Hefei Second People's Hospital, Hefei, China.
The First Affiliated Hospital of China University of Science and Technology (Anhui Provincial Hospital), Hefei, China.
Medicine (Baltimore). 2024 May 31;103(22):e38223. doi: 10.1097/MD.0000000000038223.
Evidence from observational researches have suggested that mental diseases are able to affect thyroid diseases. However, the causal relationship between mental diseases and the risk of thyroid diseases still remains unclear. Herein, we conducted a two-sample Mendelian randomization (MR) statistical analysis method to assess the causality between mental diseases and thyroid diseases. Initially, publicly available genome-wide association studies summary data were leveraged to obtain single-nucleotide polymorphisms based on set parameters. Subsequently, a two-sample MR was utilized to analyze causal relationships between mental diseases (Alzheimer disease, bipolar disorder, major depressive disorder, Parkinson disease, schizophrenia) and thyroid diseases (hyperthyroidism/thyrotoxicosis, hypothyroidism) with removing outliers based on MR-PRESSO method. Finally, 8 regression MR methods (inverse variance weighted [IVW], IVW fixed effects, c, MR Egger, weighted median, penalized weighted median, simple mode, weighted mode) were performed to evaluate bias and effectiveness, of which IVW was considered as the primary method. Our results demonstrated that most of mental diseases have no causal relationships with thyroid diseases except bipolar disorder and hyperthyroidism/thyrotoxicosis based on IVW method [odds ratio: 0.999, 95% confidence interval: 0.998-1.000, P = .028], and bipolar disorder and hypothyroidism based on IVW method [odds ratio: 0.997, 95% confidence interval: 0.995-0.999, P = .002]. Then we subsequently conducted a consistent robustness analysis to assess heterogeneity and horizontal pleiotropy. Our method reports causal relationships exist mental diseases and the risk of thyroid diseases. Subsequent researches are still warranted to determine how mental diseases influence the development of thyroid diseases.
观察性研究的证据表明，精神疾病能够影响甲状腺疾病。然而，精神疾病与甲状腺疾病风险之间的因果关系仍不清楚。在此，我们采用了两样本 Mendelian randomization (MR) 统计分析方法来评估精神疾病与甲状腺疾病之间的因果关系。首先，利用公开的全基因组关联研究汇总数据，根据设定的参数获得基于单核苷酸多态性。然后，采用两样本 MR 分析精神疾病（阿尔茨海默病、双相情感障碍、重度抑郁症、帕金森病、精神分裂症）与甲状腺疾病（甲状腺功能亢进/甲状腺毒症、甲状腺功能减退）之间的因果关系，并基于 MR-PRESSO 方法去除异常值。最后，采用 8 种回归 MR 方法（逆方差加权 [IVW]、IVW 固定效应、c、MR Egger、加权中位数、惩罚加权中位数、简单模式、加权模式）评估偏差和有效性，其中 IVW 被视为主要方法。我们的结果表明，除双相情感障碍和甲状腺功能亢进/甲状腺毒症与精神疾病之间存在因果关系外，大多数精神疾病与甲状腺疾病之间没有因果关系（比值比：0.999，95%置信区间：0.998-1.000，P=0.028），双相情感障碍和甲状腺功能减退与 IVW 方法[比值比：0.997，95%置信区间：0.995-0.999，P=0.002]。然后，我们进行了一致性稳健性分析以评估异质性和水平多效性。我们的方法报告了精神疾病与甲状腺疾病风险之间存在因果关系。需要进一步的研究来确定精神疾病如何影响甲状腺疾病的发展。
