Visual and Systemic Outcomes in Pediatric Ocular

Myasthenia Gravis
STACY L. PINELES, ROBERT A. AVERY, HEATHER E. MOSS, RICHARD FINKEL, THANE BLINMAN,
LARRY KAISER, AND GRANT T. LIU
PURPOSE:

To evaluate visual and systemic outcomes in

pediatric patients with purely ocular myasthenia gravis
(OMG) treated at the Childrens Hospital of Philadelphia.
DESIGN: Retrospective chart review.
METHODS: Pediatric patients with OMG seen at a single
institution over a 16-year period with a minimum follow-up
of 1 year were reviewed. Associations of demographic and
clinical characteristics with disease resolution, amblyopia,
and development of generalized symptoms of myasthenia
gravis were analyzed.
RESULTS: Thirty-nine patients were identified, with a
mean age of 5.4 4.8 years and mean follow-up of 4.8
4.3 years. Fifteen patients were treated with pyridostigmine
only, 19 (49%) also received steroids, and 15 (38%)
underwent thymectomy. Four patients (10%) received steroid-sparing immunosuppressive therapy. Resolution occurred in 10 patients, and generalized symptoms eventually
occurred in 9 patients. Although 10 patients were treated
for amblyopia, only 1 had amblyopia at the final visit. There
was no correlation between sex or age with amblyopia or
development of generalized symptoms. Thymectomy, when
performed before the onset of generalized symptoms,
showed a trend toward protection from the development of
generalized symptoms (P .07).
CONCLUSIONS: In our series, 24% of patients had
disease resolution and 23% had generalized symptoms.
Our larger cohort confirms previous findings that treated
and untreated pediatric patients with OMG have a
relatively low risk of developing generalized symptoms
and that related amblyopia is readily reversible. Although
our treatments were more aggressive than those previSupplemental Material available at AJO.com.
Accepted for publication May 5, 2010.
From the Neuro-ophthalmology Service, Childrens Hospital of Philadelphia, and the Division of Neuro-ophthalmology, Departments of
Neurology and Ophthalmology, University of Pennsylvania School of
Medicine, Philadelphia, Pennsylvania (S.L.P., R.A.A., H.E.M., G.T.L.);
Division of Neurology, Childrens Hospital of Philadelphia, and Departments of Neurology and Pediatrics, University of Pennsylvania School of
Medicine, Philadelphia, Pennsylvania (R.F.); the Department of Surgery,
Childrens Hospital of Philadelphia, Philadelphia, Pennsylvania (T.B.);
and the Department of Surgery, University of Pennsylvania School of
Medicine, Philadelphia, Pennsylvania (L.K.). Dr Kaisers current affiliation is the University of Texas Health Science Center at Houston, and
the Department of Cardiothoracic and Vascular Surgery, University of
Texas Medical School at Houston, Houston, Texas.
Inquiries to Grant T. Liu, Division of Neuro-Ophthalmology, Hospital
of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA
19104; e-mail: gliu@mail.med.upenn.edu
0002-9394/$36.00
doi:10.1016/j.ajo.2010.05.002

2010 BY

ously reported, our rates of amblyopia and development

of generalized symptoms are comparable. (Am J Ophthalmol 2010;150:453 459. 2010 by Elsevier Inc.
All rights reserved.)

YASTHENIA GRAVIS IS AN ACQUIRED AUTOIM-

mune disorder in which acetylcholine receptors

(AChRs) within the neuromuscular junction of
skeletal muscle are targeted by autoantibodies. This results
in disrupted neuromuscular transmission and symptoms of
fluctuating and fatigable weakness.1 Myasthenia gravis is
relatively uncommon in the pediatric population, with children accounting for approximately 10% to 15% of cases
annually.2 Approximately 90% of children with myasthenia
gravis will have ophthalmic features, such as ptosis or
ophthalmoplegia, in isolation, or in association with other
systemic weakness.1 In 10% to 15% of children, weakness
is strictly limited to the extraocular muscles, resulting in a
diagnosis of ocular myasthenia gravis (OMG).1 One group
has suggested that in 50% of pediatric patients with OMG,
systemic or bulbar symptoms will develop within 2 years.3
Data regarding the treatment and prognosis of children
with OMG at initial presentation is limited. There are a
few case series, of fewer than 25 patients, that specifically
describe the characteristics and outcomes of relatively
small cohorts of pediatric patients with OMG.3 6 Because
we have seen a relatively larger number of children with
OMG at the Childrens Hospital of Philadelphia and many
of our patients have undergone more aggressive treatment
regimens than those published by previous groups, we
sought to compare our outcomes with those previously
reported. Finally, we set out to evaluate risk factors for the
development of undesirable outcomes, such as amblyopia
and development of generalized symptoms.

