Effects of Testosterone in Mediating the Relationship Between Daytime Napping and Osteoporosis in European Populations: A Mendelian Randomization Study

We aimed to explore the causal effect of daytime napping on the risk of osteoporosis and the mediation role of testosterone in explaining this relationship. Summary data for Mendelian randomization (MR) analysis were obtained from the IEU OpenGWAS database. Univariable MR(UVMR) analysis and multiple sensitivity analyses were applied to explore the casual relationship between daytime napping and bone mineral density (BMD)/osteoporosis. We also conducted multivariable Mendelian randomization (MVMR) analysis to evaluate the correlation between testosterone-associated single-nucleotide variations and BMD/osteoporosis. Then, mediation analysis was performed to explore whether the association between daytime napping and BMD/osteoporosis was mediated via testosterone. Genetically predicted daytime napping was significantly associated with femoral neck BMD (β [95% CI]: 0.2573 [0.0487, 0.4660]; P = 0.0156), lumbar spine BMD (β [95% CI]: 0.2526 [0.0211, 0.4840]; P = 0.0324), and osteoporosis (OR [95% CI]: 0.5063 [0.2578, 0.9942]; P = 0.0481). β and 95%CIs indicate the standard deviation (SD) unit of BMD increase per category increase in daytime napping. OR and 95%CIs represent the change in the odds ratio of osteoporosis per category increase in daytime napping. We observed a potentially causal effect of more frequent daytime napping on higher BMD and a lower risk of osteoporosis. Daytime napping was causally associated with a higher level of bioavailable testosterone (β [95% CI]: 0.1397 [0.0619, 0.2175]; P = 0.0004). β and 95%CIs represent the change in the SD of testosterone per category increase in daytime napping. Furthermore, the causal effects of daytime napping on BMD/osteoporosis were partly mediated by bioavailable testosterone. Daytime napping can efficiently increase BMD and reduce the risk of osteoporosis, and testosterone plays a key mediating role in this process.