Objective To investigate the potential causal relationships between coronavirus disease-2019(COVID-19)susceptibility,hospitalization and severe case with late-onset myasthenia gravis(LOMG)based on Mendelian randomization(MR).Methods The public data from non-overlapping genome-wide association studies were screened,COVID-19 susceptibility,hospitalization and severe case served as the exposure data,and LOMG as the outcome data.The inverse-variance weighted(IVW)method was mainly adopted to evalu-ate the causal effect,which was supplemented by the methods such as MR-Egger method,weighted median method,weighted model and simple model.The sensitivity analysis was performed.Results The genetically predicted severe case of COVID-19 had the positively causal relationship with LOMG(OR=1.01,95％CI:1.00-1.03,P=0.046).The sensitivity analysis results revealed the study results were steady(P＞0.05).No heterogeneity or horizontal pleiotropy was found.Conclusion Severe case of COVID-19 may be associated with an increased risk of LOMG.