Genetically proxied liability to migraine and risk of intracranial aneurysm and subarachnoid hemorrhage

BackgroundObservational studies have reported inconsistent relationships between migraine and the risk of intracranial aneurysm and subarachnoid hemorrhage (SAH).ObjectiveTo determine whether genetic liability to migraine is associated with risk of intracranial aneurysm and aneurysmal SAH.MethodsThis study was designed as a two-sample Mendelian randomization (MR) analysis. Genetic associations with migraine were obtained from a large-scale meta-analysis of five population and clinic-based genome-wide association studies of migraine (102,084 cases and 771,257 controls of European ancestry). Genetic associations with intracranial aneurysm and SAH were obtained from a meta-analysis of 22 population-based genome-wide association studies (7495 cases and 71,934 controls of European ancestry). Findings were corroborated by sensitivity analyses and replicated in an independent sample of combined cases of intracranial aneurysm and SAH (3529 cases and 234,948 controls of Asian ancestry from the Biobank Japan and China Kadoorie Biobanks). In secondary analyses, we investigated the outcomes of extracranial aneurysm in the FinnGen and UK Biobank cohorts (up to 7466 combined cases of thoracic and abdominal aortic aneurysm).ResultsGenetic liability to migraine was associated with increased risk of the combined outcome of unruptured intracranial aneurysm and SAH (odds ratio [OR] of outcome per doubling in migraine liability 1.19, 95% confidence interval [CI] 1.06-1.35; p = 0.005). This finding was replicated in an independent sample (OR 1.15, 95% CI 1.02-1.30; p = 0.027), and there were similar associations across the component outcomes of unruptured intracranial aneurysm (OR 1.20, 95% CI 1.01-1.42; p = 0.035) and SAH (OR 1.18, 95% 1.04-1.33; p = 0.008). These findings were consistent in sensitivity analyses robust to violations of the MR assumptions. In reverse MR analyses, genetic liability to intracranial aneurysm was not associated with migraine (OR 1.03, 95% CI 0.99-1.07; p = 0.141). In a secondary analysis, there were similar associations of genetic liability to migraine with all forms of aortic aneurysm (OR for combined thoracic and aortic aneurysm 1.18, 95% CI 1.10-1.27; p = 8.49 x 10-6).ConclusionGenetic liability to migraine was associated with increased risk of intracranial and extracranial aneurysms, supporting a causal relationship between liability to migraine and these traits. Further work is needed to identify the biological mechanisms and clinical relevance of these findings.