Analysis of mortality and related factors in 2195 adult myasthenia gravis patients in a 10-year follow-up study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/neuroindia.NI_804_16
Source of Support: None, Conflict of Interest: None
Objective: To analyze the mortality and potential risk factors for death in myasthenia gravis (MG) patients.
Keywords: Age of onset, disease duration, myasthenia gravis, oral immunosuppressants, thymectomy
Myasthenia gravis (MG) is an autoimmune neuromuscular disease, clinically characterized by fluctuating weakness and fatigability of affected muscles, in which different kinds of autoantibodies play an important role. The advent of immunomodulating therapy, including administration of prednisone, immunosuppressants, intravenous immunoglobulin (IVIg), plasmapheresis, and/or performace of thymectomy has significantly improved the outcome of the disease in the past decades, especially resulting in a dramatic decline in mortality. However, death related to MG still remains a major concern. In order to depict the mortality rate, as well as to study the factors that might potentially contribute to death in MG, we have retrospectively reviewed the medical records of patients with adult-onset MG (AMG), who were treated and followed-up for 10 years at our hospital.
The MG patients at our hospital were followed up since 2003. The MG patients who initially developed the disease and were older than 18 years were included in this study. This included a cohort of 2195 AMG patients. As there was no mortality observed during the period of the study in childhood-onset MG (CMG) patients, the CMG patients were not included in the study. The documented data included the demographic spectrum of the participants, clinical types, disease duration, comorbidities, ancillary findings, treatment, and outcome. The period of evaluation of the patients was until December 2013. Death of patients occurring both at hospitals and as well as at the residence was regarded as the endpoint of this study. The causes of death were determined by the descriptions on death certificates issued by neurologists at hospitals or by doctors who had last seen the patient in the residential areas where the patients were residing.
The study was permitted by the Committee of Clinical Investigations of our hospital. The data collection, assessments, and follow-up of all the participants were performed after the informed consent was signed.
The diagnosis of MG was established based on the typical presentation of MG, and a dramatic response to intramuscular injection of a bolus of neostigmine sulfate. Repetitive nerve stimulation was used as a diagnostic tool for the diagnosis of the disease. Anti-acetylcholine receptor antibodies (AchRabs) were routinely determined using enzyme-linked immunosorbent assay (ELISA) [RSR Limited, Cardiff, UK]. The reliability of the ELISA method had been verified by simultaneously testing the AchRabs titers in 100 clinically diagnosed MG patients using radioimmunoassay (RIA) [RSR limited, Cardiff, UK]. Due to the hospital policy limiting the use of radioactive isotopes at Chinese hospitals, the ELISA method, rather than the radioimmunoassay method, was routinely used for the diagnosis of MG. The severity of the disease was evaluated according to the Myasthenia Gravis Association of America (MGFA) classification. Thymic abnormalities were checked by chest computed tomography (CT) or magnetic resonance imaging (MRI). The World Health Organization (WHO) 2004 thymic histopathologic classification was used to classify thymomas. All the patients were checked for comorbidities such as autoimmune thyroid disease.
An established therapeutic regimen was used to treat all MG patients in our Department of Neurology. At the beginning, oral pyridostigmine in a daily dosage of 180–240 mg was usually given. If the symptoms did not significantly improve or the patient continued to deteriorate despite administration of oral pyridostigmine for a few weeks to months, oral prednisone in a daily dosage of 1 mg/kg (totally less than 60 mg/d for individuals) was given, after the patients were suitably informed regarding the changes made in the therapeutic regimen. The prednisone dosage was tapered based on a fixed schedule after the symptoms significantly improved and it was eventually discontinued after the symptoms had completely resolved for at least 6 months. If the symptoms relapsed, oral prednisone was re-started. If both pyridostigmine and prednisone could not stop the progression of the disease or the sufferers had continued difficulty in activities of daily living, a trans-sternal extended thymectomy was an optional method adopted. In addition, the thymectomy was usually performed for a thymomatous MG case, who did not have any contraindications to undergo the procedure. Immunosuppressants such as azathioprine in a daily dosage of 150 mg, or FK506 in a daily dose of 3 mg (maintaining its serum concentration in the range of 5–10 μg/ml), were usually administered for patients with a generalized form of MG (GMG) who did not show a satisfactory response to oral prednisone or who developed severe side-effects, such as uncontrolled hypertension. Intravenous administration of immuno-gammaglobulin (IVIg) in a dose of 0.4 g/kg/d for 5 consecutive days, or plasma-exchange was administered in patients having a severely progressive form of GMG, in patients who had a myasthenic crisis, or in patients with the generalized type of MG when deemed necessary during the perioperative period. The comorbidities such as hypertension or diabetes were routinely treated and monitored.
