Study on causal association of depression and multiple types of osteoarthritis based on two-sample bidirectional Mendelian randomization
Objective To study the causal association between depression with knee,hip,hand,spine and early osteoarthritis by using the Mendelian randomization(MR).Methods The public genome-wide asso-ciation study(GWAS)database was used to select the data of depression and OA(knee,hip,hand,spine and early stage),which had the precise diagnostic standards.When the exposure was set as depression and the out-come was multiple OA,it was positive analysis,and vice versa.The inverse variance weighted(IVW)served as the main outcome indicator,which was combined with the simple weighting model,MR-Egger intercept method,weighted median and weighted mode methods for conducting the results judgment.The analysis re-sults conducted the pleiotropy,heterogeneity,outliers and sensitivity tests to judge the robustness of the re-sults.Results The forward direction analysis:depression showed the causal association with knee OA(IVW:P=0.010,OR=1.65,95％CI:1.13-2.41),spine OA(IVW:P＜0.001,OR=2.92,95％CI:1.70-5.01)and early OA(IVW:P=0.002,OR=5.23,95％CI:1.83-14.90),no causal association was found between depression with hip OA and hand OA.The reverse direction analysis:the causal association between knee OA,hip OA and hand OA with depression was undefined.Due to the small number of instrumental variables(IVs)extracted from the spine and early OA at the time of exposure,the causal association between the two and depression was not investigated from the reverse direction analysis.The above analyses in the tests of plei-otropy,heterogeneity,outliers and sensitivity did not found the obvious abnormality,therefore the results were reliable.Conclusion There are significantly unidirectional positive causal association between depression with knee O A,spine OA and early OA.
万方数据
维普
