White matter hyperintensities in migraine

Case-control studies showed that migraine is associated with ischemic stroke. Studies assessing the prevalence of white matter lesions (WML) in migraine however, reported conflicting results, possibly attributable to selection bias by recruiting only severe cases. We conducted a population-based MRI study of migraineurs with aura (N=161), migraineurs without aura (N=134) and controls (N=140), 34-63 years of age. Among women, the presence of severe WML was increased in migraineurs (OR=2.0; 95% CI, 1.0-4.2); this risk increased with attack frequency and younger age, but did not differ by sub-type. No association was found in men. Introduction Migraine is a chronic, multifactorial neurovascular disorder, characterized by recurrent disabling attacks of severe headache, autonomic nervous system dysfunction, and in up to one-third of patients, neurological aura symptoms. At least 6% of the males and 15% of the females in the adult general population have at least one migraine attack per year. Little is known about the effects of migraine attacks on the brain, although case-control studies have shown that migraine is a risk factor for ischemic stroke, among young women. Clinic-based MRI studies also suggested an increased prevalence of cerebral WML in migraine patients.'" However, the evidence is conflicting, possibly reflecting selection bias in clinical studies only recruiting severe cases. We conducted an MRI study to examine the association of WML and migraine in a population-based sample.. Materials and methods The sample of patients and controls is from the GEM study, which is based on a cohort of 6039 Dutch adults aged 20-60 yrs, including 863 migraineurs, who participated in a population-based study of cardiovascular risk factors? Migraine was assessed according to IHS criteria5 in a multistage procedure that included a semi-structured clinical interview by telephone. Migraine diagnosis was conducted blind to an individual's profile of cardiovascular risk factors and disease, which were assessed by interview and clinical exam at the study center. From the cohort aged 30-60 yrs, we randomly selected for MR imaging of the brain, 134 migraineurs without aura (MO), 161 migraineurs with aura (MA) and 140 controls who were frequency matched to the cases by sex, 5-yr age strata and region of residence. Total brain MR images were acquired on a 1.0 T and a 1.5 T system with 48 contiguous 3 mm axial slices. Pulse sequences were comparable between the systems, and included a proton density and T2-weighted FSE and a FLAIR sequence. Blinded to clinical status, a neuroradiologist read all images to assess WML. Periventricular white matter lesions (PVWML) were rated on a scale ranging from 0 (no PVWML), 1 (pencil-thin lining), 2 (smooth halo or thick lining), to 3 (large confluent lesions), for frontal and posterior horns and bands; the three regional scores were added up (range 0-9). Deep (subcortical) lesion (DWML) were rated using a semi-quantitative scale that takes into account a fixed volume of small (<3 mm), medium (4-10 mm) and large (>I0 mm) WML. The total DWML-volume distribution was categorized into quintiles. Using logistic regression we compared the association of migraine to risk of DWML using the group with no lesions as the reference, Risk estimates were similar up to the fourth quintile. Therefore, we collapsed the first four quintiles and show the results for the upper quintile (high-DWML-load) versus the rest. We examined the risk (Odds Ratios [OR], 95% confidence intervals [CI]) for high WML load by migraine status (yedno), subtype of migraine (MA and MO vs. controls) and average number of attacks per month ( 4 and 21 vs. controls), and controlled for demographic and cardiovascular risk factors. Results There were no differences in the grades of severity of PVWML between the diagnostic groups (Table 1); this did not vary by sex. Fifty-nine percent of subjects had at least one DWML. There were no differences between migraineurs and controls when lesion size was not considered. Table 1' Controls MO MA Deep lesions (N=140) (N=134) (N=161) P Value 2 1 lesion 82(59) 84(63) 89(55) 0.44 High-DWML-load 23 (16) 33 (25) 34(21) 0.24 Periventricular score 0.43 0 30(21) 22(16) 35(22) 1-2 99(71) 95(71) 114(71) 3-6 l l(7.9) 17(13) 12(7.5) * Values are numbers of subiects (%);P values are from Pearsons's chi-square test (unadjusted) In gender-specific analyses, migraineurs had significantly more often high-DWML-load than controls (24% vs. 13%; P=0.03). After adjusting for confounders, among women, compared to controls, migraineurs were at twofold-increased risk of high-DWML-load for 1.0-4.31). Table 2 shows that the risk increased with increasing attack frequency (Puend<0.05) and younger age. There was no association of high-DWML-load to migraine in men (OR, 0.7 [95% CI, 0.3-1.81). This did not vary by subtype or attack frequency (data not shown). Table 2 All women Women Variable Nt OR(95% CI)* Nt OR(95% CI)* Migraine history both MO (OR, 2.1 [95% CI, 1.0-4.71) and MA (OR, 2.0 [95% CI, (N=3 17) <50 yrs (N=194) NOS 13/100 1.0 3/67 1.0 Yes 51/217 2.0 (1.0-4.2) 22/127 5.2 (1.4-19) Attacks /month < I 23/121 1.6 (0.7-3.4) 12/76 4.6 (1.1-18) 2 1 28/96 2.6 (1.2-6.0) 10/51 6.9 (1.6-30) t Number of subjects with severe deep white matter lesions: OR=Odds ratio. CI=confidence interval; $ Reference group for all analyses Discussion Small vessel vasculopathy, caused by the effects of cardiovascular risk factors, is commonly thought to underlie the pathology of WML. However, since we controlled for cardiovascular risk factors, our findings are more likely to suggest an independent role of migraine related factors in the pathogenesis of WML in migraineurs. Cardiac abnorinalities, changes in coagulation factors, and local changes during migraine attacks, such as excessive neuronal activation, neurogenic inflammation, neuropeptide release or excitotoxity, could contribute to an increased risk for structural brain changes in migraine. Whether prevention of migraine attacks, rather than just treating should have a higher priority, has to he discussed. ConclusionFemale, but not male migraineurs are at increased risk of WML. Theeffect increased with increasing attack frequency. References1. Igarashi H, et al. Cephalalgia 1991;11:69-74.2. Pavese N, et al. Cephalalgia 1994; 14:342-5. 3. De Benedittis G, et al. Headache 1995;35:264-8. 4. Launer LJ, et al. Neurology 1999;53:537-42.5. Classification and diagnostic criteria for headache disorders.Cephalalgia 1988;s Suppl7: 1-96.