Objectives: Although observational studies have suggested a link between hypothyroidism and myasthenia gravis (MG), a causal relationship has not been established. We aimed to investigate the causal association using a two-sample Mendelian randomization (MR) study.
Methods: Using summary statistics from genome-wide association studies involving 494,577 and 38,243 individuals, single-nucleotide polymorphisms exhibiting no linkage disequilibrium (r2 ≤ 0.001) and displaying significant differences (p ≤ 5 × 10-8) were selected for hypothyroidism and MG. To assess the potential causality relationship between hypothyroidism and MG, MR analysis was conducted using inverse variance weighted (IVW), weighted median method, and MR-Egger. The MR-Egger regression, heterogeneity test, pleiotropy test, and leave-one-out sensitivity test were employed to examine sensitivity analyses. In addition, validation datasets were used to validate the relevant results.
Results: Genetic liability to hypothyroidism was positively associated with MG (IVW, OR: 1.36, 95% CI: 1.17-1.58, p = 7.53 × 10-05; weighted median, OR: 1.19, 95% CI: 0.70-2.02, p = 0.522; MR-Egger, OR: 1.19, 95% CI: 0.98-1.45, p = 0.080). Among the three MR methods, the correlation between hypothyroidism and MG genetic prediction was consistent. The independent validation set (IVW, OR: 466.47, 95% CI: 4.70 -46,285.95, p = 0.01) further supported this. Additionally, bidirectional studies showed that using IVW, there was no reverse causality (OR: 1.104, 95%CI: 0.96-1.27, p = 0.170).
Discussion: This MR study showed that hypothyroidism can increase the risk of MG. Further investigation into the underlying mechanisms of this potential causality is warranted to offer novel therapeutic options for MG in the future.
Keywords: Mendelian randomization; causality; genome-wide association study; hypothyroidism; myasthenia gravis.