Objective

To disentangle whether sleep disturbances have a causal effect on the risk of osteoarthritis (OA) using genetically based approaches.

Method

We performed univariable and multivariable Mendelian randomization (MR) analyses using publicly released genome-wide association studies summary statistics to estimate the causal associations of sleep disturbances with OA risk. The inverse-variance weighted (IVW) method was utilized as primary MR analysis, whereas complementary methods including weighted median, weighted mode, MR-Egger regression, and MR pleiotropy residual sum and outlier (MR-PRESSO) were applied to detect and correct for the presence of pleiotropy.

Results

There were 228 independent instrumental variables (IVs) for insomnia and 78, 27 and 8 IVs for sleep duration, short sleep duration and long sleep duration, respectively. Univariable MR analysis suggested that genetically determined insomnia or short sleep duration exerted a causal effect on overall OA in an unfavorable manner (Insomnia: OR = 1.22, 95%CI = 1.15–1.30, P = 8.05 × 10⁻¹⁰; Short sleep duration: OR = 1.04, 95%CI = 1.02–1.07, P = 2.20 × 10⁻³). More compelling, increasing genetic liability to insomnia or short sleep duration was also associated with OA risk, after accounting for effects of insomnia or short sleep duration on body mass index, type 2 diabetes and depression individually, and in a combined model considering all three confounders.

Conclusions

Findings suggested consisted evidence for an adverse effect of increased insomnia or short sleep duration on OA risk. Strategies to mitigate sleep disturbances may be one of the cornerstones protects against OA.

中文翻译：

---

睡眠障碍对骨关节炎风险因果影响的证据：单变量和多变量孟德尔随机化研究

客观的

使用基于基因的方法来解开睡眠障碍是否对骨关节炎 (OA) 的风险有因果影响。

方法

我们使用公开发布的全基因组关联研究汇总统计数据进行了单变量和多变量孟德尔随机化 (MR) 分析，以估计睡眠障碍与 OA 风险的因果关系。使用逆方差加权 (IVW) 方法作为主要 MR 分析，而补充方法包括加权中位数、加权模式、MR-Egger 回归和 MR 多效性残差和异常值 (MR-PRESSO) 用于检测和校正多效性的存在。

结果

失眠有 228 个独立工具变量 (IV)，睡眠时间、短睡眠时间和长睡眠时间分别有 78、27 和 8 个 IV。单变量 MR 分析表明，基因决定的失眠或睡眠时间短对整体 OA 产生不利影响（失眠：OR = 1.22, 95%CI = 1.15–1.30, P  = 8.05 × 10 ^-10；睡眠时间短： OR = 1.04，95%CI = 1.02–1.07，P  = 2.20 × 10 ^-3）。更引人注目的是，在单独考虑失眠或睡眠时间短对体重指数、2 型糖尿病和抑郁症的影响后，以及在考虑所有三个混杂因素的综合模型中，增加对失眠或睡眠时间短的遗传倾向也与 OA 风险相关.

结论

研究结果表明，失眠增加或睡眠时间短对 OA 风险有不利影响。减轻睡眠障碍的策略可能是防止 OA 的基石之一。