METHODS
RECORDS WERE REVIEWED FOR ALL PATIENTS 1 TO 18 YEARS

of age with at least 1 year of follow-up and a diagnosis of

OMG who were seen by one of the coauthors (G.T.L.)
from July 1993 through September 2009 at the Childrens
Hospital of Philadelphia. Pediatric OMG is also often
referred to as juvenile OMG to differentiate it from
congenital myasthenic syndromes and neonatal myasthenia. To meet inclusion into the study, patients had to

ELSEVIER INC. ALL

RIGHTS RESERVED.

453

as a confirmatory test. After the diagnosis of pediatric

OMG is established, the first step in our treatment protocol is to determine whether any treatment is indicated. In
general, children are treated if their ptosis or ophthalmoplegia is potentially amblyogenic. Therefore, if the ptosis is
observed by us, or a parent, to be occluding the visual axis
at any time, or if the child has any degree of manifest
strabismus, treatment is initiated. However, if reliable
visual acuity measurements can be obtained and visual
acuity is symmetric between both eyes, then we might
consider a period of careful observation. Additionally,
anomalous head postures attributed to ptosis or strabismus
and diplopia are indications for treatment. Those with
mild ptosis and normal ocular alignment in primary gaze
usually are left untreated and are observed.
All patients requiring treatment initially are started on
pyridostigmine (2 to 4 mg/kg/day in divided doses). Depending on the severity of the symptoms, children are
followed up approximately 1 to 8 weeks later. Additionally,
a chest magnetic resonance imaging examination usually is
ordered to evaluate for thymic hyperplasia or thymoma. If
the symptoms do not improve, or if there is any worsening
of the symptoms, steroids are recommended. If thymus
abnormalities are found, a thymectomy is considered at
this time. After the symptoms are controlled by steroids, an
attempt to wean the steroids often is undertaken. If the
symptoms cannot be controlled on a low dose of steroids,
then other immunosuppressants are introduced, and
thymectomy is considered. At our institution, thymectomies for younger children with myasthenia gravis are
performed transthorascopically without a sternotomy, and
in older children, they are performed transcervically.
Because these procedures are considered relatively safe and
leave minimal scars, we have a low threshold for noninvasive thymectomy, even in children with solely OMG, as
a means by which to reduce the amount of medication
required, to minimize chronic steroid side effects, and
possibly to decrease the risk of generalized symptoms.9 11
For each patient, data were recorded from the first clinic
visit, including: the age at presentation, ocular and systemic symptoms, visual acuity, stereoacuity, ocular ductions, ocular alignment, and the presence of ptosis. For
verbal children, visual acuity was measured using projected
optotypes appropriate for age (ie, Snellen letters, H-O-T-V,
or Lea symbols). For preverbal testing, Teller acuity was
measured. If Teller acuity could not be measured, then
fixation behaviors were tested. Subsequent visits were
reviewed for changes in visual acuity, ptosis, strabismus,
ocular alignment, and ductions. In addition, data regarding
various treatments initiated, onset of new symptoms (including generalized ones), or resolution of symptoms was
noted. As a rule, we tend to use the term generalized
symptoms instead of systemic generalization because we can
not be sure whether systemic generalization is masked by
treatment. Generalized symptoms include muscular weakness, generalized fatigue, or bulbar symptoms (difficulty

