Repeated Migraine Attacks Raise Risk of Ischemic Brain Lesions

Daniel M. Keller, PhD

September 14, 2009

September 14, 2009 (Philadelphia, Pennsylvania) — A large Dutch study of patients with migraine has shown an increased risk for ischemic brain lesions and white matter lesions compared with control patients; the risk is especially pronounced in migraine without aura.

"The more attacks you have had, the more likely it is that you will have brain lesions," lead investigator Michel Ferrari, MD, PhD, told attendees when he delivered the Special Lecture here at the 14th International Headache Congress. Dr. Ferrari is professor of neurology and chair at Leiden University Medical Center in the Netherlands.

This study, Cerebral Abnormalities in Migraine, an Epidemiological Risk Analysis II (CAMERA II), was a 9-year magnetic resonance imaging (MRI) follow-up of 286 patients from CAMERA I, a 1-year cross-sectional study of 435 migraineurs. Dr. Ferrari presented the crude unadjusted data, as the first analysis of the CAMERA II data was completed only last week.


All of the original CAMERA I subjects were invited to participate in the follow-up. Most of the people who declined (n = 149) did so because of lack of interest, refusal to undergo MRI, or illness. Nonparticipating migraineurs tended more to be smokers and to have a lower educational level and a more unfavorable cardiovascular risk profile compared with those who participated. Nonparticipating control patients just tended to be older.

Within the CAMERA II study population, migraineurs (n = 203) were a bit older than the 83 control participants (57 vs 55 years, respectively; P = .032) and had a higher daily use of analgesics (12% vs 2%; P = .012). The migraine group was made up of 71% women vs 69% for the control group, which was not a statistically significant difference.

The investigators used MRI scanners similar to the 1.5-Tesla ones used in the CAMERA I study in 1999 to 2000. "We did that because if you would start using now a 3-Tesla modern machine and then compared it with what was being used 9 years ago, you would obviously find more abnormalities, so we decided not to do that," Dr. Ferrari explained.

Deep white matter lesions progressed during the 9 years, but for each time point (CAMERA I or CAMERA II), there were no significant differences in the volume of deep white matter lesions between any of the groups — controls, all migraine, or migraine with or without aura — when both sexes were analyzed together.

However, there was an effect of sex, with no differences for the men but a difference for women in both volume of deep white matter lesions between groups and progression of lesions over time. Female migraineurs in the CAMERA II study had larger lesions than control participants (0.38 vs 0.23 mL, respectively; P = .04). This finding was especially pronounced for female migraineurs without aura (0.43 mL; P = .002 vs controls). Sample sizes were controls, 55; migraine with aura, 81; migraine without aura, 64.

Similarly, female migraineurs without aura had the greatest progression of lesion volumes during 9 years. There were no differences in periventricular white matter lesions and no interaction by sex.

Among the 203 migraineurs in the study, the number with any brain or posterior circulation territory lesion increased from 18 to 31 during the 9 years, and the number with posterior circulation territory infarcts went from 11 to 18 vs little or no increase for control participants. Contrary to a few previous reports of lesions disappearing during time, Dr. Ferrari said, "None of the lesions we had observed in the cerebellum in CAMERA I had they were persistent lesions." The number of new posterior circulation territory lesions was 10 for the migraine group compared with 0 for controls (P = .039).

Migraineurs with an infarct (n = 18) tended to have worse cardiovascular risk profiles than subjects without an infarct (n = 184). The former group was significantly older; they had more hypertension; more of them used statins, antihypertensive drugs, and platelet inhibitors; and more of them had a history of clinical stroke. Dr. Ferrari said subsequent analyses will be done to control for these risk factors. All lesions detected were subclinical, and analyses of cognitive testing are also yet to be performed.

Dr. Ferrari told Medscape Neurology that the clinical utility of this result is that migraine patients are often seen in the clinic with white matter lesions, and clinicians do not know what to do about them. This can lead to a range of tests for multiple sclerosis, cerebral autosomal dominant arteriopathy with subcortical infarcts, leukoencephalopathy, and other conditions that are not necessary. "I think it's important to emphasize that just plain migraine can cause these effects, and that you don't need to go through all kinds of investigations for diseases which are even more severe," he advised.

