Causal Effects of Type 2 Diabetes and Glycemic Traits on Dementia and Stroke: A Mendelian Randomization Study including Imaging Endpoints

Aims/hypothesisType 2 diabetes mellitus (T2D) and glycemic dysregulation have been associated with cognitive decline and cerebrovascular disease, but causal relationships remain uncertain. We used Mendelian randomization (MR) to investigate the potential causal effects of T2D and glycemic traits, including hemoglobin A1c (HbA1c), fasting glucose (FG), fasting insulin (FI), and 2-hour glucose (2h-glu), on risk of dementia subtypes, stroke subtypes, and structural brain changes. MethodsWe conducted a two-sample MR analysis using summary statistics from large genome-wide association studies in European populations. Outcomes included all-cause dementia, Alzheimers disease (AD), vascular dementia (VaD), ischemic stroke, lacunar stroke, and five MRI-derived brain phenotypes (total brain volume, total white and grey matter volumes, hippocampal volume, and white matter hyperintensities (WMH)). The inverse-variance weighted method was the primary analysis; MR-Egger, weighted median, and Mendelian Randomization Pleiotropy RESidual Sum and Outlier (MR-PRESSO) were used for sensitivity analyses. Multivariable MR adjusted for body mass index (BMI). ResultsGenetic liability to higher HbA1c was associated with increased AD risk (odds ratio (OR) = 1.35, 95% CI: 1.08-1.68) and lower hippocampal volume ({beta} =-0.489, SE = 0.152). T2D was associated with increased risk of ischemic stroke (OR = 1.14, 95% CI: 1.11-1.17), lacunar stroke (OR = 1.15, 95% CI: 1.09-1.21), and lower grey matter volume ({beta} =-0.032, SE = 0.011). FG, FI, and 2h-glu were also associated with stroke subtypes. These associations persisted after BMI adjustment. No consistent associations were observed with VaD or WMH. Conclusions/interpretationThis study provides genetic evidence that HbA1c may causally contribute to AD and regional brain atrophy, while T2D and other glycemic traits are primarily linked to stroke risk. Incorporating brain MRI outcomes provides insight into distinct vascular and neurodegenerative pathways linking dysglycemia to brain health. Research in contextO_ST_ABSWhat is already known about this subject?C_ST_ABSO_LIType 2 diabetes (T2D) and hyperglycemia have been associated with increased risk of dementia, particularly Alzheimers disease (AD), in observational studies. C_LIO_LIObservational designs are prone to confounding and reverse causality, limiting causal inference. C_LIO_LIPrevious Mendelian randomization (MR) studies on glycemic traits and dementia risk have produced inconsistent results and rarely examined brain structural changes as intermediate outcomes. C_LI What is the key question?O_LIDo genetically predicted T2D and glycemic traits (hemoglobin A1c, fasting glucose, fasting insulin, 2-hour glucose) causally influence dementia risk, stroke subtypes, and neuroimaging biomarkers of brain structure? C_LI What are the new findings?O_LIGenetically predicted higher HbA1c levels were associated with increased risk of AD and with reductions in hippocampal and white matter volumes. C_LIO_LIT2D and other glycemic traits showed stronger associations with ischemic and small vessel stroke, suggesting vascular mechanisms. C_LIO_LIThis study combines clinical and imaging outcomes to distinguish neurodegenerative from cerebrovascular pathways in diabetes-related brain changes. C_LI How might this impact clinical practice in the foreseeable future?O_LIThe findings underscore the importance of optimizing long-term glycemic control even before the onset of T2D to reduce the risk of neurodegenerative and vascular brain pathology in individuals with diabetes. C_LI