Zhao Xunying, He Lin, Wu Xueyao, Zhang Li, Xiao Jinyu, Xiao Changfeng, Qu Yang, Zhu Jingwei, Qin Chenjiarui, Huang Deqin, Shen Pengyue, Han Tao, Fan Mengyu, Li Jiayuan, Burgess Stephen, Jiang Xia
Department of Epidemiology and Health Statistics and West China Institute of Preventive and Medical Integration for Major Diseases, West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, Sichuan, China.
Longquanyi District of Chengdu Maternity and Child Care Health Hospital, Chengdu, Sichuan, China.
BMC Med. 2025 May 28;23(1):305. doi: 10.1186/s12916-025-04139-2.
While the protective role of body mass index (BMI) in bone mass has been well-documented, the divergent associations between BMI and estimated bone mineral density (eBMD), attributed to its highly heterogeneous nature, remain insufficiently understood.
Leveraging the hitherto largest genome-wide summary statistics, we conducted a two-sample Mendelian randomization (MR) to re-evaluate the effect of genetically predicted BMI on eBMD. Then, MR-Clust was applied to examine the potential presence of distinct causal pathways underlying the BMI-eBMD link. Utilizing tissue-partitioned MR, we estimated the distinct effects of separated tissue-specific subcomponents of BMI on eBMD, further supplemented by multivariable MR of body composition phenotypes on eBMD.
We reconfirmed the significant positive association between genetically predicted BMI and eBMD (β = 0.13, P value = 1.28 × 10). Potential distinct causal pathways contributing to the observed total effect were identified by MR-Clust, with some exerting a protective effect while others leading to its deterioration. Tissue-partitioned MR suggested a marginally independent protective association between skeletal muscle-tissue instrumented BMI and eBMD (β = 0.14, P value = 4.98 × 10) after accounting for adipose-tissue instrumented BMI, which was supported by the independent association between genetically predicted lean mass and eBMD after accounting for other body composition phenotypes.
Our results shed preliminary insights into the intricate relationship between obesity and bone mass, highlighting divergent causal pathways underlying the association between BMI and eBMD. Our findings emphasize the potential importance of precision obesity management over merely a general indicator as BMI in future public health strategies for osteoporosis prevention.
虽然体重指数（BMI）对骨量的保护作用已有充分记录，但由于BMI具有高度异质性，其与估算骨密度（eBMD）之间存在的不同关联仍未得到充分理解。
利用迄今为止最大规模的全基因组汇总统计数据，我们进行了两样本孟德尔随机化（MR）分析，以重新评估基因预测的BMI对eBMD的影响。然后，应用MR-Clust来检验BMI与eBMD之间潜在的不同因果途径。利用组织分区MR，我们估计了BMI分离的组织特异性亚成分对eBMD的不同影响，并通过身体成分表型对eBMD的多变量MR进行了进一步补充。
我们再次证实了基因预测的BMI与eBMD之间存在显著正相关（β = 0.13，P值 = 1.28 × 10）。MR-Clust确定了导致观察到的总效应的潜在不同因果途径，其中一些发挥保护作用，而另一些则导致其恶化。组织分区MR表明，在考虑脂肪组织测量的BMI后，骨骼肌组织测量的BMI与eBMD之间存在略微独立的保护关联（β = 0.14，P值 = 4.98 × 10），这在考虑其他身体成分表型后基因预测的瘦体重与eBMD之间的独立关联中得到了支持。
我们的结果初步揭示了肥胖与骨量之间的复杂关系，突出了BMI与eBMD之间关联的不同因果途径。我们的研究结果强调了精准肥胖管理在未来骨质疏松症预防公共卫生策略中的潜在重要性，而不仅仅是将BMI作为一个一般指标。
