The current study assessed the clinical burden of MG in a representative population of over 1000 patients in England over two decades using linked data from the CPRD and HES. The study showed that patients diagnosed with MG often experience severe MG-related events, with 39% being hospitalized at least once for MG, 25% experiencing at least one exacerbation, and 18% experiencing at least one myasthenic crisis. Many of these events occurred within the first 2–3 years after MG was diagnosed, indicating that in most cases, the disease is ultimately controlled by treatment, spontaneously subsides, or both. We also found that for several years after diagnosis—even beyond the first 2–3 years—patients who were refractory to conventional treatment continued to experience more exacerbations and MG-related hospitalizations than patients with non-refractory disease. More frequent comorbidities, some of which may have been due to treatments, especially long-term corticosteroid use, further added to the burden of refractory MG.
Few other longitudinal studies have examined how the burden of MG changes over time, and all were completed several decades ago [3, 12, 13]. In addition, sample sizes were modest in most of these studies, except for a study by Grob et al., which included 1976 patients with MG in the US over the six decades from 1940 to 2000 [3]. Two recent longitudinal studies of claims data, one in the US [15] and the other in Japan [16], focused on the burden of illness in patients with refractory and non-refractory MG, although analyses were limited to the first year after diagnosis. In line with the current study and our previous analysis of healthcare resource utilization in this same population [19], the US and Japanese claims studies showed more frequent exacerbations, hospitalizations, and other healthcare resource utilization in patients with refractory vs non-refractory MG. Unlike the US and Japanese claims studies, however, the current study did not find a difference in the proportion of patients experiencing myasthenic crises. This might be related to differences in definitions of myasthenic crisis or refractory status, although insufficient numbers of patients with refractory disease may have precluded making inferences. Also, in the current study, patients did not meet the definition of refractory MG if they died within the first 2 years after diagnosis, which is when most myasthenic crises occur.
Comorbidities contribute to the burden of MG, lengthen hospital stays, and increase the risk of death. In patients with MG, reported comorbidities include cardiovascular disease, hyperlipidemia, hypertension, diabetes mellitus, respiratory disorders, and concomitant autoimmune diseases [7, 8]. The current study provides data on comorbidities before and after MG was diagnosed. The most common comorbidities seen during the baseline and follow-up periods were renal disease, chronic pulmonary disease, diabetes, malignancies, hypertension, myocardial infarction, and congestive heart failure. These are all common in older age, and as expected in this population, which was mostly >60 years of age. However, renal disease, hypertension, psoriasis, and psoriatic arthritis were more common in patients with refractory than non-refractory MG. Diabetes, renal disease, hypertension, congestive heart failure, psoriasis, and psoriatic arthritis were more frequent in patients with refractory MG than in age- and sex-matched non-MG controls. The study of US claims data also reported more frequent diabetes, cardiac arrhythmias, and severe infections in patients with refractory than non-refractory MG [15]. Some of the comorbidities associated with refractory MG are likely adverse effects of long-term treatment with systemic corticosteroids and other immunosuppressive therapies [10, 25]. Also, increased rates of psoriasis and psoriatic arthritis in patients with refractory MG are consistent with more frequent concomitant autoimmune conditions in these patients [7].
The current study assessed the burden of MG in patients receiving conventional therapies (e.g. corticosteroids, immunosuppressive therapies, intravenous immunoglobulins, plasmapheresis) in 1997–2016. Since then, a humanized monoclonal antibody (eculizumab) that inhibits terminal complement activation has been approved for treatment of refractory gMG, and other targeted therapies are being developed.
One limitation of this study is that it was performed using healthcare data from the English primary care setting and therefore might not be generalizable to other countries. Nonetheless, this study provides extensive and long-term longitudinal data about periods before and after a diagnosis of MG was recorded. The more than 1000 patients with MG identified in this study represent one of the larger sets described to date, although low numbers of patients with refractory MG made it difficult to make inferences in some cases, such as for myasthenic crises and certain comorbidities.
Another potential limitation of this study is that the accuracy of the results depended on the completeness and precision of encoding by general practices for the CPRD and by hospitals for the HES. Further, the definition of refractory MG was adapted from existing guidelines [1, 26–28] for the databases used and, as such, depended mostly on records of treatments prescribed by general practitioners. Treatments administered in hospitals or by specialists are not recorded in CPRD or HES, which may have reduced the number of patients who should have been considered refractory. Additionally, because the reasons for procedures (e.g. intravenous immunoglobulin administration) cannot be determined from the HES, the definitions of exacerbations and crises may have resulted in a slight overestimation of their occurrence. Including clinical criteria may have improved the accuracy of detecting refractory cases, but such information was not available in the databases used for this study.