Abstract Body: Introduction: Migraines are a known risk factor for stroke among reproductive-aged women, though the relation of migraines to stroke risk among postmenopausal women remains unclear. We assessed the association between a history of diagnosed migraine and incident stroke among postmenopausal women and investigated age differences in the association.

Methods: We included women enrolled in the Women’s Health Initiative (WHI), a large longitudinal cohort study of postmenopausal women in the U.S. We excluded women with a history of stroke or TIA at baseline or with missing data on history of migraine or key covariates. The primary exposure was a history at baseline of self-reported migraine diagnosed by a physician. The primary outcome was incident stroke (total, ischemic [IS], or hemorrhagic [ICH]). Multivariable Cox proportional hazards models were used to test the cause-specific hazard ratios (HRs) between migraine history and total stroke, IS (overall and by subtype), and ICH, sequentially adjusted for age and traditional and female-specific risk factors. Using multivariable Cox proportional hazard models, we then quantified the association between migraine history and total stroke by age in 5-year groups at baseline.

Results: 130,277 participants were included. The median age at baseline was 61 years (IQR 56-67 years) for those with a migraine history compared to 63 years (IQR 57-69 years) for those without. Overall, 5,743 strokes occurred over a median follow-up period of 19.9 years (IQR 9.1-25). In multivariable-adjusted models (Table), there was a marginal association between migraine history and total stroke (HR=1.07; 95% CI, 0.99-1.17) and a significant association between migraine history and IS (HR=1.12; 95% CI, 1.02-1.23), most pronounced in the cardioembolic category. Migraine was not associated with risk of ICH (HR=0.85, 95%CI, 0.67-1.09). Risk appears to be of greatest magnitude in early (HR 1.26, 95% 0.93-1.72) and late (HR 1.16, 95%CI 0.89-1.50) postmenopausal years (Figure), though the differences by age group were not statistically significant.

Conclusions: Over a 20-year follow-up period, postmenopausal women with self-reported histories of migraine had a higher risk of IS, but not total or ICH, suggesting that migraine history contributes to ischemic stroke risk during the postmenopausal years. Along with other known risk factors, migraine history should be considered in stroke risk factor screening and prevention efforts after menopause.