749

Update on Ocular Myasthenia Gravis

Meabh O’Hare, MBBCh, BAO1 Christopher Doughty, MD1

1 Department of Neurology, Brigham & Women’s Hospital, Boston Address for correspondence Meabh O’Hare, MBBCh, BAO,
Department of Neurology, Brigham and Women’s Hospital, 60
Semin Neurol 2019;39:749–760. Fenwood Rd, Boston, MA 02115 (e-mail: mohare1@partners.org).

Abstract Myasthenia gravis is an antibody-mediated autoimmune disorder of the post-synaptic

Keywords neuromuscular junction resulting in fluctuating, fatigable weakness. Most patients first
► ocular myasthenia present with extraocular symptoms (diplopia and/or ptosis), and in 15% of cases
gravis symptoms will remain restricted to only the extraocular muscles (ocular myasthenia
► ptosis gravis [OMG]). The history and clinical examination are of the utmost importance in

Downloaded by: Collections and Technical Services Department. Copyrighted material.

► diplopia correctly identifying OMG patients, as supportive serologic or electrodiagnostic
► neuromuscular studies are frequently nondiagnostic. In this review, we outline a diagnostic approach
junction to OMG (focusing on key clinical features), discuss therapeutic options, and highlight
► acetylcholine recent developments in the understanding of OMG.
receptor antibody

Myasthenia gravis (MG) is an antibody-mediated autoimmune lipoprotein receptor-related protein-4 (LRP-4) antibody-asso-
disorder affecting the postsynaptic neuromuscular junction, ciated OMG, the emerging phenomenon of checkpoint inhibi-
resulting in fluctuating weakness that worsens with exertion tor induced myasthenia gravis, and the controversy regarding
and improves with rest.1 Extraocular symptoms—diplopia the role of early immunotherapy in potentially reducing the
and/or ptosis—are among the initial presenting complaints risk of generalization in OMG.
in up to 85% of MG patients,2,3 and are the only symptoms
present at disease onset in up to 50%.3 The majority of these
Pathophysiology
patients will go on to develop generalized muscular weakness
(generalized myasthenia gravis [GMG]), but in approximately The underlying pathophysiology of MG relates to disruption
15% of cases, the disease remains restricted to the ocular of the normal structure and function of the neuromuscular
muscles (ocular myasthenia gravis [OMG]).2,4 Ocular muscles, junction (NMJ).6 In the healthy state, an action potential (AP)
for the purpose of defining OMG, include the extraocular reaching a motor nerve terminal results in Ca2þ entry via
muscles controlling eye movements (i.e., the superior, inferior, P/Q-type Ca2þ channels. Synaptic vesicles containing acetyl-
medial, and lateral recti, and the superior and inferior obli- choline (ACh) then fuse with the presynaptic membrane,
ques), the levator palpebrae, and the orbicularis oculi.5 releasing their contents. ACh interacts with acetylcholine
Certain unique challenges are presented in making the receptors (AChRs) on the postsynaptic membrane, resulting
diagnosis of OMG, as it is common for supportive tests such in cation-specific channel opening that generates a localized
as serologic or electrophysiologic studies to be nondiagnostic end-plate potential (EPP). Normally, this EPP is more than
when disease is restricted to the extraocular muscles. The sufficient to activate postsynaptic voltage-gated Naþ chan-
clinician must rely heavily on the history and clinical exami- nels and trigger a muscle fiber AP, resulting in intracellular
nation in making the diagnosis. Therefore, the importance of Ca2þ release and muscle fiber contraction.6,7 This built-in
recognizing classic OMG features and avoiding common diag- redundancy wherein the EPP amplitude exceeds that re-
nostic pitfalls cannot be overemphasized. In this review, the quired to trigger an AP is termed the “safety factor.”6
authors outline a diagnostic approach to OMG focusing on A healthy postsynaptic NMJ is a highly organized, tightly
simple bedside exam maneuvers and discuss treatment strat- folded structure with acetylcholine receptors (AChRs) densely
egies suitable for affected patients. New developments in the concentrated at the apices of these folds, and voltage-gated
understanding of OMG are highlighted, including discussion of Naþ channels concentrated in the troughs (►Fig. 1). Structural

Issue Theme Neuro-Ophthalmology; Copyright © 2019 by Thieme Medical DOI https://doi.org/

Guest Editor, Sashank Prasad, MD Publishers, Inc., 333 Seventh Avenue, 10.1055/s-0039-1700527.
New York, NY 10001, USA. ISSN 0271-8235.
Tel: +1(212) 584-4662.
750 Update on Ocular Myasthenia Gravis O’Hare, Doughty

represent false negative testing in patients that harbor an

autoantibody that is not detected by conventional means. For
example, the diagnostic yield of AChR Abs testing is significant-
ly increased by using a cell-based assay, rather than the typical
radioimmunoprecipitation technique.12,13 Others may have
novel autoantibodies that have not yet been described. Recent-
ly, antibodies directed against cortactin (a postsynaptic protein
required for clustering of AChRs) have been identified in 24% of
double seronegative OMG patients.14 The pathogenicity and
specificity of cortactin antibodies remains to be determined.15
The primary site of AChR Ab production is the thymus
gland, which plays a key role in MG pathogenesis.8,9 The
thymus is the site of normal T-cell maturation, and is critical
in the establishment of central immune tolerance.9 Normally
the thymus undergoes involution in adulthood,9 but 70% of MG
Fig. 1 Electron microscopy of the neuromuscular junction (stained
for AChR). (A) Normal individual. (B) A patient with moderately severe
patients have evidence of thymic follicular hyperplasia,16 and

Downloaded by: Collections and Technical Services Department. Copyrighted material.

