Causal relationship between dementia and delirium: Insights from a bidirectional two-sample Mendelian randomization analysis

Our previous study found dementia as a significant risk factor for delirium development in elderly patients with hip fracture. However, the causal relationship between dementia and delirium remains unclear. To assess the causal relationship between delirium and dementia, we conducted a bidirectional Mendelian randomization (MR) analysis. Inversevariance weighted (IVW), weighted median, MR Egger, weighted mode, and simple mode were employed to conduct the MR analysis. Heterogeneity was assessed using the Cochran Q statistic in MR-Egger and IVW methods. Horizontal pleiotropy was examined via the MR pleiotropy residual sum and outliers (MR-PRESSO) and MR-Egger intercept tests. The forward MR analysis revealed a significant association between unclassified dementia (1.604 (1.326–1.941), p = 1.12 × 10−6), Alzheimer's disease (1.259 (1.128–1.405), p = 4.10 × 10−5), and dementia with Lewy bodies (1.121 (1.026–1.225), p = 0.011) with an increased risk of delirium. In the reverse MR analysis, delirium was also suggested to increase the risk of unclassified dementia (1.133 (1.066–1.204), p = 6.31 × 10−5) and vascular dementia (1.246 (1.075–1.444), p = 0.003). These significant results were further validated in the multivariable MR analysis. No evidence of heterogeneity or horizontal pleiotropy was observed (p > 0.05). (1) Limited to European populations. (2) Sample population overlap between delirium and dementia. (3) Not all dementia subtypes were causally associated with delirium. This study provides genetic evidence supporting a causal relationship between dementia and delirium, indicating that dementia may influence the risk of delirium while delirium may also increase the risk of dementia.