Long-term outcome of 424 childhood-onset myasthenia gravis patients
The objective of this study was to describe the clinical characteristics, outcome and factors that may affect the outcome of childhood-onset myasthenia gravis (CMG) patients in China. We have followed up 424 patients with CMG for at least 5 years at Tongji Hospital. At the end of follow-up, the outcome of all the patients was measured according to MGFA Post-intervention Status. In this study, the patients have been followed up for 9.8 ± 5.4 years. The mean onset age was 5.4 ± 3.6 years. Ocular myasthenia gravis (OMG) was the major type of CMG within 2 years after onset (95 %). Thymic hyperplasia was found in 116 patients, and thymoma was confirmed in 6 patients. Acetylcholine receptor antibodies were elevated in 69.5 % of the patients. All the patients were routinely treated. Thymectomy was performed in 34 patients (8.0 %). At the end of follow-up, seventy-one patients (16.7 %) were significantly improved, 66 patients (15.6 %) remained unchanged, 53 patients (12.5 %) were worsened, and 234 patients (55.2 %) were exacerbated. Importantly, fifty OMG patients (12.4 %) had transformed into generalized myasthenia gravis (GMG) over 2 years after onset. Thymectomy did not effectively reduce the transformation from OMG to GMG. However, GMG cases significantly benefited from the surgery. This study indicated that the cases with autoimmune CMG account for over 50 % in Chinese MG population. The long-term follow-up discloses that CMG patients have a low percentage of improvement, and a high percentage of worsening and exacerbation. The treatment should not be withdrawn too early after the patients obtain complete stable remission. More studies are needed to gain better control of CMG symptoms.
KeywordsChildhood-onset myasthenia gravis Ocular myasthenia gravis Transformed generalized myasthenia gravis Thymus Long-term outcome
Myasthenia gravis (MG) is an organ-specific autoimmune disease characterized by varying degrees of fluctuating muscle weakness and fatigability in which autoantibodies against nicotinic acetylcholine receptor (AChR) on endplates play an important role. In the Western countries, the disease mainly affects adults, and just 10–15 % of all the MG cases were childhood-onset MG (CMG) in North America . However, in China, CMG cases were estimated to account for more than half of the MG population . Both the childhood-onset and adult-onset MG (AMG) share common aspects in clinical presentation and immunological profiles, but differences in onset age distribution, clinical types and the outcome between the two different groups have rarely been reported [3, 4, 5].
Previous studies suggested that about half of AMG patients who initially present with ocular muscle involvement may converse into generalized muscle weakness within 2 years after onset [6, 7, 8]. However, as far as we knew, there have been few reports about the long-term outcome of CMG cases, especially the transformation from ocular type to the generalized forms of the disease. Here, we retrospectively reviewed the clinical characteristics and outcome of a cohort of 424 CMG patients who have been followed up for at least 5 years at Tongji Hospital, to see the long-term outcome and the potential factors which may influence the outcome.
Methods and patients
The study has been approved by the Committee of Clinical Investigation at Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (No. 20140903) and conformed to the principles of the Declaration of Helsinki.
All patients were diagnosed to have MG before the age of 14 years and have followed up since Jan, 1987 to Dec 2013 at the Department of Neurology, Tongji Hospital. The informed consents have been signed by the patient’s parents or custodies for the permission of clinical evaluation and follow-up prior to their inclusion in this study. The inclusion criteria of the study were all the MG cases who developed MG symptoms before age of 14 years and had been followed up for a minimum of 5 years (ranged from 5 to 45 years) at the Hospital, the cases who presented as a transitory form of MG in neonates due to the passive transfer of acetylcholine receptor antibodies (AChR-abs) from their mothers who had MG were excluded.
The MG diagnosis was made based on the typical clinical presentation characterized by fluctuating weakness and fatigability of the affected muscle weakness, and the dramatic response to intramuscular injection of a bolus of neostigmine sulfate. In addition, repetitive nerve stimulation (RNS) was only performed in children older than 8 years. Thymic abnormalities have been checked by computed tomography (CT), magnetic resonance imaging (MRI) or chest X-ray. AchR-abs were determined using ELISA (RSR Limited, Cardiff, UK). In addition, thyroid function, anti-thyroid antibodies and connective tissue diseases associated autoantibodies were determined as well in all the cases. In the cases who presented as atypical manifestations, screening of mutations for Kearns–Sayre syndrome (KSS) and congenital myasthenic syndromes was performed and MRI of extra-ocular muscles was routinely carried out to exclude the possibility of atypical Graves’s disease [9, 10, 11].
