Wednesday October 28 2009
Although migraines raised stroke risk it still remained low
Women taking the pill are more at risk of a stroke, the Daily Express has warned. The paper reports new research that shows that women who suffer from migraines double their risk if they take the pill. Several news sources have covered this research, although each has focused on different aspects of its findings.
The study behind these stories is a systematic review analysis of studies that estimated the association between migraine, stroke and events such as heart attacks. The research found that any type of migraine was associated with stroke but not other events. Migraines with aura (visual distortions, such as seeing flashing lights) were responsible for this link, as highlighted in the BBC’s coverage.
The study’s authors advise that being under 45, female, smoking or a user of oral contraceptives may be stroke risk factors. The review was well-conducted, but there are shortcomings associated with pooling the results from studies with different methodologies. This research adds weight to a growing body of evidence suggesting that the risk of stroke is greater in people with migraine.
Although the the increase in stroke risk associated with migraine with aura seems high, the overall risk remains extremely low in this population, which should offer some reassurance to people who get migraines.
Where did the story come from?
The research was carried out by Drs Markus Schurks and colleagues from various academic and medical institutions in the USA, France and Germany. The study was funded by a grant from the Merck pharmaceutical company, which was declared to play “no role in the study design or in the collection and analysis of the data”. The research was published in the peer-reviewed British Medical Journal.
What kind of scientific study was this?
This was a systematic review and meta-analysis of observational studies to investigate the association between migraine and cardiovascular disease, including stroke, heart attack and death from cardiovascular causes.
The researchers combed a variety of well-known medical databases for studies published up until January 2009. They only included studies that had a case-control or cohort design, and those that were cross-sectional but had only analysed cardiovascular events that occurred after the onset of the migraine. Their report only analysed studies where the main purpose was investigating the association between a migraine (or suspected migraine) and cardiovascular events.
The researchers pooled the findings from the gathered studies, using meta-analysis to derive an estimate of the odds associated with cardiovascular events and incidence of migraine. One way to determine how appropriate it was to pool this data is to estimate the heterogeneity between their study designs, (i.e. how different they are from one another). Some differences between studies would be expected to occur naturally, by chance.
If studies are very heterogeneous, then it may be appropriate to pool their results and run further analysis using complex statistical processes. These include a technique called meta-regression, which can be used to explore the reasons for the differences between studies.
The results from their meta-analyses are broken down into a number of different subgroups, such as men versus women, different types of cardiovascular events, women currently using oral contraceptives and smokers versus non-smokers. As a result of this analysis by subgroup there are several findings from this study that could have been reported on. This explains the different angles explored in various newspapers.
What were the results of the study?
The study found that any type of migraine was associated with a near-doubling of the risk of ischaemic stroke [RR 1.73, 95% CI 1.31 to 2.29].
There was no association between migraine and heart attack, or between migraine and death due to cardiovascular disease. Exploration of populations included in the studies showed that the association of ischaemic stroke and any migraine seemed to be stronger for people under 45, women, smokers and those using oral contraceptives.
The research divided studies into those specifying migraine that occurred with and without aura (visual distortions). It found that only migraine with aura was associated with an increased risk of ischaemic stroke.
What interpretations did the researchers draw from these results?
The researchers say that the most consistent evidence from their study was the increase in risk of ischaemic stroke in people with migraine, and that this seemed to be driven in particular by migraine with aura. Given that risk is raised for certain subgroups, they say that “young women who have migraine with aura should be strongly advised to stop smoking, and methods of birth control other than oral contraceptives may be considered”.
What does the NHS Knowledge Service make of this study?
This systematic review and meta-analysis provides high quality evidence for the association between migraine and strokes.
There are several limitations to the study, many of which the researchers themselves highlight:
- There was significant heterogeneity between the studies included in several of the sub-analyses here. This may indicate that it is not appropriate to pool the findings in this way.
- The researchers highlight the fact that the methods of diagnosing migraine differed across the trials, with some using self-administered questionnaires, headache checklists and insurance databases. Some studies did not differentiate between migraines with and without auras. These are all potential sources of heterogeneity that may have made pooling the results through meta-analysis less appropriate.
- There were a number of subgroup analyses carried out in this research. It is possible that some of the significant findings were due to chance alone.
- As the researchers included case-control studies they were unable to determine the absolute risks (the actual rates of stroke in these populations). They do say that other cohort studies suggest that the absolute risks for cardiovascular disease among people with migraine are “considerably low”. One 2007 study of women suggested that after adjusting for age, there were 18 additional vascular events attributable to migraine with aura per 10,000 women per year, less than 0.2%.
The researchers say that their findings agree in the main with those of another meta-analysis, which found a twofold increased risk of ischaemic stroke among people with migraine, and that this risk was greater in people aged under 45 and in women taking oral contraceptives. Their finding that this was only apparent in people with migraine with aura was different from that in the previous study, which found a similar risk both with and without auras.
From a clinical perspective, the researchers note that firm evidence on the association between migraine and other ischaemic vascular events is lacking. They say that patients should be “treated the same as any other patient without migraine: they should be screened for traditional cardiovascular risk factors, such as hypertension, adverse lipid profile, and increased risk of coronary heart disease, and if appropriate, these risk factors should be modified”.