GMG. The NMJ in myasthenia demonstrates deficiency of AChR about 15% have thymoma.17 The prevalence of thymic pathol-
staining as well as degeneration of the postsynaptic architecture. ogy in patients with purely ocular symptoms has not been
Reproduced from: Engel129 with permission. AChR, acetylcholine clearly described. Thymomas are more commonly identified in
receptors; GMG, generalized myasthenia gravis; NMJ, neuromuscular
late-onset MG (i.e., >50 years old), and are strongly associated
junction.
with both AChR Ab seropositivity and generalized disease.8
The presence of antibodies targeting components of striated
proteins including low-density lipoprotein receptor-related muscle (including titin and ryanodine receptors)8 is also
protein 4 (LRP-4) and muscle-specific receptor tyrosine kinase predictive of thymoma, particularly in younger patients.18
receptor (MuSK) are required for the normal clustering of It is not entirely clear why MG involves extraocular
AChRs.6 The structural organization of the NMJ is critical for muscles so prominently.19 Extraocular muscles may be
the coupling of the EPP to AP generation.6 In healthy NMJs, the more prone to fatigability due to their rapid firing rate.
quantity of ACh released from the presynaptic membrane Motor end plates within extraocular muscles may also
progressively decreases under conditions of sustained activity have physiologic characteristics that confer increased sus-
(a normal phenomenon termed “synaptic rundown”). Howev- ceptibility to the effects of MG, such as lower concentration
er, due to the normal built-in safety factor, EPPs remain large of AChRs and the presence of multiterminal muscle fibers
enough to trigger an AP. In MG, disruption of normal NMJ which lack secondary synaptic folding.19,20 In addition,
function results in a loss of the safety factor, meaning that complement regulatory genes are expressed differently in
some EPPs are of insufficient amplitude to trigger a muscle extraocular muscles, which may make their NMJs more
fiber AP. The decremental muscle response with sustained susceptible to complement-mediated tissue injury.19
neural activity manifests clinically as the characteristic fatiga-
bility seen in this condition.7
Clinical Presentation and Examination
In MG, a number of pathogenic autoantibodies have been
Findings
identified, targeting different components of the NMJ. Anti-
bodies targeting the AChR (AChR Abs) are most common and The estimated incidence of MG is approximately 10 cases per
result in complement-mediated destruction of the postsyn- million person-years, with a prevalence of about 80 cases per
aptic membrane.8 In addition, AChR Abs block the action of million people.21 The age of onset of MG is bimodal in women
ACh and cause crosslinking of AChRs, thereby increasing the (peaking in the 30s and then in the 70s) but is unimodal in
rates of receptor degradation.6,8,9 The pathogenic mecha- men (peaking in the 70s).21 A similar distribution is seen in
nism of anti-MuSK Abs, which are primarily immunoglobu- OMG.22 OMG is slightly more common in men, with a male:
lin G4 (IgG4) subtype Abs that do not fix complement,8 is less female ratio of about 3:2.22
clear. However, MuSK is known to be critical for the mainte- Patients with OMG complain of drooping of the eyelids
nance of both pre- and postsynaptic structure and function, and/or doubling or blurring of their vision. Fatigability of
and anti-MuSK Abs appear to impair normal MuSK activity, these symptoms is a key feature – patients often describe
leading to decreased ACh release, as well as loss of normal feeling best first thing in the morning and worst in the
AChR clustering.10 Antibodies directed at LRP-4 result in evening, and may report improvement after rest or napping.
impaired NMJ function owing to interference with MuSK Diplopia can take a variety of forms, such as vertical, hori-
activation and signaling, as well as complement fixation and zontal, oblique, or any combination of these. Diplopia that
postsynaptic membrane destruction.11 varies in character over time is highly suggestive of MG.
It is important to be aware that a significant proportion of On examination, ophthalmoplegia of any combination of
OMG patients have no detectable antibodies against these extraocular muscles may be found, including any isolated
targets, and the pathophysiology underlying these cases extraocular muscle. OMG can accordingly be confused with
remains unclear. Many of these “seronegative” cases may disorders such as neuropathies of cranial nerves III, IV, or VI;

Seminars in Neurology Vol. 39 No. 6/2019

 Update on Ocular Myasthenia Gravis O’Hare, Doughty 751

should not be overrelied upon. Common differential consider-

ations are outlined in ►Table 1. Careful examination of pupil-
lary function and eye closure is particularly important; as
discussed above, pupillary involvement should automatically
prompt consideration of a cranial nerve III palsy rather than
MG as the cause of ophthalmoplegia/ptosis. Conversely, weak-
ness of eye closure in association with ophthalmoplegia makes
MG more likely, as it is not seen with cranial nerve III, IV, or VI
palsies or with thyroid ophthalmopathy. Given that intracra-
nial mass lesions such as tumors or aneurysms can mimic OMG
through effects on cranial nerves or their nuclei, imaging of the
brain is sometimes necessary in cases of diagnostic uncertain-
Fig. 2 In this patient with bilateral ptosis due to ocular myasthenia ty.33 Decompensated underlying strabismus (e.g. caused by
gravis, taping up the right eyelid has resulted in worsening ptosis of congenital esophoria or IV nerve palsy) can also mimic OMG, as
the left eyelid, to the point of complete eye closure (“curtain sign”). patients often report symptoms that worsen with fatigue;
however, serial examinations in these cases do not demon-

Downloaded by: Collections and Technical Services Department. Copyrighted material.

