Until recent years it was believed that migraine with aura was a disorder causing intermittent neurological symptoms, with no impact on brain structure. However, recent MRI studies have reported increased cortical thickness of visual and somatosensory areas in patients with migraine with aura, suggesting that such structural alterations were either due to increased neuronal density in the areas involved, or a result of multiple episodes of cortical spreading depression as part of aura attacks. Subsequent studies have yielded conflicting results, possibly due to methodological reasons, e.g. small number of subjects. In this cross-sectional study, we recruited females aged 30-60 years from the nationwide Danish Twin Registry. Brain MRI of females with migraine with aura (patients), their co-twins, and unrelated migraine-free twins (controls) were performed at a single centre and assessed for cortical thickness in predefined cortical areas (V1, V2, V3A, MT, somatosensory cortex), blinded to headache diagnoses. The difference in cortical thickness between patients and controls adjusted for age, and other potential confounders was assessed. Comparisons of twin pairs discordant for migraine with aura were also performed. Comparisons were based on 166 patients, 30 co-twins, and 137 controls. Compared with controls, patients had a thicker cortex in areas V2 [adjusted mean difference 0.032 mm (95% confidence interval 0.003 to 0.061), V3A [adjusted mean difference 0.037 mm (95% confidence interval 0.008 to 0.067)], while differences in the remaining areas examined were not statistically significant [adjusted mean difference (95% confidence interval): V1 0.022 (-0.007 to 0.052); MT: 0.018 (-0.011 to 0.047); somatosensory cortex: 0.020 (-0.009 to 0.049)]. We found no association between the regions of interest and active migraine, or number of lifetime aura attacks. Migraine with aura discordant twin pairs (n = 30) only differed in mean thickness of V2 (0.039 mm, 95% CI 0.005 to 0.074). In conclusion, females with migraine with aura have a thicker cortex corresponding to visual areas and our results indicate this may be an inherent trait rather than a result of repeated aura attacks.

Migraine with aura is considered a putative risk factor for silent brain infarcts and white matter hyperintensities in women. In an MRI study of Danish female twins aged 30–60 years, Gaist et al. compare 172 affected women with 139 controls, but find no evidence to support the proposed associations.

Preclinical and clinical proof-of-concept studies have suggested the effectiveness of pharmacological modulation of inflammatory cytokines in ischemic stroke. Experimental evidence shows that targeting tumor necrosis factor (TNF) and interleukin (IL)-1 holds promise, and these cytokines are considered prime targets in the development of new stroke therapies. So far, however, information on the cellular expression of TNF and IL-1 in the human ischemic brain is sparse. We studied 14 cases of human post-mortem ischemic stroke, representing 21 specimens of infarcts aged 1 to > 8 days. We characterized glial and leukocyte reactions in the infarct/peri-infarct (I/PI) and normal-appearing tissue (NAT) and the cellular location of TNF, TNF receptor (TNFR)1 and TNFR2, IL-1α, IL-1β, and IL-1 receptor antagonist (IL-1Ra). The immunohistochemically stained tissue sections received a score reflecting the number of immunoreactive cells and the intensity of the immunoreactivity (IR) in individual cells where 0 = no immunoreactive cells, 1 = many intermediately to strongly immunoreactive cells, and 2 = numerous and intensively immunoreactive cells. Additionally, we measured blood TNF, TNFR, and IL-1 levels in surviving ischemic stroke patients within the first 8 h and again at 72 h after symptom onset and compared levels to healthy controls. We observed IL-1α and IL-1β IR in neurons, glia, and macrophages in all specimens. IL-1Ra IR was found in glia, in addition to macrophages. TNF IR was initially found in neurons located in I/PI and NAT but increased in glia in older infarcts. TNF IR increased in macrophages in all specimens. TNFR1 IR was found in neurons and glia and macrophages, while TNFR2 was expressed only by glia in I/PI and NAT, and by macrophages in I/PI. Our results suggest that TNF and IL-1 are expressed by subsets of cells and that TNFR2 is expressed in areas with increased astrocytic reactivity. In ischemic stroke patients, we demonstrate that plasma TNFR1 and TNFR2 levels increased in the acute phase after symptom onset compared to healthy controls, whereas TNF, IL-1α, IL-1β, and IL-1Ra did not change. Our findings of increased brain cytokines and plasma TNFR1 and TNFR2 support the hypothesis that targeting poststroke inflammation could be a promising add-on therapy in ischemic stroke patients.

No abstract