TABLE 1. Criteria for Diagnosing Pediatric Ocular

Myasthenia Gravisa
1. One of the following must be present:
a. Ptosis in one or both upper lids not resulting from local
disease, with or without fatigability and recovery with
rest, by history or examination.
b. Extraocular muscle weakness, with or without fatigability
and recovery with rest, and the absence of an
alternative explanation for the duction abnormality, often
including negative neuroimaging in questionable cases.
2. No pupillary abnormalities except from prior ocular disease,
surgery, or physiologic anisocoria.
3. Positive results from 1 of the following confirmatory tests:
a. Fatigue of the affected muscle with worsening of ptosis
after prolonged upward gaze, or worsening of duction
abnormality after prolonged gaze using the affected
extraocular muscle,
b. Recovery of ptosis or ductional deficits after prolonged
rest,
c. Positive edrophonium test,
d. Abnormal repetitive nerve stimulation,
e. Abnormal acetylcholine receptor binding antibody level.
4. Age at presentation between 1 and 18 years.
5. Lack of generalized symptoms at disease onset.
a

Modification of previously published criteria for adult ocular

myasthenia gravis.7,8

satisfy all of the criteria described in Table 1 for the

diagnosis of pediatric OMG. Children with congenital or
neonatal myasthenia (with an onset less than 6 months of
age) were excluded. Our criteria represent a modification
of the Myasthenia Gravis Foundation of America clinical
classification7 and other published criteria for adults.8
Modifications were based on differences in evaluating
children and adults with OMG, for example, relative
difficulty in obtaining single-fiber electromyography in
children. Patients were excluded from this retrospective
review if they had systemic or bulbar weakness preceding
ocular symptoms.
In general at our institution, pediatric patients suspected
of having OMG first undergo AChR antibody testing as a
confirmatory test. If the antibody level is within normal
limits, patients often are sent for repetitive nerve stimulation testing. In the pediatric population, single-fiber electromyography of the orbicularis oculi usually is not
tolerated; therefore, repetitive nerve stimulation of the
upper extremity often is performed. If both of the preceding tests demonstrate negative results, patients may undergo edrophonium testing, in which edrophonium is
injected intravenously, and then measurements of eyelid
position, ocular ductions, and alignment are repeated in
attempt to evaluate for improvement. Finally, if diagnostic
testing shows negative, equivocal, or unobtainable results,
an improvement in ptosis or extraocular muscle weakness
after a period of rest (i.e., a so-called rest test) often is used
454

AMERICAN JOURNAL

OF

OPHTHALMOLOGY

OCTOBER 2010

TABLE 2. Patient Characteristics for Pediatric Ocular

Myasthenia Gravis
Characteristic

Values

Gender
Age at onset (yrs)
Follow-up (yrs)
Abnormal AchR antibody level
Positive edrophonium test
Abnormal repetitive nerve
stimulation
Generalized symptoms
Disease resolution
Amblyopia diagnosis at any
time
Amblyopia at final visit

18 male, 21 female
5.4 4.8 (range, 1 to 17)
4.8 4.3 (range, 1 to 17)
15/31 abnormal results (48%)
16/18 positive results (88%)
2/7 abnormal results (29%)
9/39 (23%)
10/39 (26%)

RESULTS
A SUMMARY OF PATIENT CHARACTERISTICS IS PRESENTED

10/39 (26%)
1/39 (3%)

AChR acetylcholine receptor.

FIGURE. Bar graph showing the distribution of patient ages at

the time of presentation with ocular myasthenia gravis. Shaded
areas indicate the number of patients who experienced generalized symptoms at any point during their clinical course.

breathing, swallowing, or chewing). Disease remission was

diagnosed in patients who were completely asymptomatic
and had not received any treatments for at least 1 year,
without ptosis, extraocular muscle weakness, strabismus, or
generalized symptoms. Amblyopia was defined as a 2-line
or more difference in best-corrected visual acuity between
each eye. In preverbal children in whom Teller acuity
testing was not performed, amblyopia was defined as a
failure to hold fixation with either eye. Additionally, any
treatment-related complications were noted. The patients
disease status at the final visit was classified using a
modification of the Myasthenia Gravis Foundation of
America postintervention status categories.7 Patients were
categorized into one of the following final outcomes:
complete stable remission, ocular minimal manifestations
VOL. 150, NO. 4

(MMs) with medications, ocular MM without medications, or persistent generalized symptoms.