internuclear ophthalmoplegia; or vertical gaze palsy. An es- strate dynamic changes in the pattern of ocular misalignment.
sential differentiating feature is ophthalmoplegia that fatigues It is also important to inquire about exposure to certain
or changes in character. Fatigability should be sought on drugs that can induce MG. Historically, D-penicillamine
examination even if patients do not volunteer it, with the (used in the treatment of rheumatoid arthritis) was known
caveat that it is not specific to MG. Sustained upward gaze may to induce AChR Ab-positive MG in up to 7% of treated
provoke or exacerbate ptosis or diplopia,23 and persistent gaze patients.34,35 In contemporary practice, MG is now increas-
in the direction in which diplopia is most symptomatic may ingly being recognized as a complication of immune check-
also provoke or exacerbate symptoms.24 Examining the eyes at point inhibitors (ICPis). These medications are being
different times during the patient encounter may reveal increasingly used to upregulate the immune system to target
changing patterns of extraocular weakness. In cases of mild a variety of malignancies. ICPis can lead to off-target im-
ophthalmoplegia, the alternate crosscover test or Maddox rod mune-related adverse events (irAEs) affecting essentially any
testing can be helpful in demonstrating a subtle phoria that system in the body. The incidence of MG after ICPi use has
worsens with sustained gaze. been estimated at 0.12 to 0.2%.36,37 This can present as
Ptosis can be bilateral or unilateral; unilateral ptosis that isolated OMG, but more commonly patients with ICPi-relat-
alternates between sides is particularly suggestive of MG ed MG present with more severe generalized symptoms,
(►Fig. 2).25 Because of the equal bilateral innervation to the frequently requiring ventilator support.36–38 ICPis have also
levator palpebrae muscles (Hering’s law), passively raising a been reported to provoke severe exacerbations in those with
seemingly unilateral ptotic eyelid may sometimes unmask known myasthenia gravis, even patients previously with
subtle ptosis on the other side (a phenomenon called the only OMG,39 or in pharmacologic remission.40
“curtain sign”).26,27 Weakness of orbicularis oculi, leading to Symptoms of ICPi-associated MG typically begin after the
weak or incomplete eye closure (the “peek sign”), is also very first or second cycle of treatment, a median of 5 weeks after
common in OMG.28 Strength of other facial, bulbar, or limb initiation of treatment.36–38 In addition, ICPis can induce
muscles should be carefully examined, as weakness of any of myositis, which commonly involves bulbar and oculomotor
these would indicate GMG. muscles, making the distinction from MG challenging. More-
Bedside assessments of pupillary function are normal in over, overlap of MG and myositis frequently occurs.36,41
MG, which can help to distinguish it from other causes of When overlap is suspected, a decremental response to
complex ophthalmoparesis, including third nerve involve- slow repetitive nerve stimulation and positive AChR Abs
ment. However, pupillometric studies in patients with MG suggest MG, whereas an elevated creatine kinase (CK) sug-
have demonstrated subtle subclinical changes in the velocity gests myositis. Cases of MG with or without myositis are
of the pupillary response,29,30 as well as fatigability of the sometimes accompanied by concurrent myocarditis,36 so
pupillary response (as demonstrated by a reduction in the screening with troponin levels, electrocardiogram, and/or
oscillatory rate of hippus, following sustained light expo- echocardiography may be considered in those with respira-
sure31), and fatigability of the accommodation reflex, indi- tory symptoms or an elevated CK.42 A major difference
cating minor involvement of intrinsic ocular muscles.32 between ICPi-associated MG and idiopathic MG is that the
illness may be monophasic; initial treatment with pyridos-
tigmine and corticosteroids (or intravenous immunoglobu-
Differential Diagnosis
lins [IVIg] or plasma exchange in severe cases) may result in
A broad differential diagnosis should be considered for symptom resolution without the need for corticosteroid-
patients presenting with symptoms and signs suggestive of sparing agents. If initial symptoms are restricted to OMG or
OMG, particularly those with ambiguous exam findings and only mild generalized weakness, the American Society of
negative diagnostic testing. As discussed further below, ab- Clinical Oncology (ASCO) guidelines allow for consideration
normalities on electrodiagnostic testing are nonspecific, and of restarting the ICPi once symptoms have resolved.42

Seminars in Neurology Vol. 39 No. 6/2019

752 Update on Ocular Myasthenia Gravis O’Hare, Doughty

Table 1 Mimics of ocular myasthenia gravis

Disease Distinguishing features

Thyroid ophthalmopathy Thyroid eye disease (also known as Grave’s orbitopathy) can cause variable
ophthalmoplegia but typically does not cause ptosis or weakness of orbicularis
oculi. It may be associated with proptosis or periorbital edema.120 In some cases,
thyroid ophthalmopathy and OMG can coexist.
Mitochondrial myopathies Chronic progressive external ophthalmoplegia (CPEO) causes bilateral, sym-
metric ptosis and severe restriction of eye movements. Despite the degree of
ophthalmoparesis, diplopia is often absent.121 Associated findings indicative of
cerebellar and/or retinal pathology may also be seen when CPEO occurs in the
context of Kearns–Sayre syndrome.122
Cranial neuropathies Any process affecting cranial nerves III, IV, and/or VI can be confused with OMG.
Intracranial mass lesions (including the cavernous sinus), brainstem infarction,
trauma, infection, and Miller–Fisher variant acute inflammatory demyelinating
polyneuropathy can all affect multiple cranial nerves and thereby mimic OMG.
Pupillary involvement is not seen with MG and accordingly should raise a
high degree of concern for an alternative pathology, especially a compressive

Downloaded by: Collections and Technical Services Department. Copyrighted material.

lesion affecting cranial nerve III.
Oculopharyngeal muscular dystrophy (OPMD) OPMD causes progressive bilateral ptosis which may be asymmetric. Retrospec-
tively identifying an insidious onset (e.g. in old photographs) can be helpful.
Ophthalmoparesis occurs but diplopia is less common. Patients often also
experience dysarthria and dysphagia.123
Myotonic dystrophy Myotonic dystrophy (most commonly type 1) can cause bilateral ptosis and
some degree of ophthalmoparesis. Grip or percussion myotonia is usually evident
on exam, as is weakness of neck flexion and distal limb muscles. Other features
include dysphagia, dysarthria, cataracts, and cardiac arrhythmias.124
Congenital myasthenic syndromes Congenital myasthenic syndromes are rare genetic disorders resulting in NMJ
dysfunction. Some may present in adulthood, mimicking seronegative autoim-
mune MG and should be considered if usual MG treatment is ineffective.125
Congenital myopathies Certain congenital myopathies (centronuclear myopathy and core myopathies)
can present in adulthood with prominent ophthalmoplegia and/or ptosis.
Prominent bulbar symptoms typically accompany the extraocular weakness, as
well as frequent respiratory involvement.126
Decompensated phoria An underlying tendency toward mild ocular misalignment is common, and
typically overcome by compensatory mechanisms. These mechanisms can
decompensate over time, leading to episodes of blurred or double vision.127
Convergence insufficiency Convergence insufficiency can cause symptomatic diplopia due to an inability to
maintain binocular fusion with near vision. This is often a primary idiopathic
condition, but can occur as a consequence of traumatic, ischemic, neurode-
generative or metabolic insults.128

Abbreviations: MG, myasthenia gravis; NMJ, neuromuscular junction; OMG, ocular myasthenia gravis.