The severity of the disease at onset is classified as type I, IIa, IIb, III or IV according to the modified Osserman classification .
The clinical paradigms were documented for each child, including demographic data, age of onset, duration, types, remission period, relapses, triggers of relapses or worsening of the symptoms, associated diseases, and the responses to therapy.
A therapeutic regimen is usually used to treat all the children patients at Department of Neurology, Tongji Hospital. Initially, oral pyridostigmine is usually employed for all MG cases to improve the symptoms. If the symptoms are not improved significantly or become deteriorated after administration of oral pyridostigmine for a few months, oral prednisone at a daily dosage of 0.75 mg/kg will be started after informing the patients’ custodies. The prednisone will be tapered on schedule after the symptoms are significantly improved and eventually discontinued after the symptoms have totally disappeared for at least 6 months. If the symptoms relapse, oral prednisone will be re-started. If both pyridostigmine and prednisone does not stop the progression of the disease or/and the symptoms severely affect the daily life of the sufferers, thymectomy will be an optional method based on the aspect of thymic abnormalities on chest imaging and AChR-abs titers. Intravenous immunoglobulin (IVIG) or plasma exchange will be indicated for severely progressive cases with generalized forms of MG. Immunosuppressants are not usually administered for CMG cases that do not show satisfactory responses to both pyridostigmine and prednisone because the Chinese custodies are very reluctant to take these agents.
According to our protocol, all patients were usually asked to visit the neurologists for evaluation and adjustment of the therapeutic regimen at least every 2 months. If patients had been symptom free for at least 6 months, the visit intervals would be 6 months or longer. Telephonic interviews or online communications were usually used if the patients were asymptomatic for years.
At the end of the follow-up, the current clinical status was evaluated as complete stable remission (CSR), pharmacologic remission (PR) and minimal manifestation (MM). The change of clinical status was determined as improved, unchanged, worsened, exacerbated and death of MG according to MGFA post-intervention status .
To analyze the factors which may potentially affect the long-term outcome, the patients have been divided into two subgroups according to transformation of the ocular type into the generalized types 2 years after onset of the disease. The ocular myasthenia gravis (OMG) group included the patients who still presented as isolated ocular symptoms throughout the follow-up period, and the transformed form of MG (TMG) group consisted of the ocular CMG cases that developed generalized symptoms 2 years after onset of MG.
All AMG cases were treated as the same protocol, except thymectomy was encouraged in the cases with thymoma or with generalized myasthenia gravis (GMG), and immunosuppressants such as azathioprine or FK606 were more commonly prescribed for the cases that did not show satisfactory responses to pyridostigmine and prednisone.
For comparison study, the t test and the χ2 test were used. Differences are considered significant when two-sided p values are less than 0.05. Bivariate correlation analysis was used to analyze the correlation of stratified intervals between time of diagnosis and time of thymectomy to the clinical outcome in the thymectomized patients. All statistical analyses were done with the SPSS 19.0 (SPSS Inc. Chicago, IL, USA).
The general aspects and outcome of 424 CMG patients
The clinical characteristics of 424 CMG patients
Age at onset
5.4 ± 3.6 (1–14)
11.3 ± 6.2
Follow-up duration (years)
9.8 ± 5.4
Neostigmine test (+)
Based on chest MRI or CT, thymic hyperplasia was radiologically suggested in 116 cases, and thymomas were highly suspected in 23 patients. At last, thymectomy was carried out in 34 patients. Thymomas were pathologically confirmed in only 6 cases out of the 34 thymectomized patients. AChR-abs determination revealed that 69.5 % of the detected cases (89/128) were positive. All the patients had received pyridostigmine and prednisone therapy, and five patients had been treated with azathioprine. IVIG was used in two cases that experienced severely generalized weakness triggered by respiratory infection.
