To elucidate the precise upstream and downstream regulatory mechanisms of inflammatory factors in osteoporosis (OP) progression and to establish a causal relationship between inflammatory factors and OP. We conducted bidirectional Mendelian randomization (MR) analyses using data for 41 cytokines obtained from three independent cohorts comprising 8293 Finnish individuals. Estimated bone mineral density (eBMD) data were derived from 426,824 UK Biobank White British individuals (55% female) and fracture data from 416,795 UK Biobank participants of European ancestry. The inverse variance-weighted method was the primary MR analysis approach. We employed other methods as complementary approaches for mutual corroboration. To test for pleiotropy and heterogeneity, we used the MR-Egger regression, MR-pleiotropy residual sum and outlier global test, and the Cochrane Q test. Macrophage inflammatory protein (MIP)-1α and interleukin (IL)-12p70 expression associated negatively and causally with eBMD (β = −0.017 [MIP-1α], β = −0.011 [IL-12p70]). Conversely, tumor necrosis factor-related apoptosis-inducing ligand was associated with a decreased risk of fractures (Odds Ratio: 0.980). Additionally, OP influenced the expression of multiple inflammatory factors, including growth-regulated oncogene-α, interferon-gamma, IL-6, beta nerve growth factor, and IL-2. Finally, we discovered complex bidirectional causal relationships between IL-8, IL-10, and OP. Specific inflammatory factors may contribute to OP development or may be causally affected by OP. We identified a bidirectional causal relationship between certain inflammatory factors and OP. These findings provide new perspectives for early prediction and targeted treatment of OP. Larger cohort studies are necessary in the future to further validate these findings.

阐明炎症因子在骨质疏松症（OP）进展中的精确上下游调控机制，并建立炎症因子与OP之间的因果关系。我们使用从 8293 名芬兰人组成的三个独立队列中获得的 41 种细胞因子的数据进行了双向孟德尔随机化 (MR) 分析。估计骨矿物质密度 (eBMD) 数据来自 426,824 名英国生物银行白人个体（55% 女性），骨折数据来自 416,795 名欧洲血统的英国生物银行参与者。逆方差加权方法是主要的 MR 分析方法。我们采用其他方法作为补充方法来相互印证。为了测试多效性和异质性，我们使用了 MR-Egger 回归、MR-多效性残差和和异常值全局检验以及 Cochrane Q 检验。巨噬细胞炎症蛋白 (MIP)-1α 和白细胞介素 (IL)-12p70 表达与 eBMD 负相关且因果相关（β = -0.017 [MIP-1α]，β = -0.011 [IL-12p70]）。相反，肿瘤坏死因子相关的凋亡诱导配体与骨折风险降低相关（优势比：0.980）。此外，OP影响多种炎症因子的表达，包括生长调节癌基因-α、干扰素-γ、IL-6、β神经生长因子和IL-2。最后，我们发现了 IL-8、IL-10 和 OP 之间复杂的双向因果关系。特定的炎症因子可能会导致 OP 的发生，或者可能受到 OP 的因果影响。我们发现某些炎症因子与 OP 之间存在双向因果关系。这些发现为OP的早期预测和靶向治疗提供了新的视角。未来需要进行更大规模的队列研究来进一步验证这些发现。