Diagnosis and Testing overestimates given inherent limitations of the case-

control design of the included studies. One retrospective
An overview of the diagnostic approach is illustrated in cohort study (in which the ice test was routinely per-
►Fig. 3. Although serologic and electrophysiologic testing formed on all patients presenting with ptosis) reported
play an important role in the diagnosis of MG, it is important sensitivity of 92% and specificity of 79% for OMG.46
to recall that serologic studies are frequently negative in • The “rest test”: the patient lies down in a dark room with
OMG, and typical electrodiagnostic findings can be nonspe- eyes closed for 30 minutes. The test is positive if this results
cific. Accordingly, a number of simple bedside maneuvers, in an objective improvement in ptosis or ophthalmopare-
listed below, have an important role in the evaluation of sis.47 A single case–control study reported very high sensi-
patients with suspected OMG: tivity and specificity for this test (99 and 91%, respectively).45
• Cogan’s lid twitch: this sign is provoked by asking the patient
• The “ice test”: applying ice to a ptotic eyelid results in to gaze downwards (thus resting the levator palpebrae), and
temporary improvement in ptosis due to the enhance- then make a saccade back to the primary position. The upper
ment of neuromuscular transmission at cold tempera- eyelid briefly overshoots into retraction before then settling
tures (►Fig. 4).43 Ice should be applied for at least into the ptotic position.48 In one cohort of patients with
2 minutes, and an improvement of at least 2 mm in ptosis isolated ptosis (bilateral or unilateral), sensitivity and speci-
is considered a positive result.44 Pooled estimates of ficity of Cogan’s lid twitch for OMG were 50 and 91.7%,
sensitivity and specificity for this test have been reported respectively, with false positives seen in levator dehiscence
as 94 and 97%, respectively45; however, these are likely and mitochondrial myopathy.49 A false positive Cogan’s lid

Seminars in Neurology Vol. 39 No. 6/2019

 Update on Ocular Myasthenia Gravis O’Hare, Doughty 753

Downloaded by: Collections and Technical Services Department. Copyrighted material.

Fig. 3 Suggested diagnostic algorithm for suspected OMG. AChR, acetylcholine receptors; CMAP, compound muscle action potential; GMG,
generalized myasthenia gravis; LRP, lipoprotein receptor-related protein; MRI, magnetic resonance imaging; MuSK, muscle-specific receptor
tyrosine kinase receptor; OMG, ocular myasthenia gravis; RNS, repetitive nerve stimulation; SFEMG, single-fiber electromyogram.

twitch has also previously been reported with parasellar specificity is estimated to be 98 to 100%.45 False positives
meningiomas50 and dorsal midbrain gliomas.51 have been reported in cases of motor neuron disease and LEMS,
• The forced eyelid closure test (Bienfang’s test): this is a as well as in asymptomatic patients with other autoimmune
modification of the Cogan’s lid twitch in which sustained diseases, or thymoma.53 The sensitivity of serologic testing,
voluntary contraction of the orbicularis oculi for 5 to however, is poor in OMG. While as many as 90% of GMG
10 seconds (to ensure a period of complete levator palpe- patients are AChR Ab positive, only about 50% of those with
brae relaxation) that enhances the appearance of a transient OMG will test positive.45,54–59 Different AChR subtypes have
upward lid twitch when eyes are reopened.52 In one study, been identified: binding, modulating, and blocking AChR Abs.
sensitivity was 94% and sensitivity was 91% for the diagnosis Binding Abs are the most important of these, but testing for
of OMG. False positives were seen in Lambert–Eaton myas- modulating Abs in addition leads to slightly increased test
thenic syndrome (LEMS) and decompensated phoria.52 sensitivity.53,60 AChR blocking Abs are of lower clinical utility,
as they are not detected in isolation.53,60
Serologic Testing Approximately 40 to 50% of generalized MG patients
AChR Ab positivity is the most specific diagnostic test for negative for AChR Abs will test positive for anti-MuSK
OMG–in cases with suggestive symptoms and signs, the Abs,61–63 with a female predominance.63 Bulbar, neck, and

Seminars in Neurology Vol. 39 No. 6/2019

754 Update on Ocular Myasthenia Gravis O’Hare, Doughty

Fig. 4 This patient with ocular myasthenia gravis has baseline asymmetric bilateral ptosis (A). After application of the ice pack (B), the ptosis
objectively improves (C). (Images courtesy Sashank Prasad, MD).

Downloaded by: Collections and Technical Services Department. Copyrighted material.

respiratory involvement are prominent features seen in this this rate results in a depletion of the presynaptic ACh
phenotype.58,61,62 Initial descriptions of the clinical pheno- vesicles, the safety factor present in healthy NMJs results
type associated with anti-MuSK seropositivity suggested in stable CMAP amplitudes. In MG, however, repetitive
that extraocular muscles were less commonly affected,61,64 stimulation results in EPPs that dip below the threshold
but more recent reports indicate at least some ocular in- required to generate each all-or-nothing muscle fiber AP,
volvement in the majority of cases.58 However, anti-MuSK so summative CMAPs decrease in amplitude with repetitive
Abs are only rarely found in isolated OMG (detected in 5% of stimulation. Typically the first CMAP is compared with the
one OMG cohort56). fourth or fifth CMAP, and a decrement of >10% is considered
Anti-LRP-4 Abs have been detected in about 20% of a positive test.75,76 Repetitive nerve stimulation (RNS) has
“double-negative” MG (i.e., negative for both AChR and relatively low sensitivity for OMG (with abnormalities
MuSK Abs),65 although estimates of LRP-4 seropositivity detected in about 30% of cases).45,77 RNS test sensitivity is
vary widely between studies (range: 2–50%).66–68 Clinically, improved by focusing the study on muscles that are clinically
anti-LRP-4 MG tends to present with milder symptoms than weak, so testing orbicularis oculi offers higher yield in
AChR-positive MG, frequently with isolated ocular involve- suspected OMG.78
ment.65 In one cohort of patients with double-negative OMG, After 1 to 2 seconds of RNS, ACh is mobilized from
27% were LRP-4 positive.65 MG antibody assays do not a secondary store, increasing the EPP and resulting in stabi-
routinely include LRP-4 antibodies, so sending this separate lization or slight improvement of the CMAP decrement.76 In
test in seronegative OMG patients may be useful. LRP-4 MG, exercise of muscles also leads to fatigability and more
antibodies are not specific to MG – they have been detected profound decrement of the CMAP amplitude.76 A decremen-
in up to 23% of patients with motor neuron disease,69,70 as tal response to slow RNS can also be seen in other conditions,
well as in cases of LEMS and neuromyelitis optica.65 such as LEMS, motor neuron disease, and other neuropathic
conditions.77 RNS should therefore always be performed in
Edrophonium (“Tensilon”) Testing conjunction with routine nerve conduction studies and
The interpretation of this test requires an objective, measur- electromyography to ensure these other conditions are not
able deficit (most commonly ptosis or ocular misalignment) missed.76 RNS is also technically demanding; any movement
which should improve after IV administration of edropho- of the patient, recording electrodes, or nerve stimulator can
nium (an acetylcholinesterase inhibitor with rapid onset and artifactually create the false impression of a decremental
offset71). False positive results have been reported in motor response (►Fig. 5).77
neuron disease, LEMS, and central lesions including brain- Single-fiber electromyography (SFEMG) is the most sen-
stem glioma and pineal germinoma.72–74 Edrophonium test- sitive test for OMG (abnormal in about 95% of OMG patients if
ing has largely been supplanted by serologic testing given the more than one muscle tested79,80), leading some to argue
small associated risk of bradycardia and bronchospasm; that it should be used in place of RNS in suspected OMG
testing requires cardiac monitoring and atropine available cases.75 However, SFEMG is not available at all centers, is
at the bedside.72 highly operator dependent, and requires close patient coop-
eration.76 Variability in the AP interval between two muscle
Electrodiagnostic Studies fibers in the same motor unit, termed “jitter,” is the key
Given the high specificity of serologic testing, electrodiag- abnormality detected on SFEMG in MG. Abnormal jitter is not
nostic testing is most useful in seronegative cases of sus- specific to NMJ disorders; it is seen in a wide range of nerve
pected OMG. When testing for MG, slow repetitive nerve and muscle disorders, and must be interpreted in the appro-
stimulation (RNS) at 2 to 5 Hz is used (►Fig. 5). Changes in priate clinical context.
the resulting compound muscle action potential (CMAP) Finally, in cases where serologic and electrodiagnostic
with each stimulation are measured.75 Although RNS at testing is nondiagnostic, but significant clinical suspicion for