The outcome of 424 patients
Total (n = 424)
Improved (n = 71)
Unchanged (n = 66)
Worse (n = 53)
Exacerbation (n = 234)
Onset age (years old)
5.4 ± 3.6
4.7 ± 3.7
5.3 ± 3.5
6.6 ± 4.4
5.4 ± 3.4
11.3 ± 6.2
8.8 ± 3.9
11.4 ± 7.8
14.5 ± 7.6
11.4 ± 5.6
The comparison between OMG and TMG
Transformation from OMG to GMG was an important endpoint at the end of the follow-up. In addition to 21 cases that developed generalized symptoms of MG within 2 years after onset, 50 out of the 403 OMG cases (12.4 %) had transformed into GMG 2 years after the onset, and the remaining 353 cases still manifested as the isolated ocular symptoms. In light of the intervals that the transformation happened from the disease onset, transformation happened in 11 patients (22.0 %) 2–5 years, in accumulated 32 cases (64.0 %) 2–10 years, and in 44 cases (88.0 %) 2–20 years after the onset. Longer follow-up period revealed that the transformation occurred in 6 cases over 20 years after the onset. The cohort of CMG patients showed that the transformation occurred about 11.0 years after onset (ranged from 2 to 35 years).
The comparison of TMG and OMG groups
TMG (n = 50)
OMG (n = 353)
Age of onset (years old)
6.6 ± 4.3
5.2 ± 3.5
14.3 ± 7.6
11.0 ± 5.9
Other autoimmune diseases
Transformed time (years)
11.0 ± 7.1
Preceding events which may trigger relapses were reported in over 50 % of the CMG cases. Out of them, discontinuation of drugs (the intervals between the withdrawal and relapses of symptoms were ranged from 1 month to 21 years) was the predominant one (154 cases). Other preceding events were respiratory tract infection (24 cases), vaccination (7 cases), surgery (6 cases), pregnancy or delivery (5 cases), administration of other drugs (5 cases), alcohol intake (3 cases), trauma (3 cases), emotional stress (3 cases) and other diseases (3 cases).
Thymectomy did not show definite efficacy in CMG patients with OMG
The outcome of 34 thymectomized CMG patients
Age at onset (years old)
6.5 ± 4.3
Classification before thymectomy
Age at thymectomy (years old)
15.7 ± 8.5
Course at thymectomy (years)
9.1 ± 7.8
Classification after thymectomy
The aspects of CMG patients differ from those of AMG cases
Comparison of CMG and AMG patients
CMG (n = 424)
AMG (n = 220)
Onset age (years old)
5.4 ± 3.6
32.2 ± 12.5
Disease course (years)
11.3 ± 6.2
9.2 ± 5.8
Follow-up duration (years)
9.8 ± 5.4
9.6 ± 5.2
Presentation at onset
Pyri + pred
Rate of transformation
The comparison of major series of CMG reports
R. Melbourne Chambers 
Age of onset (mean, years old)
Follow-up length (mean, years)
Time of transformation (mean, years)
Transformation to a generalized form from OMG has not been as clearly delineated in CMG patients as in AMG cases . Interestingly, the rate of transformation (12.4 %) in the present study was significantly lower than the rate (32.5 %) in the 220 AMG patients who have been followed up in the same way at the Hospital (Table 5), and by far lower than the expected rate (up to 80 %) in AMG population . This difference between the CMG and AMG highlights the idea that CMG may constitute a distinctive clinical entity different from AMG patients in some aspects such as the rare presence of thymoma, high percentage of OMG and a tendency to relapse in CMG patients. However, the auto-immunological characteristics of CMG patients were similar to those of AMG sufferers, such as the presence of AChR-abs and the responsiveness to immunotherapy. In our study, all the patients at last had received both pyridostigmine and prednisone therapy (Table 6), whether the combined therapy was the reason for the lower transformation rate in the CMG patients needs more detailed clinical trials, because there were a few reports suggesting that treatment with prednisone alone or in combination with azathioprine at onset of MG was able to reduce the possibility of transformation from OMG to GMG . Unfortunately, no responsible factors were identified to predict the transformation. Whether the higher frequent thymoma linked with the higher GMG percentage in our AMG patients implies a possible role of thymic abnormalities in the transformation needs to be uncovered further.