Logistic regression models were created to evaluate for
associations between disease outcomes (resolution, generalized symptoms, and amblyopia) and individual patient
characteristics (age, sex, follow-up length, diagnostic test
responses, and treatment type). Analysis was carried out
using Stata-10 statistical software (StataCorp, College
Station, Texas, USA). A P value of .05 or less was
considered a statistically significant association.

in Table 2. Thirty-nine patients meeting the study inclusion criteria were identified. Eighteen patients (46%) were
male, and the mean age at presentation was 5.4 4.8 years
(range, 1 to 17 years), with 7 patients (18%) who were 12
years of age or older at the time of presentation. The age
distribution is presented in the Figure. The mean follow-up
time was 4.8 4.3 years (range, 1 to 17 years).
On presentation, all of our cases had ptosis. Strabismus
in primary position was present in 22 (56%) patients, and
ocular duction deficits were present in 20 (51%) patients.
AChR antibody testing was performed in 31 patients, 15 of
whom had an abnormally high level (48% sensitivity).
Repetitive nerve stimulation was performed in 7 patients,
with 2 patients showing an abnormal response (29%
sensitivity). Finally, edrophonium testing was performed in
18 patients, 16 of whom had a positive response with
improvement in ptosis or ophthalmoplegia (88% sensitivity).
Characteristics of patients who had abnormal versus normal
results of AChR antibody testing are presented in Supplemental Table (Supplemental Material at AJO.com).
Generalized symptoms developed in 9 (23%) patients,
manifested by generalized hypotonia (n 1), bulbar
symptoms (n 6), and extremity weakness (n 2). The
mean time to generalized symptoms was 0.75 0.6 years
(range, 1 month to 2 years). The age distribution for those
patients who experienced generalized symptoms is depicted
by shading in the Figure. Ten subjects (26%) had complete
resolution of symptoms without medication for at least 1
year. The mean time to resolution was 2.3 1.5 years
(range, 3 months to 5 years). At the final follow-up visit,
of those patients whose disease had not resolved, 7 had
MM without any medications, 21 had MM with medication, and 1 patient had generalized symptoms (systemic
weakness). During the follow-up period, 10 patients (26%)
were treated for amblyopia with occlusion therapy. At the
final visit, 1 patient (3%) had residual amblyopia ( 2
lines difference in visual acuity). The age of OMG onset
for patients treated for amblyopia ranged from 1 to 6 years.
At the final follow-up visit, 6 (15%) had manifest strabismus, and 18 (46%) had some degree of ptosis. Two cases
had thymic hyperplasia. One case of thymic hyperplasia

PEDIATRIC OCULAR MYASTHENIA GRAVIS

455

TABLE 3. Treatment Outcomes for Pediatric Ocular Myasthenia Gravis

Amblyopia at
Any Time
during
Follow-up

Treatment

No treatment (n 5)
Pyridostigmine (n 15)
Pyridostigmine steroid (n 3)
Pyridostigmine steroid, other
immunosuppressive (n 1)
Pyridostigmine steroid
thymectomy other
immunosuppressive (n 15)
Total

0
1
1

Outcome at Most Recent Follow-up

Generalized Symptoms
at Any Time during
Follow-up

CSR

0
1
1

0
1
8
6
Generalized amblyopia 4
Generalized only 2
Amblyopia only 4
10
9

MM without
Medications

MM with
Medications

Generalized
Symptoms

Amblyopia

2
4
2

3
2
0

0
9
1

0
0
0

0
0
0

0
2

0
2

0
11

1 (systemic weakness)
0

0
1

10

21

CSR complete stable remission (the patient has had no symptoms or signs of ocular myasthenia gravis and was without pharmacologic
therapy for at least 1 year); MM minimal ocular manifestations (no symptoms of functional limitations from myasthenia gravis, but some
weakness on examination of some muscles).

