Abstract
Observational studies have documented association between rapid eye movement sleep behavior disorder (RBD) and neurodegenerative diseases, but the causal relationship remains to be established. In this study, we utilized a bidirectional 2-sample Mendelian randomization (MR) approach to assess the potential causal connection between RBD and 6 neurodegenerative conditions, including Alzheimer disease (AD), amyotrophic lateral sclerosis (ALS), dementia with Lewy bodies (DLB), frontotemporal dementia (FTD), multiple system atrophy (MSA), and Parkinson disease (PD). Inverse variance weighting (IVW) was performed as the main MR analysis, and additional 4 MR methods were performed to reinforce the robustness of the final MR estimates. Sensitivity analyses were conducted to detect possible heterogeneity or pleiotropy. Ten single-nucleotide polymorphisms (SNPs) for RBD, 54 SNPs for AD, 11 SNPs for ALS, 4 SNPs for DLB, 3 SNPs for MSA, and 12 SNPs for PD were selected as instrumental variables in this study. No suitable instrumental variables for FTD could be identified from genome-wide association studies dataset using a threshold of P < 5 × 10-8. Genetically predicted RBD was causally associated with an increased risk of PD (IVW: odds ratios = 1.093, 95% confidence intervals = [1.031-1.159], P = .003). In the reverse MR analysis, genetically predicted PD was found to be causally increased the incidence of RBD (IVW: odds ratios = 1.758, 95% confidence intervals = [1.355-2.281], P = 2.176 × 10-5). No heterogeneity or pleiotropy was observed in Cochran Q test, MR-Egger intercept test, or MR-PRESSSO Global test in the determination of the above 2 MR analyses. However, this bidirectional MR study did not identify any causal relationship between RBD and AD, ALS, DLB, FTD, and MSA. This MR study supported a bidirectional causal relationship between RBD and PD, mutually increasing the incidence of each. However, current genetic evidence did not support causal associations between RBD and AD, ALS, DLB, FTD, and MSA in either direction. However, these null findings should be interpreted with caution due to the limited sample sizes of the genome-wide association studies summary data used, highlighting the need for larger genetic studies to investigate these relationships further.