强直性脊柱炎与听力损失之间的关联：一项孟德尔随机化研究

文献信息

中文摘要

观察性研究提示强直性脊柱炎（AS）与听力损失之间可能存在关联。然而，由于治疗可能引起药物性耳毒性，这种关联仍存在争议。因此，我们采用孟德尔随机化（MR）方法，从遗传学角度评估AS与听力损失之间的关联。本MR研究使用了关于AS和听力损失的全基因组关联研究汇总数据，包括感音神经性听力损失（SNHL）和传导性听力损失（CHL）。主要分析采用逆方差加权法进行。主要分析显示AS对SNHL（比值比[OR]=1.121，95%置信区间[CI]：1.042-1.206，P=.002）和CHL（OR=1.387，95% CI：1.123-1.712，P=.002）均存在因果效应。排除6号染色体变异后，与CHL的关联仍然存在（OR=1.675，95% CI：1.212-2.315，P=.002），而对SNHL的效应不再显著（OR=1.054，95% CI：0.942-1.181，P=.357）。主要逆方差加权分析显示SNHL对AS无因果效应（OR=0.984，95% CI：0.892-1.086，P=.747）。敏感性分析未发现水平多效性的证据。本MR研究提示AS与CHL和SNHL风险升高存在因果关联。与CHL的关联似乎独立于6号染色体人类白细胞抗原区域，而SNHL的风险在很大程度上由该区域驱动。这些发现提示，对于有症状的AS患者应考虑进行针对性的听力学评估。需要进一步研究以探索AS与不同类型听力损失之间关联的具体机制。

英文摘要

Observational studies have suggested a potential link between ankylosing spondylitis (AS) and hearing loss. However, this association remains controversial due to potential drug-induced ototoxicity from treatments. Therefore, we employed a Mendelian randomization (MR) approach to evaluate the association between AS and hearing loss from a genetic perspective. This MR study utilized summary data from genome-wide association studies on AS and hearing loss, including sensorineural hearing loss (SNHL) and conductive hearing loss (CHL). The primary analysis was conducted using the inverse-variance weighted method. The primary analysis showed a causal effect of AS on both SNHL (odds ratio [OR] = 1.121, 95% confidence interval [CI]: 1.042-1.206, P = .002) and CHL (OR = 1.387, 95% CI: 1.123-1.712, P = .002). After excluding chromosome 6 variants, the association with CHL persisted (OR = 1.675, 95% CI: 1.212-2.315, P = .002), while the effect on SNHL was no longer significant (OR = 1.054, 95% CI: 0.942-1.181, P = .357). The primary inverse-variance weighted analysis showed no causal effect of SNHL on AS (OR = 0.984, 95% CI: 0.892-1.086, P = .747). Sensitivity analyses found no evidence of horizontal pleiotropy. This MR study suggests that AS is causally associated with higher risks of CHL and SNHL. The association with CHL appears independent of the chromosome 6 human leukocyte antigen region, whereas the risk for SNHL is largely driven by this region. These findings suggest that targeted audiometric evaluations should be considered for symptomatic patients with AS. Further studies are required to explore the specific mechanisms linking AS to different types of hearing loss.