prine (n 1), cyclosporine (n 1), mycophenolate

mofetil (n 1), or monthly intravenous immunoglobulin
(n 4). All of the cases that received long-term steroidsparing treatments had generalized symptoms before initiation of the agent. The reason for proceeding to
thymectomy in those patients who underwent thymectomy
was either a lack of symptomatic improvement on pyridostigmine and steroids (n 8), generalized symptoms
(n 6), or thymic hyperplasia with failure to improve
with pyridostigmine (n 1). There were no reported
complications of treatment other than diarrhea associated
with pyridostigmine.
Risk factor analysis is presented in Table 4. The presence of generalized symptoms was not associated significantly with any of the following risk factors: age (P .12),
gender (P .91), AChR antibody status (P .44), results
of repetitive stimulation (P .3), or results of edrophonium testing (P .3). In addition, none of these factors
seemed to increase significantly the risk of amblyopia (age,
P .07; gender, P .8; AChR antibody status, P .8;
result of repetitive stimulation, P .4; or result of
edrophonium test, P .3), or increase the likelihood of
disease resolution (age, P .35; gender, P .5; AChR
antibody status, P 1.0; results of repetitive stimulation,
P .8; or results of edrophonium testing, P .3). Finally,
there was a statistically significant association between
both generalized symptoms and amblyopia with the use of
steroids and thymectomy (P .05) when all patients who
received the treatment (i.e., steroids or thymectomy) were
compared with all of those who never received steroids or
thymectomy. However, this significant trend was lost when
patients who had generalized symptoms or amblyopia
before thymectomy were removed from the analysis. Additionally, there was a trend toward thymectomy preventing generalized symptoms (n 0/9) when patients who

TABLE 4. Risk Factor Analysis for Pediatric Ocular

Myasthenia Gravisa

Association

General symptoms
Gender
Onset age
AChR antibody status
Disease resolution
Gender
Onset age
AChR ab
Amblyopia
Gender
Onset age
AChR ab
Strabismus

Odds Ratio (95%

Confidence Interval)

P Value
(All Patients)

0.91 (0.2 to 4.1)

0.82 (0.63 to 1.1)
2.17 (0.3 to 16)

.91
.12
.44

1.6 (0.36 to 7.3)

1.07 (0.92 to 1.25)
1 (0.2 to 5)

.5
.35
1

1.23 (0.3 to 5.2)

0.7 (0.53 to 1.02)
0.8 (0.11 to 5.8)
0.8 (0.2 to 3.4)

.8
.07
.8
.8

AChR acetylcholine receptor.

Comparisons made between patient groups with and without
risk factor of interest.
a

was identified on preoperative magnetic resonance imaging

(enlarged thymus), and the second case was identified in
the postthymectomy pathologic analysis (follicular hyperplasia). No thymomas (epithelial cell tumor of the thymus)
were identified in our series.
A summary of treatment regimens in our cases is
presented in Table 3. Medical treatment was initiated in
all but 5 cases. All untreated cases were considered too
mild to warrant therapy. The remaining 34 patients initially were treated with pyridostigmine. Steroid treatment
was added to pyridostigmine in 19 cases. In addition, 15
patients underwent thymectomy. Finally, 4 patients received long-term steroid-sparing agents such as azathio456

AMERICAN JOURNAL

OF

OPHTHALMOLOGY

OCTOBER 2010

TABLE 5. Summary of Studies of Pediatric Ocular Myasthenia Gravis

Author(s), Year

No. of Patients

Schmidt 198217

Generalized
Symptoms

Resolution

Amblyopia

Treatment Regimen

33%

78%

Not described

66% P only
11% P S
11% P T
Regimen for all 34 myasthenics:
12% P only
20% P S
12% P I
56% P T
36% P only
14% P S
36% P T
14% P S T
8% P only
50% P S
25% S only
71% P only
14% P S
14% P S T
13% no treatment
38% P only
8% P S
3% P S I
38% P S T

Mullaney and associates, 20003

34 (all myasthenics),
14 ocular

36%

43%

McCreery and associates, 20025

14

43%

50%

7%

Kim and associates, 20034

24

8%

13%

8%

Ortiz and Borchert, 20086

21

14%

19%

10%

Current study

39

23%

24%

3%

I immunosuppression; P pyridostigmine; S steroids; T thymectomy.

had thymectomy after the onset of generalized symptoms

(n 6) were considered during their prethymectomy state
(i.e., they had generalized symptoms, amblyopia, or both
before thymectomy) and compared with the remaining 24
patients who never received thymectomy (30 patients in
total compared with the 9 described above, P .07). The
mean time to thymectomy after disease onset in this group
of patients was 2.5 2.4 years (range, 0.5 to 7 years).
There was a significant association of increased follow-up
with generalized symptoms (P .03), but not with disease
resolution or amblyopia at the final visit.
Subgroup analysis within patient groups who sought treatment before or after the age of 10 years did not reveal any
significant associations with generalized symptoms, disease
resolution, or amblyopia when gender, presence of strabismus
at onset, AChR antibody status, repetitive nerve stimulation
results, edrophonium results, and treatment method were
evaluated. Additionally, evaluation of those patients with
more than 2 years follow-up did not reveal any significant
associations with the above risk factors.