Seminars in Neurology Vol. 39 No. 6/2019

 Update on Ocular Myasthenia Gravis O’Hare, Doughty 755

OMG remains, an empiric trial of pyridostigmine can be

considered. A symptomatic response to this can help support
a clinical diagnosis of OMG. Serologic testing should also be
repeated in 6 to 12 months, as some patients initially
categorized as seronegative will develop detectable circulat-
ing antibodies on repeated testing.81

Additional Testing
All patients diagnosed with MG (OMG or GMG) should
undergo screening imaging to evaluate for thymoma, typi-
cally with computed tomography (CT) or magnetic reso-
nance imaging (MRI) of the chest.82 Other autoimmune
diseases are frequently comorbid in patients with MG.
Thyroid disease is most common among these, affecting up
to 22% at presentation.83 It is therefore reasonable to obtain
screening thyroid function tests at the time of diagnosis.

Downloaded by: Collections and Technical Services Department. Copyrighted material.

Risk of Conversion to Generalized
Myasthenia
The clinical course over the first 3 years seems to be critical in
determining the long-term manifestations of the disease.84
About 15% of patients initially presenting with ocular symp-
toms will remain purely ocular, while about 85% will general-
ize.2,4,85 The vast majority of this generalization occurs within
the first year of symptoms. One large case series demonstrated
that 87% of generalization occurs in the first year and 94% in the
first 3 years.84 However, a recent case series using a more
restricted definition of OMG—isolated ocular symptoms for
>24 months—still demonstrated a rate of progression to GMG
of 21% at a median of 83 months (range, 24–158 months).56
Risk factors for the development of generalized symptoms
include female gender and seropositivity, with a much greater
risk of generalization in MuSK-positive cases than in AChR-
positive cases.56,86 A retrospective cohort study of 101 OMG
patients (not treated with any form of immunosuppression)
identified that significant predictive factors for generalization
included seropositivity, thymic hyperplasia, and the presence
of comorbidities including other autoimmune diseases.87
Electrodiagnostic evidence of subclinical generalized dis-
ease has been examined as a risk factor for clinical generali-
zation of OMG. One prospective study of 39 OMG patients
within 4 months of symptom onset showed that SFEMG
abnormalities of the extensor digitorum communis (EDC)
Fig. 5 (A) 3-Hz repetitive nerve stimulation in a patient with myasthenia
were common, occurring in 70% of cases. A higher percentage
gravis results in a >10% decrement in amplitude when comparing the first
and fifth compound muscle action potential (CMAP). (B) Following
of the patients with abnormal SFEMG converted to GMG (57
10 seconds of sustained voluntary contraction of the muscle, the test is vs. 18%), although this did not reach statistical significance.88
repeated and less decrement (“repair”) is observed. This phenomenon of Subsequent larger retrospective studies have confirmed the
transient improvement (“postexercise facilitation”) occurs because maxi- high rate of subclinical SFEMG abnormalities in OMG
mal voluntary contraction leads to calcium accumulation in the pre-synaptic
patients (occurring in 68–82% of cases) but have shown no
nerve terminal, facilitating ACh release and thereby counterbalancing the
depletion of ACh vesicles. (C) At a more delayed time point following
predictive value of these findings.89,90
exertion (i.e., 3 minutes after 60 seconds of exercise), the decremental Patients with OMG that do not generalize tend to have a
CMAP response is more pronounced—a phenomenon termed “postexercise benign clinical course, with ocular symptoms typically reach-
exhaustion.” (D) Repetitive stimulation can be technically challenging. ing peak severity within the first 1 to 3 years.2 In a retrospec-
Here, artefactual decrement in sequential CMAPs is demonstrated; this can
tive review of 78 OMG cases, 54 cases remained purely ocular
occur as a result of patient, electrode, or stimulator movement. True
pathophysiologic decrement leads to a characteristic “U-shaped” pattern of
over the follow-up period (mean duration, 8.3 years); of these,
sequential CMAPs, as is seen in panels A and C, whereas artefactual 54% went into remission (defined as the absence of disability,
decrement leads to a more haphazard, random pattern. with or without pharmacologic treatment), 33% showed

Seminars in Neurology Vol. 39 No. 6/2019

756 Update on Ocular Myasthenia Gravis O’Hare, Doughty

clinical improvement, and 13% remained stable.83 However, day.33,101 Pyridostigmine is a purely symptomatic interven-
patients who do not achieve minimal manifestations of disease tion with no effect on the natural history of the disease.101
status do report reduced quality of life, and more severe ocular Unfortunately, only 20 to 40% of OMG patients will exhibit a
symptoms at onset correlate with an increased likelihood of an satisfactory response to pyridostigmine monotherapy.101 Side
unfavorable response to therapy.91 effects range from unpleasant (diarrhea, increased salivation,
urinary urgency) to more dangerous (e.g. bronchospasm in
those with underlying obstructive lung disease).101
Treatment
While OMG symptoms of ptosis and diplopia can be trouble- Corticosteroids
some and even disabling for the patient, they are usually not In patients with persistent symptoms despite pyridostig-
life-threatening. Therefore the role of aggressive immuno- mine, corticosteroids are commonly used as the next line of
modulating therapies is debated in this population. However, therapy, typically prednisone/prednisolone.33 Corticoste-
balanced with this concern is the understanding that a roids exert their therapeutic effect in MG by reducing anti-
significant proportion of OMG patients will develop gener- body production and inhibiting CD4 þ T cell activation.101 Up
alized symptoms, with some data suggesting a possible to 80% of OMG patients will show significant symptomatic
reduction in the rate of generalization with the early use improvement with corticosteroids.101 Data from a retrospec-