A major concern for treating CMG children is the side effects of long-term use of prednisone, especially the growth failure and Cushing’s reaction [22, 23]. Neurologists and patients’ custodies have been struggling in balancing the improvement of MG symptoms and the side effects. Long-term therapy with prednisone should be theoretically administrated to prevent relapses of MG symptoms, but in reality the failure to grow taller in very young CMG kids is the major reason for failure of treatment. What we usually do is to taper after patients’ symptoms are significantly improved, and at last to discontinue prednisone when they have been asymptomatic for at least 6 months. However, 65.8 % patients (154 out of the 234 exacerbated patients) who stopped taking pyridostigmine or prednisone after they obtained so-called CSR for longer than 1 year became exacerbated (relapsed). The high tendency to relapse induced by discontinuation may indicate that discontinuation of drugs was a dominant risk factor triggering exacerbation. In attempt to prevent the relapses or transformation, azathioprine and thymectomy have been suggested to the CMG patients. However, the Chinese custodies are traditionally reluctant to accept any immunosuppressants for treating the patients with isolated ocular symptoms. The efficacy of azathioprine alone or in combination with low dosage of prednisone needs more controlled studies.
Thymectomy is generally considered an important option for treating MG in adult MG patients if medical treatment proved unsatisfactory [24, 25, 26]. However, it is not generally recommended for non-thymomatous OMG patients [24, 27, 28]. The 34 CMG patients in our study did not generally benefit very much from the surgery (Table 5). Only six out of the CMG patients with GMG had achieved significant clinical improvement, implying that thymectomy could be suggested for CMG patients with GMG (Table 5). Unfortunately, the transformation in the CMG patients had not been efficiently stopped after thymectomy and pharmaceutical therapy, and the symptomatic improvement was not evidently observed in 26 CMG cases with OMG. Since the thymectomized cases had not received the operation until they showed refractory responses to pharmaceutical treatment, whether the timing of the surgery influences the efficacy of thymectomy needs further clinical trials.
Relapses after remission constituted the major concern about the CMG. After the relapses, re-usage of prednisone was proved to be effective in majority of the exacerbated cases. Until now, we failed to pinpoint the factors which may predict the relapsing of MG symptoms. However, half of the cases experienced a preceding event (trigger) before relapses. Discontinuation of prednisone was the major trigger. Notably, infection (especially the upper respiratory tract infection, even the minor ones) was a very common discernible trigger. The possible role of infections even the occult ones in triggering activity of autoimmune diseases has been speculated [29, 30, 31, 32, 33, 34]. The challenging question is after an OMG child has attained so-called pharmaceutical remission for 1 year or longer on low dosage of prednisone, we lack guidelines or markers to discontinue the corticotherapy since long-term usage of prednisone would handicap the children.
Though there were several limitations in the retrospective study, such as the incomplete data of determination of AChR-abs in all cases, and lacking the data of CMG cases who did not receive prednisone therapy, the large number of CMG cases and nearly 10 years follow-up indicate that OMG is far more frequent than GMG in Chinese children. The CMG has a high tendency to relapse after remission. Efforts should be done to reduce the relapses such as developing new drugs or methods with less side effects, and further evaluating efficacy of thymectomy in the treatment of ocular CMG.
We thank Dr. Minhong Sun for her assistance in statistical work. The research was partly supported by a grant from Natural Science Foundation of China (No. 81271320 to Dr. Bu).
Conflicts of interest
All the authors declare no potential conflicts of interest, including relevant financial interests, activities, and relationships.
The Bu Bitao's study concerning the client's right to privacy. The study was approved by the Ethical Commitee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology and was conducted according to the principles of the Declaration of Helsinki. Written informed consent was obtained from the subject, and his study considered Declaration of Helsinki as a statement of ethical principles.
- 13.Jaretzki AR, Barohn RJ, Ernstoff RM, Kaminski HJ, Keesey JC, Penn AS, Sanders DB (2000) Myasthenia gravis: recommendations for clinical research standards. Task force of the Medical Scientific Advisory Board of the Myasthenia Gravis Foundation of America. Neurology 55:16–23CrossRefPubMedGoogle Scholar