DISCUSSION
OUR APPROACH TO THE TREATMENT OF PEDIATRIC OMG

results in similar outcomes as those published by other

groups.3 6 Although our risk factor analysis identified
VOL. 150, NO. 4

steroids and thymectomy as being significantly associated

with poorer outcomes, this finding is likely the result of the
retrospective and nonrandomized nature of our study; the
patients who received more aggressive treatments were
treated as such because their disease was more difficult to
control, and therefore potentially was worse. Additionally,
when those patients who underwent thymectomy after the
onset of generalized symptoms were removed from the analysis, the significance of this association was lost. Therefore,
this finding cannot be interpreted in the context of this study
and could be answered only in a prospective, randomized trial
or possibly in a larger, multicenter, retrospective review.
Importantly, when the subgroup of patients (n 9) who had
not experienced generalized symptoms before thymectomy
were evaluated separately, there was a trend toward protection from the development of generalized symptoms from
thymectomy (P .07).
Several groups have reported findings in pediatric patients with OMG (Table 5). Notably, in 2003, Kim and
associates reported the outcomes of a cohort of 24 Korean
children with isolated OMG.4 A very large proportion of
this group of patients received steroids (75%), yet none of
them underwent thymectomy. The rates of undesirable
outcomes, such as generalized symptoms (8%) and amblyopia (8%), were superior to our outcomes. However,
although they report disease remission in 16% of cases,
none of their patients achieved complete remission of eye

PEDIATRIC OCULAR MYASTHENIA GRAVIS

457

movement abnormalities. In response to the study by Kim

and associates, in 2008, Ortiz and Borchert asserted that
aggressive treatment with steroids was not necessarily
indicated in pediatric OMG when they reviewed their
outcomes in 21 preadolescent patients from Childrens
Hospital of Los Angeles.6 This group of patients was
treated in a less aggressive fashion, with only 29% of
patients receiving steroids and 10% of patients undergoing
thymectomy. However, although their rate of final amblyopia was 3 times as high as that of subjects from our study
(10% vs 3%) and was similar to that of patients in the
study by Kim and associates,4 their rate of complete
resolution (20%) was superior to that of Kim and associates
and almost equivalent to ours.
Interestingly, our protocol for treatment seems similar to
the approach taken by Ortiz and Borchert.6 However, the
number of patients who actually received additional immunosuppressives and thymectomy treatments was much
higher, and therefore implies that either our patients did
not respond as well to pyridostigmine, that our threshold
for recommending thymectomy is lower, or both. If our
patients did not respond as well to pyridostigmine and
therefore required additional agents, one might consider
subtle differences, whether based on geography, race, or
age, that may dictate a patients disease severity or response to pyridostigmine. Although we did not track the
ethnic background of our cases, it is well known that race
plays a role in genetic susceptibility and disease severity.12
Additionally, the mean age of onset in our cohort was
older than that of Ortiz and Borchert (5.4 vs 2.2 years),
and because prepubertal OMG has a higher rate of spontaneous remission,12 it is possible that their cohort had a
better prognosis at the outset. This theory is offset by the
Figure, which demonstrates that most of the patients who
had generalized symptoms were younger than 10 years.
An alternative explanation for our more liberal use of
thymectomy is that our group has a lower threshold for
recommending thymectomy, given the less invasive (transthoracic and transcervical) approaches that are available
to us.10,11 Although there is no randomized controlled
trial, there has been some evidence that thymectomy is
effective in controlling juvenile OMG.13 Our data seem to
support this assertion, because none of the patients who
underwent thymectomy before experiencing generalized
symptoms experienced generalized symptoms after surgery.
Clinicians should keep in mind that the effect of thymectomy can take months to years for a full effect.11 Although

