For years, clinicians have noted an apparent overlap between migraine and glaucoma. Both are common chronic conditions, both have vascular components, and several observational studies have suggested that people with migraine may be more likely to develop glaucoma. But association is not causation — and a new genetic analysis argues that migraine itself is unlikely to be a direct causal risk factor.
In a study published in Medicine, a team from The People’s Hospital of Yubei District of Chongqing, China, used bidirectional Mendelian randomization to examine whether migraine causes glaucoma, or whether glaucoma predisposes to migraine. Their conclusion is direct: the analysis found no evidence of a causal relationship in either direction.
The question is clinically relevant because previous literature has been mixed. Some studies have reported that those with migraines have a higher risk of developing glaucoma, including one cited by the authors showing a 1.2-fold higher risk after adjustment for demographic and socioeconomic factors. Others have suggested stronger links in chronic severe migraine or in patients younger than 50 years. A meta-analysis cited in the paper concluded that migraine was associated with a 24 percent higher risk of glaucoma. However, the paper also observes that studies in Chinese, Japanese, and Armenian populations have reported no significant association.
To address this uncertainty, the authors turned to Mendelian randomization, an approach that uses genetic variants as instrumental variables to test causal relationships. Because genetic variants are randomly allocated before birth, MR can reduce the confounding and reverse-causation biases that complicate conventional observational epidemiology.
The team used genome-wide association study data from the FinnGen database, restricting exposure and outcome cohorts to individuals of European ancestry to minimize population stratification bias. Migraine, migraine with aura, migraine without aura, glaucoma, open-angle glaucoma, and angle-closure glaucoma were all included in the bidirectional framework.
The sample sizes for the research were substantial: the migraine dataset included 184,654 individuals, including 8,547 cases; glaucoma included 218,792 individuals, including 8,591 cases; open-angle glaucoma included 4,433 cases; and angle-closure glaucoma included 588 cases.
Using inverse variance weighting as the primary analysis method, the investigators found no causal effect of overall migraine on overall glaucoma, open-angle glaucoma, or angle-closure glaucoma. The reported estimates were close to null: migraine to glaucoma OR 0.94, 95 percent CI 0.84–1.05; migraine to open-angle glaucoma OR 0.90, 95 percent CI 0.77–1.04; and migraine to angle-closure glaucoma OR 0.77, 95 percent CI 0.52–1.14. Migraine with aura and migraine without aura also showed no significant causal effect on glaucoma outcomes.
The reverse analyses told the same story. Neither glaucoma overall nor its subtypes showed evidence of a causal effect on migraine, migraine with aura, or migraine without aura. Sensitivity testing reinforced the null result: MR-Egger, inverse variance weighted heterogeneity tests, MR-PRESSO, funnel plots, and leave-one-out analyses did not suggest that the findings were driven by pleiotropy, heterogeneity, or a single genetic instrument.
The study authors do not dismiss biological overlap. Migraine and glaucoma both involve neurovascular regulation, and vascular dysregulation, impaired optic nerve head perfusion, trigeminovascular activation, inflammatory mediators, and genetic pathways such as matrix metalloproteinases have all been proposed as shared mechanisms. But shared biology does not necessarily mean one disease causes the other.
The study's focus on European ancestry restricts generalizability;further research in diverse populations is needed, the authors note. For ophthalmologists, however, the practical message is useful: migraine may remain relevant to a patient’s vascular history and symptom burden, but this genetic evidence does not support treating it as a causal glaucoma risk factor. Prior epidemiological associations may reflect confounding, shared mechanisms, or coincidence rather than direct causation.
