[A cohort study and Mendelian randomization study on association between osteoporosis and Alzheimer's disease]

Objective: To analyze the association between osteoporosis and the risk for Alzheimer's disease (AD). Methods: A total of 471 922 study subjects were selected from the UK Biobank database, including 12 818 osteoporosis cases and 459 104 controls. Cox proportional hazard regression model and competing risk model were used to evaluate the association between osteoporosis and AD after adjusting confounding factors. Furthermore, a Mendelian randomization (MR) study was conducted by using the data of two published genome-wide association studies, and 1 050 highly relevant single nucleotide polymorphisms were identified from the bone mineral density data as instrumental variables. The association between bone mineral density and the risk for AD was evaluated by using inverse variance weighted method, MR-Egger regression, and weighted median estimator method. Additionally, sensitivity analyses were performed. Results: After adjusting for confounders, no significant association between osteoporosis and an increased risk for AD was found in the cohort study (Cox proportional hazard regression model analysis: HR=1.10, 95%CI: 0.78-1.56, P=0.588). The MR analysis revealed no association between bone mineral density and the risk for AD (inverse-variance weighted: OR=1.03, 95%CI: 0.98-1.09, P=0.252), and the results remained robust in multiple sensitivity analyses. Conclusion: The study result does not support the association between osteoporosis and risk for AD.