the thymus often appears normal on magnetic resonance

imaging, a high rate of follicular hyperplasia (89%) has
been found in postthymectomy specimens of juvenile
generalized myasthenic patients.14 Finally, it is possible
that the frequency of generalized symptoms in our thymectomy group is artificially low in that several of the patients
underwent thymectomy more than 2 years after disease
onset, at which time their risk of generalization is lowered
already.
Overall, our rate of the development of generalized
symptoms (23%) was lower than early case series of
pediatric OMG (36% to 43%)3,5 and that of adult OMG
(31% to 49%).15,16 These rates corroborate the notion that
generalized symptoms are less common in pediatric OMG
than in adult OMG. In the adult OMG literature, it has
been suggested that immunosuppressive agents may reduce
the conversion of OMG to generalized myasthenia gravis.3,5,8 This trend was not observed in our cases; however,
the retrospective nature of our study would preclude
interpretation related to this point. Although we have a
relatively long follow-up period, it is certainly possible that
some of our cases will have recurrence of the disease or
generalized symptoms at a future time, because untoward
events are possible after many years of disease quiescence.
Importantly, this study is retrospective, and therefore
cannot be used to assess definitively the role that various
treatments may have in inducing disease remission or
preventing undesirable outcomes. Furthermore, our population had a wide range of ages, presenting symptoms, and
follow-up length. However, our data are useful in that they
form the largest single-center cohort of pediatric OMG
patients described to date.
Taken together, the results of our study combined with
those of the studies described above indicate that pediatric
OMG is a readily treatable disease in most patients. Poor
outcomes such as generalized symptoms (8% to 43%) and
amblyopia (0% to 10%) are comparable among 3 groups
from different institutions,4 6 regardless of treatment type.
Importantly, we were unable to identify patient characteristics that are associated with amblyopia, and thus clinicians must remain vigilant in all children who with OMG
who are younger than 8 to 10 years. Although it would be
difficult to perform because of the relative paucity of cases,
a prospective, multicenter study of treatment outcomes in
pediatric OMG would be useful to assess accurately the
role that more aggressive treatment regimens may play in
preventing undesirable outcomes.

PUBLICATION OF THIS ARTICLE WAS SUPPORTED BY GRANT SUPPORT FROM THE NATIONAL INSTITUTES OF HEALTH,
Bethesda, Maryland; PTC Therapeutics; the SMA Foundation; and Genzyme Corporation (all for Dr. Finkel). The remaining authors indicate no
financial support. Dr Finkel has served as an expert witness. Dr Liu has served as a consultant for Ipsen and has received lectureship honoraria from Merck
and Endo. The remaining authors indicate no financial conflict of interest. Involved in design of study (S.L.P., R.A.A., H.E.M., G.T.L., R.F.); Conduct
of study (S.L.P., R.A.A., H.E.M., G.T.L., R.F.); Collection of data (S.L.P., R.A.A., G.T.L., R.F., T.B., L.K.); management and analysis of data (S.L.P.,
R.A.A., H.E.M., G.T.L., R.F.); interpretation of data (S.L.P., R.A.A., H.E.M., T.B., L.K., R.F., G.T.L.); and Preparation, review, and approval of the
manuscript (S.L.P., R.A.A., H.E.M., R.F., T.B., L.K., R.F., G.T.L.). This study was approved by the Childrens Hospital of Philadelphia Institutional
Review Board and adhered to the Declaration of Helsinki and all relevant federal and state privacy laws.

458

AMERICAN JOURNAL

OF

OPHTHALMOLOGY

OCTOBER 2010

REFERENCES
1. Andrews P. Autoimmune myasthenia gravis in childhood.
Semin Neurol 2004(1);24:101110.
2. Phillips L, Torner J, Anderson M, Cox G. The epidemiology
of myasthenia gravis in central and western Virginia. Neurology 1992;42(10):1888 1893.
3. Mullaney P, Vajsar J, Smith R, Buncic J. The natural history
and ophthalmic involvement in childhood myasthenia gravis
at The Hospital for Sick Children. Ophthalmology 2000(3);
107:504 510.
4. Kim J, Hwang J, Hwang Y, Kim K, Chae J. Childhood ocular
myasthenia gravis. Ophthalmology 2003;110(7):1458 1462.
5. McCreery K, Hussein M, Lee A, Paysse E, Chandran R,
Coats D. Major review: the clinical spectrum of pediatric
myasthenia gravis: blepharoptosis, ophthalmoplegia, and
strabismus. A report of 14 cases. Binocul V Strabismus Q
2002;17(5):181186.
6. Ortiz S, Borchert M. Long-term outcomes of pediatric ocular
myasthenia gravis. Ophthalmology 2008;115(7):12451248.
7. Jaretzki A, Barohn R, Ernstoff R, et al. Myasthenia gravis:
recommendations for clinical research standards. Neurology
2000;55(1):16 23.
8. Kupersmith M. Does early immunotherapy reduce the conversion of ocular myasthenia gravis to generalized myasthenia gravis? J Neuro-Ophthalmol 2003;23(4):2003.