Downloaded by: Collections and Technical Services Department. Copyrighted material.

of immunotherapies.83,92–95 Additional recent observational tive cohort study suggests that the rate of achieving complete
data also suggest that early initiation of immunosuppressive remission from ocular symptoms may be higher in those
therapy is associated with a greater likelihood of successful treated with steroids compared to pyridostigmine alone (70
resolution of symptoms. Those treated with corticosteroids vs. 21%). However, at a median dose of 20 mg of prednisone
 steroid-sparing agents within 12 months of symptom daily, this was associated with a relatively high incidence of
onset were twice as likely to show complete resolution of adverse events including new-onset glucose intolerance or
ophthalmoparesis, with a median time to resolution of diabetes in 67%, new or worsening hypertension in 20%,
4 months (compared with 14 months in the delayed treat- weight gain in 42%, and new or worsening osteoporosis in
ment group).96 Randomized data to support a specific treat- 20%, despite appropriate calcium and vitamin-D supplemen-
ment approach are sparse, so a tailored approach based on tation.59 Although hampered by poor accrual of participants,
patient preference, treatment goals, and the anticipated side the recently completed randomized placebo-controlled
effect burden may be most appropriate.97 EPITOME (Efficacy of Prednisone for the Treatment of Ocular
Myasthenia) trial demonstrated that minimal manifestations
Nonpharmacologic Treatment Options status could be reached within 16 weeks for five out of six
In patients with mild symptoms (e.g., incomplete ptosis), prednisone-treated OMG patients in comparison to zero of
starting with a conservative, nonpharmacologic approach is five on placebo, with comparably low rates of mild adverse
reasonable and avoids potential side effects from medica- events.57
tions. For patients with symptomatic diplopia, the most Expert consensus generally favors starting at low doses of
simple and cost-effective intervention is to patch one eye. prednisone and slowly uptitrating (e.g., 5 mg increments
For those resistant to wearing a patch, one occlusive contact every 5 days) until significant symptom resolution is
lens can be used. Alternatively, customized prisms can be seen.33,103,104 The maximum dose of steroids required for
helpful if the degree of misalignment is relatively stable.98 symptom control is typically lower in OMG than in
Strabismus surgery can be performed but typically only in GMG.96,103 The median prednisone dose required for treat-
patients with stable ocular misalignment for at least ment response in the EPITOME trial, for example, was 15 mg
6 months.98,99 In one retrospective series describing nine daily.57 After symptom control is achieved, the dose is
OMG patients managed with strabismus surgery, five tapered over several weeks to the minimum effective
patients achieved single-vision postoperatively, and four dose.33,103,104 For certain patients – those prioritizing rapid
patients required a second surgery.100 Ptosis can be treated symptomatic resolution, and at lower risk of side effects –
with eyelid supports (either crutches inserted into glasses higher starting doses of 0.5 mg/kg/day can be prescribed,
frames or taping of the eyelid),98,101 although there is an followed by a slow taper once symptoms improve.97
associated risk of exposure keratopathy.98 Ptosis surgery can
be performed in cases of longstanding, stable myasthenic Other Immunotherapeutic Agents
ptosis, but there is a high rate of ptosis recurrence with need When corticosteroids are ineffective (as seen in up to 24% of
for repeat surgery. Exposure keratopathy is again a risk.98,102 OMG patients92), when side effects limit their use, or when
contraindications preclude their use entirely, additional
immunosuppressive agents, such as azathioprine, mycophe-
Pharmacologic Treatment Options
nolate mofetil, or methotrexate can be considered. These
Pyridostigmine medications may be introduced when patients require high
Pyridostigmine is an acetylcholinesterase inhibitor which doses of steroids or have difficulty weaning, or they can be
prolongs the duration of action of ACh at the NMJ.103 Dosing started at the same time as steroid therapy (especially in
is started at 30 mg two to four times per day which can be selected patients at higher risk for steroid-induced
increased gradually up to 60 to 90 mg four to five times per complications).

Seminars in Neurology Vol. 39 No. 6/2019

 Update on Ocular Myasthenia Gravis O’Hare, Doughty 757

A placebo-controlled randomized trial has shown that may be considered for AChR Ab seropositive OMG patients
using azathioprine as an adjunct to steroids in generalized who have failed medical therapy.85,103
MG leads to lower maintenance steroid doses, increased
remission rates, lower relapse rates, and decreased side Does Immunomodulatory Therapy Improve Outcomes in
effects.105 Specifically for OMG, azathioprine used alone or OMG?
in combination with prednisone is effective in improving Retrospective data suggest that in addition to treating OMG
symptoms in the majority of patients (91% positive response symptomatically, immunomodulatory therapy may alter the
reported with azathioprine and prednisone combination natural history of the disease and reduce the risk of develop-
therapy in a nonrandomized trial).83 Azathioprine is typical- ing GMG. However, there is a lack of randomized controlled
ly better tolerated than long-term steroid therapy,101 but data to support this theory. The natural tendency of the
careful monitoring is required as some side effects can be disease to improve or remit spontaneously (i.e., with no
life-threatening. Complete blood count and liver function treatment) in a minority of cases should also be noted.118
testing must be followed, as leukopenia, thrombocytopenia, Multiple retrospective case series have noted that OMG
and liver function test abnormalities can result.101 Those patients treated with immunosuppression (including cortico-
with mutations in the thiopurine methyltransferase gene are steroids, azathioprine, and thymectomy) are significantly less
at increased risk of azathioprine-induced myelosuppres- likely to develop GMG.83,92–95 In one illustrative retrospective

Downloaded by: Collections and Technical Services Department. Copyrighted material.

