Migraine and heart risk

The Daily Telegraph has reported that women with certain genes and blurred or difficult vision during migraines have a doubled risk of heart attack and stroke.These findings come from a US study on 25,000 women aged over 45, which looked at their migraine history, the presence of certain genes and their risk of cardiovascular problems.

Although this study did find that one particular group of the women did face a doubled risk of cardiovascular problems, it should be highlighted that even this doubled risk remains relatively low. Readers might also misinterpret the news article to mean that of the 1,275 women in this high risk group, 625 went on to have these problems. However, there were actually 625 first-time major cardiovascular episodes in the 25,000 women in the study as a whole.

This large study has some limitations, and more studies will be needed to confirm the findings, but it does provide impetus for future research. Meanwhile, the advice to women with migraines remains the same as for the general population: cardiovascular risk can be reduced through a healthy diet, not smoking and participating in physical activity.

Where did the story come from?

Dr Markus Schürks and colleagues from Brigham and Women’s Hospital and Universities carried out this research. The study was funded by the Donald W. Reynolds Foundation, the Leducq Foundation, the Doris Duke Charitable Foundation, F. Hoffmann La-Roche and Roche Molecular Systems, Inc., Deutsche Forschungsgemeinschaft, the National Cancer Institute and the National Heart, Lung, and Blood Institute. The study was published in Neurology, the peer-reviewed medical journal.

What kind of scientific study was this?

This study was a cross-sectional analysis of data looking at the relationship between migraine, risk of cardiovascular disease (CVD) and a particular genetic variant. The data for this analysis came from the Women’s Health Study (WHS), a large randomised controlled trial looking at the effects of aspirin and vitamin E on risk of CVD and cancer in healthy women.

The genetic variant that was looked at is known as the ACE D/I polymorphism. A polymorphism is a section of DNA that can occur in different forms. In this case, the polymorphism can occur in one of two different variations, either a small missing section of DNA (called a deletion or ‘D’) or an additional piece of DNA (an insertion or ‘I’). Each person has two copies of this gene, and each one can carry either a D variation or an I variation. This means an individual may possess two D copies, two I copies or one of each.

These particular deletions and insertions were of interest to the researchers because they lie in the gene that gives the cells instructions for making angiotensin converting enzyme (ACE). Drugs that stop ACE working are used to treat both CVD and migraine. The researchers thought that variation in the ACE gene might affect risk of these two conditions.

The researchers identified 25,000 eligible women taking part in the WHS. These women were all health workers aged 45 years or over. Before the start of the study the women had filled out questionnaires about their lifestyles and CVD risk, including questions about whether they had migraines and auras. Auras are sensations such as visual disturbances that are linked to the onset of migraine. Women who had experienced migraines in the past year were classed as having “active” migraines.

To be eligible for the current study women had to have been tested for the ACE D/I polymorphism, to have provided information about migraines, to have reported not having had a CVD event before the start of the study and to be Caucasian.

Women were followed for an average of 11.9 years in the WHS study, and the researchers identified those women who reported having a CVD event in this period. CVD events included having a heart attack or stroke, as well as cardiovascular deaths. Medical records were checked for women who reported having such an event, and for women who died, autopsy reports, death certificates or information from next of kin or relatives was obtained. Physicians reviewed these records and information, and diagnosed events based on standard criteria.

The researchers then looked to see whether there was any association between having the ACE D/I polymorphism and whether a woman had migraine (with or without aura) or experienced major CVD events (nonfatal heart attack or ischaemic stroke, or death from CVD). They did this by looking to see whether any particular form of the polymorphism was more common in those with migraine or major CVD event.

The researchers also looked at whether there was an association between migraine and major CVD events among women with different combinations of the ACE D/I polymorphism (those with two D copies, two I copies or one of each).

In their analyses the researchers adjusted for factors that could affect results, such as age, body mass index, smoking, alcohol use, exercise, having diabetes or high blood pressure, family history of heart attack before the age of 60, and use of hormone replacement therapy or oral contraception. They also adjusted for what randomised treatment the women were receiving as part of the WHS, but this did not affect results.

What were the results of the study?

At the start of the study, approximately 18% of the women (4,577 women) reported having migraines in the past, and about 13% had experienced migraines in the past year. Of the women who had had migraines in the past year (the “active” migraine group), almost 40% experienced auras.

Among the women studied, 29% had two D copies of the ACE D/I polymorphism, 25% had two I copies and 46% had one copy of each form. There was no association between the women’s ACE D/I polymorphism and migraine with or without aura.

During follow-up, there were 625 first-time major CVD events. There was no association between the ACE D/I polymorphism and risk of major CVD events. Over follow-up there were;

504 major CVD events among the 20,423 women with no history of migraines,
121 events among the 4,577 women with any history of migraines,
48 events among the 1,275 women who had migraines with auras in the past year, and
32 events among the 1,951 women who had migraines without auras in the past year.

The researchers found no significant association between active migraine without auras or having had migraines in the past but not in the past year and risk of major CVD event during follow-up. However, having active migraines with auras at the start of the study doubled the risk of major CVD events compared with women with no history of migraines.

When the researchers further divided the women according to what combinations of the ACE D/I polymorphism they carried, they found that this increase in risk was only seen in women carrying either one or two D copies of the ACE D/I polymorphism. However, overall statistical tests found that the effect of the polymorphism on the link between migraines with or without auras and risk of CVD events was not significant.

What interpretations did the researchers draw from these results?

The researchers conclude that their data did not suggest links between the ACE D/I polymorphism and migraine or CVD risk.

They say that women who experienced migraines with auras were at increased risk of CVD events, but that this was only the case among those women with a specific genetic makeup: those carrying one or two copies of the D form of the ACE D/I polymorphism. Because of the small number of CVD events within this specific group of women the researchers suggest that further research is needed to investigate this potential link.

What does the NHS Knowledge Service make of this study?

This study has a few limitations, some of which the authors discuss:

There were only a small number of CVD events within each of the compared subgroups, meaning the results for each group were more susceptible to chance. As the authors report, these findings need to be confirmed in larger studies.
Women self-reported any migraine and auras in a questionnaire, rather than using the criteria suggested by the International Headache Society, which may have led to their headaches being misclassified. For example, what one woman considers to be a migraine, another might consider to be just a headache, and this could affect results.
In addition, CVD events also appeared to be self-reported, and this could result in missing some events.
As with all studies of this type, it is possible that the results seen are being affected by a factor, or factors, other than those assessed. These are known as confounders. The researchers did take steps to adjust for potential confounders, which increases confidence in the results. However, other factors may still be having an effect.
The study only included Caucasian women, who were all health professionals. These results may not apply to men, women from other ethnic groups or those from different socio-economic backgrounds.
It is worth noting that the number of CVD events was relatively low. Although the reported doubling in risk may sound high, even a doubled risk of events would still not be particularly high.

This study provides impetus for future research, although whether the findings will have practical implications remains to be seen. As the authors point out, other population studies have found an association between migraine and ischaemic cardiovascular events, so this is not a novel finding.

The advice to women with migraines and auras remains the same as for the general population. They can reduce their cardiovascular risk by having a healthy diet, not smoking and participating in physical activity.