According to our protocol, all patients were usually asked to visit the neurologist for evaluation and adjustment of the therapeutic regimen at least every 2 months. If the symptoms completely disappeared for at least 6 months, the visit intervals were extended to every 6 months or longer. Interviews by telephones and online communication were usually used if the patients had been symptom-free for years.
The MGFA post-intervention status was applied to measure the outcome. The history of any major diseases such as heart attacks, diabetes mellitus, tuberculosis, hepatitis B, chronic obstructive pulmonary disease (COPD), and hypertension, as well as the intake of medications required for treating these diseases were documented at follow-up visits or during the telephonic assessments.
The Statistical Package for the Social Sciences (version 19) was used for statistical analysis. Multivariate analyses were performed using Cox proportional hazards model. We created a multivariable model that was adjusted for gender. Survival analysis was performed using Kaplan–Meier's method to estimate the survival function. Differences were considered as significant when P values were <0.05.
The demographic features of the 2195 AMG patients (1259 females and 936 males) are described in [Table 1]. The mean age at onset of the disease was 36.06 ± 14.48 years, and the mean age at entry into the study was 41.16 ± 13.70 years. The mean follow-up period was 111.16 ± 39.59 months. The clinical types of MG present in the patients were the type 1 form of the disease in 487 (22.19%) patients; and, the generalized type in 1708 (77.81%) patients. The AChRab titers were elevated in 70.48% of all the AMG patients. Thymoma was found in 211 patients (9.61%), and hyperplasia was observed in 115 cases (5.24%). In the cohort, 2166 patients (98.68%) had received oral prednisone therapy, 761 cases (34.67%) were given immunosuppressants, and thymectomy was performed in 583 patients (26.56%).
During the period of observation, 129 out of 2195 (5.88%) MG patients died [Table 2]. Death occurred after a mean interval of 61.47 ± 53.52 months after the onset of the disease. The causes of death included cancers (n = 41, 31.78%), MG-related factors (n = 36, 27.91%), heart attacks (n = 21, 16.28%), pulmonary diseases (14, 10.85%), stroke (n = 13, 10.07%), gastrointestinal bleeding (3, 2.33%), and other causes (n = 1, 0.78%) [Table 2]. Out of the dead patients, 85 (65.9%) died at hospitals and 44 (34.1%) died in residential communities.
The risk factors potentially associated with MG-related death have been analyzed [Table 3]. In general, the disease course, occurrence of myasthenic crisis, the severity of disease (MGFA types III and IV) at onset, elevation of AchR-abs titers, thymectomy, as well as the application of immunosuppressants were the key factors which significantly influenced the mortality (P < 0.05). However, the age at entry and clinical types IIA and IIB as well as the application of prednisone did not show a strong association with mortality in patients. When we analyzed the association of the patients' age with mortality, the range of 18–50 years of patients at entry was treated as the referent [Table 3].
