapeutic restraint until we have a better understanding of the mechanisms underlying the link between migraine and vascular disease.

At present, migraine is probably best thought of as a situation in which the medical urge to “do something” (beyond currently recommended assessments for cardiac risk and advocating a healthy lifestyle) should be resisted, especially when migraine is a patient’s sole risk factor. As Will Durant has observed, “One of the lessons of history is that nothing is often a good thing to do.”