Causal relationships between delirium and Alzheimer's disease: a bidirectional two-sample Mendelian randomization study

Abstract

Abstract Background Previous observational studies have reported that delirium has an association with an increased risk of Alzheimer's disease (AD), and that patients with AD have a higher risk of developing delirium. However, due to the limitations of observational study, it is challenging to confirm whether delirium has a causal effect on AD or reverse causation exists. Methods A bidirectional two-sample Mendelian randomization (MR) was performed to investigate the relationship between delirium and AD. Summary statistics from genome-wide association studies of delirium and AD phenotypes were utilized. Inverse-variance weighted (IVW) method was used as the main analysis approach, and additional analyses were performed using MR Egger, weighted median, simple mode and weighted mode to ensure result accuracy. Heterogeneity and horizontal pleiotropy were assessed using Cochran's Q statistics and MR Egger intercept, separately. Results The MR analyses showed that genetically predicted delirium was not associated with AD (IVW: odds ratio [OR] 0.98, 95% CI 0.91–1.05, P = 0.544; MR Egger: OR 0.98, 95% CI 0.83–1.15, P = 0.780; weighted median: OR 0.96, 95% CI 0.88–1.05, P = 0.323; simple mode: OR 0.91, 95% CI 0.80–1.04, P = 0.212; weighted mode: OR 0.93, 95% CI 0.83–1.05, P = 0.277). However, in the reverse direction, AD was associated with delirium (IVW: OR 1.32, 95% CI 1.13–1.54, P = 3.91E-04; MR Egger: OR 1.42, 95% CI 1.02–1.98, P = 5.60E-02; Weighted median: OR 1.39, 95% CI 1.18–1.63, P = 8.22E-05; Simple mode: OR 1.41, 95% CI 1.10–1.80, P = 1.41E−02; Weighted mode: OR 1.39, 95% CI 1.16–1.67, P = 3.23E-03). Conclusion Based on the results of our MR study, there is no bidirectional causality between delirium and AD, delirium is not associated with an increased risk of AD, while genetically predicted AD is a potential causal risk factor for delirium.

Keywords

• Alzheimer’s disease
• Delirium
• Mendelian randomization
• Causal relationships
• European