Using the same protocol, a disease duration shorter than 5 years and the stage of MGFA type I at onset were considered as referents. Interestingly, these factors showed differing associations with death between male and female patients [Table 3]. The duration of the disease for 5–10 years was considered to be a greater risk when compared with the duration of less than or equal to 5 years; however, the duration of the disease for longer than 15 years was considered to be a protective factor in all the patients. There was a difference between the male and female patients regarding the duration of the disease. Overall, the duration of the disease lasting for longer than 10 years was a protective factor; the duration of the disease for 5–10 years was considered as being of a lower risk relative to the duration of the disease of shorter than 5 years in the female patients; whereas, in male patients, the duration longer than 5 years was considered as being riskier [Table 3]. Unexpectedly, elevation of AchR-abs titers had a higher risk of mortality in the female patients but not in the male patients. Administration of immunosuppressants was a protective factor in the male but not in female patients.
Among the non-MG factors, the preceding history of stroke, COPD, diabetes mellitus, atrial fibrillation, hyperlipidemia, myocardial infarction, and malignant tumors were found to be closely linked with death in the MG population (hazard ratios were 3.251, 4.173, 3.738, 3.886, 1.945, 2.177, and 14.7, respectively; P< 0.05), however, other factors including peripheral vascular disease, heart failure, hypertension, chronic renal disease, and smoking were not significantly related to death (P > 0.05).
The influence of the major factors closely associated with MG-related death, including gender, an intake of predisone, AChRab titers, and thymic abnormalities were further analyzed using the Kaplan–Meier's method [Figure 1]. Accordingly, the presence of a thymic abnormality and elevated titers of AchRabs were the main factors which were closely associated with MG survival (P < 0.05); however, the gender of the patient and an intake of prednisone were not factors influencing survival (P > 0.05).
The study disclosed that the mortality rate in the AMG population was 5.88% at a 10-year follow up. The important death-related factors were the severity of disease at entry, the presence of AchR-Abs, the occurrence of thymic pathology, and the disease course. In addition, some comorbidities, including the history of stroke, COPD, diabetes mellitus, atrial fibrillation, hyperlipidemia, myocardial infarction, and a malignant tumor significantly increased the risk of death in the MG population.
The disease course was a very important factor closely associated with the MG-related deaths. The majority of deaths occurred at 5–10 years after the onset of the disease, and rarely within 5 years; however, the risk of death tended to decrease after 15 years of the prevalence of the disease. As MG is an autoimmune disease with a fluctuating course, a proper medical intervention helps in significantly improving the symptomatology in the initial stages of the disease. Some cases did not respond well to the therapy. They might not have received careful medical care because of many reasons such as the psychological stress that they were undergoing, or due to their financial constraints, and were at a higher risk of developing neurological deterioration or even dying. Standardized long-term therapy, especially the use of prednisone in addition to azathioprine or FK506 in patients who experienced transit deterioration, was mandatory to reduce the risk for death during the observation period.
Anti-AchR titers were not previously reported to correlate with disease severity and prognosis in the MG population. However, our data showed that there was a close association between the risk of MG-related death and anti-AchR antibody titers (P < 0.001; HR = 2.432) in Chinese MG patients. Interestingly, elevation of anti-AchR-abs titers was found to be a risk factor only in the female patients but not in the male ones [Table 3]. Probably, the anti-AchR-abs is not a sole player in the regulation of the course of events; sex hormones may also play a role in the pathological process.,
If MG was severe at the onset (MGFA types III and IV) it constituted a grave risk in our patients. The association of severity of the disease with a higher risk of death could be explained by the frequent involvement of the bulbar and respiratory muscles in these individuals; the patients in MGFA type I have a better outcome because only the ocular muscles are affected in them.
There is a controversy regarding the impact of thymic pathology on the long-term outcome in patients with MG., Many reports did not conclude that the presence of a thymoma is closely associated with prognosis in MG patients. However, our data suggested that the probability of survival in thymomatous patients with MG was significantly less than that in the MG patients with hyperplasia [Figure 1]. The detailed reason needs more studies.
