The present study is the first systematic review and meta-analysis that summarizes the results of all available cohort and case-control studies that reported the association between asthma and risk of developing MG. The pooled analysis found that patients with asthma had approximately 1.4-time higher odds of MG.
The underlying mechanism of the association between asthma and risk of MG is still unclarified, but there are some possible explanations. First, asthma and MG might share some common immunopathogenic pathway. Previous studies showed that MG patients had the overexpression of CD23 in germinal centers of the thymus and the increase serum level of sCD23. Whereas, the declining of serum level of sCD23 were observed after thymectomy which showed a strong correlation with clinical improvement [20, 21]. CD23, known as a low-affinity receptor for IgE (FcεRII), involved in regulation IgE synthesis, antigen presentation and B cell activation [22, 23]. IgE is believed to contribute to asthmatic manifestation and other allergic condition [24]. In addition, the dysregulation of CD4 T lymphocyte (Th1, Th2, Th17, Teg) is also considered to be involved in the pathogenesis of asthma and MG [4, 25].
Second, genetic predispositions for MG and asthma have been well documented, and with the strongest association to HLA genes. Based on the current evidence, several traits of HLA can be characterized in the background of MG by subtype specificity, gender discrepancy, ethic and geographical disparity. For example, HLA-B*08, HLA-DRB1 and HLA-DQB1 present virtually in certain MG subtypes universally [26, 27]. These genes are also shown to be associated with asthma based on genome-wide association studies [28–30]. In addition, TNF-α (gene related non-classic MHC molecules) also contribute to MG and asthma predisposition [26, 31, 32]
Lastly, asthma and MG may share some common environmental risk factors. Viral infections are important trigger of acute wheezing episodes in infancy and the inception and exacerbation of asthma [33]. They also have been suspected to play role in MG. Antiviral process can be observed in the myasthenia gravis thymus [34]. Molecular mimicry, cryptic antigens, epitope spreading, bystander activation, and polyclonal activation have all been suggested as mechanisms for the induction of MG by viral agents [35].
This meta-analysis carries some limitations that should be aware of. First, the statistical heterogeneity of the meta-analysis was moderate. Difference in study design and participant characteristics was probably the one of the main reasons for the variation. Second, nearly half of included studies have poor quality based on Newcastle-Ottawa scale. Third, the majority of the included studies relied on diagnosis codes from administrative databases to identify and diagnose asthma and MG [12, 13, 15, 16]. One study relied on self-reported questionnaires to diagnose asthma [14]. Therefore, the completeness of case identification, accuracy of the diagnoses of both diseases and MG subtype classification could be limited. Last, the small number of included studies in meta-analysis could jeopardize the validity and the interpretation of the funnel plot.