"Prudent" and "Aggressive" About CVD Risk

Commenting to Medscape Neurology on the association of cardiovascular risk factors and the risk for infarcts, Fred Sheftell, MD, president of the American Headache Society and founder and director of the New England Center for Headache in Stamford, Connecticut, said, "We have to really be prudent and careful and aggressive about reducing cardiovascular risk factors in our patients, and I think the other bottom line is women who have migraine with aura should not be on birth control pills."

Dr. Sheftell noted that migraine patients in general are more prone to cardiovascular disease in early and late life, with an elevated risk for ischemic stroke in early life and other cardiovascular problems later on, such as hypertension and hyperlipidemia. "Given the fact that migraine is a proinflammatory disorder in part and many of the cardiovascular diseases are also proinflammatory, then maybe [there is] some comorbidity there that needs to be looked at from a mechanistic point of view," he said. "I think that migraine is certainly a lot more than the attack itself, and there's a lot more to it than just the headache itself."

He cautioned, however, that patients "shouldn't get frightened to death" by these findings, noting that all the lesions Dr. Ferrari found were subclinical, and the numbers of lesions and affected patients were not very large.

Attack Frequency

The CAMERA II study supports the findings of previous studies and meta-analyses that also showed a higher incidence of ischemic brain lesions and white matter lesions among migraineurs. The original CAMERA I cross-sectional study found a greater incidence of MRI infarcts in migraine, but only in the posterior brain circulation and only for migraine with aura (8.1% incidence vs 0.7% in control participants; P = .002; odds ratio [OR], 13.8; Kruit MC et al. JAMA. 2004;291:427–434).

"The really interesting thing was this relationship with attack frequency," Dr. Ferrari told the audience. "The higher the attack frequency, the greater the risk of finding a posterior circulation infarct."

Compared with control patients, patients with migraine without aura with fewer than 12 attacks per year had an OR of less than 1, but having more than 12 attacks per year resulted in an OR of 7.1. The presence of aura raised the risk substantially, with an OR of 18.1 for those with fewer than 12 attacks per year, increasing to an OR of 25.4 for those with more than 12 attacks per year. Similarly, for white matter lesions, the researchers found that the higher the attack frequency, the greater the incidence of these lesions.

Dr. Ferrari also cited the 25-year longitudinal Age Gene/Environment Susceptibility Reykjavik Study (Scher A et al. JAMA. 2009;301:2563–2570). Subjects were interviewed in midlife (mean age, 51 years) and then reinterviewed and given MRI scans 25 years later. People without headaches in midlife were used as control patients.

Compared with these control participants, people with nonmigraine headaches or with migraine without aura were at no greater risk for infarcts later, but patients who had migraine with aura had an OR of 1.6 (95% confidence interval [CI], 1.3 – 2.2). There was a significant interaction by sex (P < .05): Men had an OR of 1 (95% CI, 0.6 – 1.8), whereas the OR was almost 2 for women (OR, 1.9; 95% CI, 1.4 – 2.6). The effect was independent of the usual cardiovascular risk factors.

Another meta-analysis of 7 clinic-based studies showed an overall OR for white matter lesions among migraineurs of 3.90 (95% CI, 2.26–6.72). Of migraineurs, 23% had white matter lesions vs 7% of control participants (P < .001; Swartz RH et al. Arch Neurol. 2004;61:1355–1368).

The study, through the Leiden Migraine Neuroimaging Programme, received support from Leiden University Medical Centre, the U.S. National Institutes of Health, the Netherlands Organisation for Scientific Research, the European Union, the Netherlands Heart Foundation, GlaxoSmithKline, Merck, Pfizer, and the Netherlands Brain Foundation. Dr. Ferrari has disclosed receiving research funding, honoraria, income for advisory board participation, and/or consultant fees within the past 5 years from the Migraine Trust (UK), Eisai, Pfizer, CoLucid, Allergan, Menarini, Pharmacia, Boehringer, Medtronic, St. Jude Medical GlaxoSmithKline, Bristol Myers-Squibb, Johnson and Johnson, Merck Sharp and Dohme, and Almirall Prodesfarma/Teva. Dr. Fred Sheftell has received payments for conducting clinical research and has received honoraria from Merck, GlaxoSmithKline, MAP Pharmaceuticals, Pfizer, and NuPath; grants to fund research; and compensation for advisory board participation and consultation from Merck, GSK, MAP, Pfizer, and NuPath.

14th International Headache Congress: Special Lecture. Presented Friday, September 11, 2009.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.