Background: Migraine and dementia are prevalent neurological disorders with overlapping pathophysiological mechanisms. The association between migraine and dementia risk remains debated, with conflicting findings from observational studies and emerging evidence from Mendelian randomization (MR) studies. Methods: This systematic review followed PRISMA guidelines. A comprehensive literature search was conducted in PubMed/MEDLINE, Embase, and Web of Science for publications from January 2021 to January 2026. Eligible studies included observational cohort studies and MR studies examining the association between migraine and dementia outcomes in adults. Risk of bias was assessed using the Newcastle-Ottawa Scale for cohort studies and a modified ROBINS-I framework for MR studies. A narrative synthesis was performed due to substantial heterogeneity. Results: Seven studies met inclusion criteria: five population-based cohort studies and two two-sample MR studies. Among cohort studies, four reported significant positive associations between migraine and dementia risk, with hazard ratios ranging from 1.21 to 1.37 for all-cause dementia, 1.29 to 1.31 for Alzheimer's disease (AD), and 1.21 to 1.35 for vascular dementia (VaD). One Swedish cohort study reported no significant associations. MR studies provided evidence supporting a causal relationship between genetically predicted migraine and increased AD risk (odds ratios 1.09–1.10), with thalamic atrophy identified as a partial mediator (28.2% of the total effect). Bidirectional MR analysis revealed that migraine increases AD risk while VaD increases migraine risk. Migraine subtype, aura status, and case definition influenced observed associations. Conclusion: Current evidence suggests migraine is associated with increased risk of dementia, particularly AD and VaD, with MR studies supporting causal relationships. Heterogeneity across studies highlights the importance of diagnostic methods, population characteristics, and dementia subtype specification. Future research should employ standardized diagnostic criteria, detailed migraine phenotyping, and investigate the potential neuroprotective effects of migraine management on cognitive outcomes.