sion.103 In addition, azathioprine confers a three-fold in- series by Kupersmith, outcomes of OMG patients treated with
creased risk of developing nonmelanomatous skin steroids were compared to those treated with pyridostigmine
cancers.106 Mycophenolate mofetil has also been shown to alone.92 All patients with symptomatic extraocular symptoms
be safe and well-tolerated as a steroid-sparing agent in not responsive to pyridostigmine alone were offered predni-
OMG,107 but a randomized placebo-controlled trial in GMG sone (in the absence of any contraindications to steroid use),
showed no benefit.108 Gastrointestinal complaints are the and the majority of patients treated with steroids started them
most prominent side-effect of mycophenolate,107 and mye- within 6 months of symptom onset. GMG developed in 50% of
losuppression or hepatotoxicity are rarely seen.103 Metho- the untreated group (mean conversion time 0.22 years from
trexate has been compared to azathioprine in GMG and OMG symptom onset; range, 0.1–0.8 years) versus 14% of the
shown to have similar efficacy and tolerability.109 treated group (mean conversion time, 5.8 years; range,
Although commonly considered as rescue therapy in 2.5–10.5 years). This suggests that steroid treatment may
generalized disease, IV immunoglobulins and plasmaphere- impact the natural history of OMG by both delaying the
sis are typically not required for OMG. Eculizumab, a biologic conversion to GMG and reducing the incidence of conversion.
agent targeting the terminal complement protein C5 is now However, the retrospective nature of the data supporting
available for use in AChR positive GMG. However, the lack of this theory makes it hard to draw firm conclusions, and the
data in OMG and seronegative patients110 preclude its use at role of immunosuppression in preventing secondary gener-
present in the OMG population. alization remains controversial.119 The difficulty of inter-
preting nonblinded retrospective studies to assess treatment
Thymectomy effect is illustrated by the case series reported by Galassi et al
Thymectomy is required for those patients with thymoma, but which indicated that 0% of untreated OMG patients
has also been shown to result in better clinical outcomes and developed secondary generalization versus 24.8% of treated
reduced corticosteroid requirements in nonthymomatous patients.56 This finding can presumably be attributed to a
AChR seropositive GMG.111,112 Thymectomy may be particu- lower risk of disease progression in mild OMG cases that
larly effective in those early in their disease course with were not symptomatic enough to warrant treatment, rather
significant follicular hyperplasia.8 The role of thymectomy in than any causal effect from the treatment itself.
nonthymomatous OMG remains highly controversial,113,114
given the potential for significant morbidity associated with
Conclusion
this invasive procedure. Some retrospective data suggest a
high rate of OMG remission following thymectomy (defined OMG is an autoimmune disease affecting the neuromuscular
variably as being asymptomatic either off medications or on junction of the extraocular muscles, resulting in a characteristic
low doses of immunosuppression), particularly when surgery presentation with fluctuating, fatigable diplopia and ptosis.
is performed within the first 12 months of symptoms.115,116 A Supportive diagnostic tests, including serologic and electro-
2017 meta-analysis examining outcomes of thymectomy spe- diagnostic studies, can help confirm the diagnosis; however,
cifically in nonthymomatous OMG concluded that complete the sensitivity of these tests is much lower when myasthenia is
stable remission (defined as being asymptomatic for >1 year of restricted to the ocular form. Therefore, the diagnosis of OMG
all pharmacologic treatment5) was achieved in about 50% of frequently remains purely clinical. The physical examination of
cases.114 Significant heterogeneity of the data included in this patients with OMG is accordingly of fundamental importance,
meta-analysis precludes definitive conclusions. In addition, both in establishing a positive diagnosis and ruling out potential
the AChR Ab seropositivity status is lacking in the majority of mimics. Treatment of OMG encompasses a range of nonphar-
included reports, and data on complication rates was not macologic, symptomatic, and immunomodulatory options, and
reported in the meta-analysis.117 Despite the lack of random- is best tailored to the individual patient based on severity of
ized prospective data supporting this practice, thymectomy symptoms, treatment goals, and comorbidities.

Seminars in Neurology Vol. 39 No. 6/2019

758 Update on Ocular Myasthenia Gravis O’Hare, Doughty

References 23 Kee HJ, Yang HK, Hwang J-M, et al. Evaluation and validation of
1 Jayam Trouth A, Dabi A, Solieman N, Kurukumbi M, Kalyanam J. sustained upgaze combined with the ice-pack test for ocular
Myasthenia gravis: a review. Autoimmune Dis 2012;2012:874680 myasthenia gravis in asians. Neuromuscul Disord 2019;29(04):
2 Grob D, Brunner N, Namba T, Pagala M. Lifetime course of 296–301
myasthenia gravis. Muscle Nerve 2008;37(02):141–149 24 Osher RH, Glaser JS. Myasthenic sustained gaze fatigue. Am J
3 Bever CT Jr., Aquino AV, Penn AS, Lovelace RE, Rowland LP. Ophthalmol 1980;89(03):443–445
Prognosis of ocular myasthenia. Ann Neurol 1983;14(05):516–519 25 Pourmand R. Myasthenia gravis. Dis Mon 1997;43(02):65–109
4 Evoli A, Tonali P, Bartoccioni E, Lo Monaco M. Ocular myasthenia: 26 Averbuch-Heller L, Poonyathalang A, von Maydell RD, Remler BF.
diagnostic and therapeutic problems. Acta Neurol Scand 1988; Hering’s law for eyelids: still valid. Neurology 1995;45(09):
77(01):31–35 1781–1783
5 Jaretzki A III, Barohn RJ, Ernstoff RM, et al; Task Force of the 27 Pelak VS, Galetta SL. Ocular myasthenia gravis. Curr Treat
Medical Scientific Advisory Board of the Myasthenia Gravis Options Neurol 2001;3(04):367–376
Foundation of America. Myasthenia gravis: recommendations 28 Osher RH, Griggs RC. Orbicularis fatigue: the ‘peek’ sign of
for clinical research standards. Neurology 2000;55(01):16–23 myasthenia gravis. Arch Ophthalmol 1979;97(04):677–679
6 Ruff RL, Lisak RP. Nature and action of antibodies in myasthenia 29 Lepore FE, Sanborn GE, Slevin JT. Pupillary dysfunction in
gravis. Neurol Clin 2018;36(02):275–291 myasthenia gravis. Ann Neurol 1979;6(01):29–33
7 Thanvi BR, Lo TCN. Update on myasthenia gravis. Postgrad Med J 30 Yamazaki A, Ishikawa S. Abnormal pupillary responses in myas-
2004;80(950):690–700 thenia gravis. A pupillographic study. Br J Ophthalmol 1976;60
8 Berrih-Aknin S, Le Panse R. Myasthenia gravis: a comprehensive (08):575–580

