Migraine increases CVD risk

A large data-mining study from Denmark found significant increases in many cardiovascular outcomes in young patients with migraine (see migraine inc cvd risk bmj2018 in dropbox, or doi.org/10.1136/bmj.k96).

Details:

--data from all hospitals (including ED visits) and outpatient clinics in Denmark, using routinely and prospectively collected administrative and medical registries, in a country with a national health insurance program (ie, these results reflect the whole population of Denmark), from 1995-2013 (19 year study)

--51032 patients with migraine, compared to 510,320 from the general non-migraine population, matched by age, sex, and calendar year

--median age 35, 71% women, 1.4% diabetes, 3.2% obesity, 0.2% hyperlipidemia, 3% hypertension, 1.3% COPD, 2.4% cancer [ie, a pretty young, low-risk group]

Results:

--incidence of outcomes, comparing migraine vs general population, absolute numbers followed by adjusted hazard ratio (HR):

   --25 vs 17 MIs; HR 1.49 (1.36-1.64)

   --45 vs 25 ischemic strokes; HR 2.26 (2.11-2.41)

   --11 vs 6 hemorrhagic strokes; HR 1.94 (1.68-2.23)

   --27 vs 18 venous thromboembolism; HR 1.59 (1.45-1.74)

   --47 vs 34 atrial fibrillation/flutter; HR 1.25 (1.16-1.36)

   --no meaningful association with peripheral artery disease or heart failure

--timing of relationship:

    ​--ischemic stroke outcomes had largest increase in the first year after migraine diagnosis, more slowly increasing thereafter

    -- others had slow gradual increase in CVD outcomes over time, especially after about 2 years

--subgroup analysis: higher overall associations in those having migraine with aura (though limited to ischemic stroke in the first year and MI throughout) and in women (though pretty minimal difference from men)

Commentary:

--migraine is remarkably common (about 15% of the population worldwide, 1 billion people, affecting mostly young people, with 90% having onset prior to age 50)

--this brings up 2 issues in interpreting the above findings:

    --the absolute numbers of people affected is small in the above study, but this finding still has significant public health impact given the huge number of patients with migraine

    ​--and, these are pretty young and healthy patients, mostly women, average age 35 and followed 19 years. the absolute cardiovasc effect of migraine is likely to be much higher as people age, and this is reinforced in this study since the cumulative incidence of migraine continues to increase for these outcomes over time (curves are splaying apart)

--this study is basically in concordance with several other studies finding migraine is associated with ischemic strokes and heart disease (eg, a 2015 meta-analysis found a 33% increased risk of MI in those with migraine), esp in women and in those with aura

--potential mechanisms include:  endothelial dysfunction, vasospasm, paradoxical embolism (higher prevalence of patent foramen ovale in those with migraine, esp with aura), hypercoagulability, platelet aggregation, use of drugs that increase cardiovasc disease (eg NSAIDs), and higher incidence of other cardiovasc risk factors in those with migraine (found in other studies; this one found similar increases in those with vs without migraine, but the actual numbers of people with risk factors was very small)

--there was a somewhat related study published recently finding that patients with migraine had a 75% increased risk of perioperative ischemic stroke, adjusted OR 1.75 (1.39-2.21), and was higher in those who had aura, adjusted OR 2.61 (1.59-4.29).  The baseline absolute risk was 2.4 ischemic strokes for every 1000 surgical patients, which increases to 4.3/1000 in those with migraine, but 3.9/1000 in those without aura and 6.3/1000 in those with aura. There was a 31% increase hospital readmission rate within 30 days in those with migraine (see migraine periop surg risk bmj2017 in dropbox, or doi: 10.1136/bmj.i6635)

--of note, the QRISK3, an upgraded  atherosclerosis risk calculator used in the UK, includes the presence of migraine in their algorithm (see http://gmodestmedblogs.blogspot.com/2017/08/a-new-atherosclerosis-risk-calculator.html​ )

--potential biases: as a large-scale data-mining study, we can only ascertain associations and not causality; there are always concerns about the accuracy and consistency of data collection in so many different areas (though they do note the positive predictive value of these outcomes in this registry is in the 90+% range); many of the drugs used to treat migraine have cardiac effects (b-blockers, calcium channel blockers, NSAIDs, tricyclics, etc); and there might be specific problems in distinguishing/labeling neurological symptoms as being stroke vs migraine

--as mentioned in other blogs, studies like this reflect back on us in the US, where we have such an incoherent medical system, with huge differences in access to care, especially in the uninsured (which is likely to increase significantly in the current political environment). This study (and those in other industrialized nations) highlights a health care system with not only universal healthcare coverage but really useful outcomes data to guide future policy/directions.​

So, this study does reinforce migraine as a cardiovascular risk factor. Which means that we probably should consider it in our general gestalt of cardiovasc disease susceptibility. Which means that we should perhaps be somewhat more aggressive in overall risk factor reduction in those who have migraine. Unanswered (but interesting) questions remain: does better control of migraine lead to less cardiovasc risk (eg does effective migraine prophylactic therapy mitigate the increased CVD risk)? what is the real mechanism by which migraine is associated with cardiovasc disease (eg, if it is through hypercoagulability, perhaps we should encourage exercise; and/or prescribe aspirin instead of a statin, especially since migraine is so prevalent in young women, and statins should not be used in women who may become pregnant)?

please feel free to circulate this to others. also, if you send me their emails at gmodest@uphams.org, i can add them to the list​​

 ​

 ​​