Myasthenic crisis is a significant risk factor responsible for death, and an emergency intervention is the key to the patients' survival. If the condition was not recognized and treated promptly, the patients were at an extremely high risk of dying. The causes of death in patients with myasthenic crisis were pulmonary infarction or cardiac arrhythmia.
Thymectomy is a very important component in the process of improving the symptom complex in MG. Some studies observed a clinical improvement in the patients with refractory MG after repetitive thymectomy. In our study, thymectomy was performed in 26.56% of MG patients, and was found to decrease the risk of death (P < 0.05, HR = 0.315). Optimum control of the myasthenic presentation before the thymectomy was the key to decreasing the occurrence of myasthenic crisis during the perioperative period. To our experience, the quantitative MG (QMG) score of less than 9 and the dosage of prednisone less than 30 mg per day were the minimum requirements for the performance of an extended trans-sternal thymectomy.
The usage of corticosteroids and immunosuppressants in the treatment of MG has greatly changed the outcome of MG patients; however, no large clinical trials have clearly demonstrated their efficacy. Importantly, oral administration of immunosuppressants was discovered to be closely associated with a decreased risk for death in our study (P < 0.05, HR = 0.447). The results of our study are consistent with the findings of other reports. The most commonly used agents in our patients included azathioprine (100–150 mg per day), FK506 (3 mg per day), or cyclosprorin (50–150 mg per day). When the potential side-effects of these agents were cautiously monitored, majority of MG patients could tolerate the drugs for a few years. No patient died of the severe side-effects of these agents in our patients. In our study, oral immunosuppressants had a protective role only in male patients but not in the female ones [Table 3]. The role of sex hormones in determining the effective difference between the outcome in males and female patients needs further evaluation.
Apart from the MG associated factors, the associated comorbidities, especially in the middle-aged and elderly MG patients, which may contribute to death in MG patients have been analyzed in our study [Table 3]. As expected, a previous history of stroke, diabetes mellitus, and myocardial infarction were the risk factors for an early death in patients suffering from MG. We are still not sure whether the usage of aspirin and statins in the patients suffering from MG with associated cardio-cerebrovascular diseases or/and diabetes mellitus may change the long-term outcome of MG. In reality, both doctors and patients have concerns regarding the use of corticosteroids in these patients because steroids may further cause disruptions in the blood pressure and glucose homeostasis in these patients.
The potential role of COPD in influencing MG-related death is not easy to determine. The acetylcholinesterase inhibitors may constrict the broncho-alveolar apparatus, which may cause patients with MG with associated COPD to deteriorate. Smoking further increases the vulnerability of the MG patients with associated COPD towards developing a lung infection, which may in turn trigger a relapse or may lead to an exacerbation of symptoms related to MG. It is, therefore, easy to understand that COPD plays an important part in influencing mortality in the MG patients [Table 3]. It is suggested that MG patients with associated CODP should be kept infection-free and should be counseled to avoid smoking. Early recognition of upper respiratory infection, prompt management of COPD in combination with respiratory infections, as well as taking precautions in increasing the dosage of pyridostigmine, constitute the basic protocol for these MG patients with associated COPD. On the other hand, the presence of peripheral vascular disease, heart failure, hypertension and chronic kidney disease were not found to be significantly related to death in the MG population.
Unfortunately, one of our patients committed suicide. MG is chronic, often fluctuating and occasionally a life-threatening disease; hence, the psychological status of the sufferers must be closely monitored. When necessary, medications and psychotherapy should be administrated to treat the emotional stress that the patient is undergoing.,
To conclude, the mortality of the MG population is very low following a systemic and protocol-based medical intervention for MG. However, many non-MG factors such as stroke, COPD, diabetes mellitus, atrial fibrillation, hyperlipidemia, myocardial infarction, malignant tumors, and the psychological status of the patients may influence the long-term outcome of MG patients.
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent was obtained from all individual participants included in the study.
Financial support and sponsorship
The study was supported by a grant of the National Science Foundation of China to Dr. Bitao Bu (No. 81271320).
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2], [Table 3]