VOL. 150, NO. 4

9. Kolski H, Kim P, Vaijsar J. Video-assisted thorascopic

thmectomy in juvenile myasthenia gravis. J Child Neurol
2001;16(8):569 573.
10. Shrager J, Deeb M, Mick R, et al. Transcervical thymectomy for
myasthenia gravis achieves results comparable to thymectomy
by sternotomy. Ann Thorac Surg 2002;74(2):320 327.
11. Shrager J, Nathan D, Brinster C, et al. Outcomes after 151
extended transcervical thymectomies for myasthenia gravis.
Ann Thorac Surg 2006;82(5):18631869.
12. Andrews P, Massey J, Howard J, Sanders D. Race, sex, and
puberty influence onset, severity, and outcome in juvenile
myasthenia gravis. Neurology 1994;44(7):1208 1214.
13. Gadient P, Bolton J, Puri V. Juvenile myasthenia gravis:
three case reports and a literature review. J Child Neurol
2009;24(5):584 590.
14. Lindner A, Schalke B, Toyka K. Outcome in juvenile-onset
myasthenia gravis: a retrospective study with long-erm follow-up of 79 patients. J Neurol Neurosurg Psychiatry 1997;
244(8):515520.
15. Bever C, Aquino A, Penn A, Lovelace R, Rowland L. Prognosis
of ocular myasthenia. Ann Neurol 1983;14(5):516 519.
16. Sommer N, Sigg B, Melms A, et al. Ocular myasthenia
gravis: response to long term immunosuppressive treatment.
J Neurol Neurosurg Psychiatry 1997;62(8):156 162.
17. Schmidt D. Prognosis of ocular myasthenia gravis in childhood. Neuro-Ophthalmology 1983;3(2):117124.

PEDIATRIC OCULAR MYASTHENIA GRAVIS

459

Biosketch
Stacy L. Pineles, MD, is currently a neuro-ophthalmology fellow at the University of Pennsylvania and the Childrens
Hospital of Philadelphia. She received her medical degree from the University of Pennsylvania, and completed her
ophthalmology residency and fellowship in pediatric ophthalmology at the Jules Stein Eye Institute at the University of
California, Los Angeles. After the completion of her fellowship, she will return to Los Angeles to join the faculty at the
Jules Stein Eye Institute.

VOL. 150, NO. 4

PEDIATRIC OCULAR MYASTHENIA GRAVIS

459.e1

Biosketch
Dr. Liu, a Professor of Neurology and Ophthalmology, received his medical degree from Columbia University. He
completed a residency at the Harvard-Longwood Neurology Program and a fellowship at the Bascom-Palmer Eye Institute.
Although he sees both adult and children with neuro-ophthalmic problems, his special interest is in pediatric
neuro-ophthalmology.

459.e2

AMERICAN JOURNAL

OF

OPHTHALMOLOGY

OCTOBER 2010

SUPPLEMENTAL TABLE. Patient Characteristics Based on Acetylcholine Receptor Antibody Level

Characteristics

AChR Ab (n 15)

AChR Ab (n 16)

% Male
Mean age at onset (yrs)
Follow-up (mos)
Generalized symptoms
Resolution
Amblyopia

7 (47%)
5.8 4.7
71 62
4 (27%)
4 (27%)
3 (20%)

7 (44%)
5.3 5.1
34 32
2 (13%)
5 (31%)
3 (19%)

AChR ab acetylcholine receptor antibody level; abnormally high results; normal results.

VOL. 150, NO. 4

PEDIATRIC OCULAR MYASTHENIA GRAVIS

459.e3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.