Downloaded by: Collections and Technical Services Department. Copyrighted material.

review of immune dysregulation and etiological mechanisms. 31 Dutton GN, Garson JA, Richardson RB. Pupillary fatigue in
J Autoimmun 2014;52:90–100 myasthenia gravis. Trans Ophthalmol Soc U K 1982;102(Pt
9 Sommer N, Tackenberg B, Hohlfeld R. Chapter 5 The immuno- 4):510–513
pathogenesis of myasthenia gravis. In: Engel AG, ed. Handbook 32 Cooper J, Pollak GJ, Ciuffreda KJ, Kruger P, Feldman J. Accommo-
of Clinical Neurology. New York, NY: Elsevier; 2008:169–212 dative and vergence findings in ocular myasthenia: a case
10 Mori S, Kubo S, Akiyoshi T, et al. Antibodies against muscle- analysis. J Neuroophthalmol 2000;20(01):5–11
specific kinase impair both presynaptic and postsynaptic func- 33 Sussman J, Farrugia ME, Maddison P, Hill M, Leite MI, Hilton-
tions in a murine model of myasthenia gravis. Am J Pathol 2012; Jones D. Myasthenia gravis: Association of British Neurologists’
180(02):798–810[published Online First: 2011/12/07] management guidelines. Pract Neurol 2015;15(03):199–206
11 Shen C, Lu Y, Zhang B, et al. Antibodies against low-density 34 Andonopoulos AP, Terzis E, Tsibri E, Papasteriades CA, Papape-
lipoprotein receptor-related protein 4 induce myasthenia gravis. tropoulos T. D-penicillamine induced myasthenia gravis in
J Clin Invest 2013;123(12):5190–5202 rheumatoid arthritis: an unpredictable common occurrence?
12 Rodríguez Cruz PM, Al-Hajjar M, Huda S, et al. Clinical features Clin Rheumatol 1994;13(04):586–588
and diagnostic usefulness of antibodies to clustered acetylcho- 35 Drosos AA, Christou L, Galanopoulou V, Tzioufas AG, Tsiakou EK.
line receptors in the diagnosis of seronegative myasthenia D-penicillamine induced myasthenia gravis: clinical, serological
gravis. JAMA Neurol 2015;72(06):642–649 and genetic findings. Clin Exp Rheumatol 1993;11(04):387–391
13 Jacob S, Viegas S, Leite MI, et al. Presence and pathogenic 36 Suzuki S, Ishikawa N, Konoeda F, et al. Nivolumab-related
relevance of antibodies to clustered acetylcholine receptor in myasthenia gravis with myositis and myocarditis in Japan.
ocular and generalized myasthenia gravis. Arch Neurol 2012;69 Neurology 2017;89(11):1127–1134
(08):994–1001 37 Kao JC, Brickshawana A, Liewluck T. Neuromuscular complica-
14 Cortés-Vicente E, Gallardo E, Martínez MÁ, et al. Clinical character- tions of programmed cell death-1 (PD-1) inhibitors. Curr Neurol
istics of patients with double-seronegative myasthenia gravis and Neurosci Rep 2018;18(10):63
antibodies to cortactin. JAMA Neurol 2016;73(09):1099–1104 38 Makarious D, Horwood K, Coward JIG. Myasthenia gravis: An
15 Gallardo E, Martínez-Hernández E, Titulaer MJ, et al. Cortactin emerging toxicity of immune checkpoint inhibitors. Eur J Cancer
autoantibodies in myasthenia gravis. Autoimmun Rev 2014;13 2017;82:128–136
(10):1003–1007 39 Cooper DS, Meriggioli MN, Bonomi PD, Malik R. Severe exacer-
16 Marx A, Müller-Hermelink HK, Ströbel P. The role of thymomas in bation of myasthenia gravis associated with checkpoint inhibitor
the development of myasthenia gravis. Ann N Y Acad Sci 2003; immunotherapy. J Neuromuscul Dis 2017;4(02):169–173
998(01):223–236 40 Lau KH, Kumar A, Yang IH, Nowak RJ. Exacerbation of myasthenia
17 Marx A, Pfister F, Schalke B, Saruhan-Direskeneli G, Melms A, gravis in a patient with melanoma treated with pembrolizumab.
Ströbel P. The different roles of the thymus in the pathogenesis of Muscle Nerve 2016;54(01):157–161
the various myasthenia gravis subtypes. Autoimmun Rev 2013; 41 Shah M, Tayar JH, Adbel-Wahab N, Suarez-Almazore ME. Myosi-
12(09):875–884 tis as an adverse event of immune checkpoint blockade for
18 Romi F, Skeie GO, Aarli JA, Gilhus NE. Muscle autoantibodies in cancer therapy. Semin Arthritis Rheum 2018;48(04):736–740
subgroups of myasthenia gravis patients. J Neurol 2000;247(05): 42 Brahmer JR, Lacchetti C, Schneider BJ, et al; National Comprehen-
369–375 sive Cancer Network. Management of immune-related adverse
19 Kaminski HJ, Li Z, Richmonds C, Ruff RL, Kusner L. Susceptibility events in patients treated with immune checkpoint inhibitor
of ocular tissues to autoimmune diseases. Ann N Y Acad Sci 2003; therapy: American Society of Clinical Oncology Clinical Practice
998:362–374 Guideline. J Clin Oncol 2018;36(17):1714–1768
20 Kaminski HJ, Maas E, Spiegel P, Ruff RL. Why are eye muscles 43 Sethi KD, Rivner MH, Swift TR. Ice pack test for myasthenia
frequently involved in myasthenia gravis? Neurology 1990;40 gravis. Neurology 1987;37(08):1383–1385
(11):1663–1669 44 Kubis KC, Danesh-Meyer HV, Savino PJ, Sergott RC. The ice test
21 Carr AS, Cardwell CR, McCarron PO, McConville J. A systematic versus the rest test in myasthenia gravis. Ophthalmology 2000;
review of population based epidemiological studies in myasthenia 107(11):1995–1998
gravis. BMC Neurol 2010;10:46–46. Doi: 10.1186/1471-2377-10-46 45 Benatar M. A systematic review of diagnostic studies in myas-
22 Peragallo JH, Bitrian E, Kupersmith MJ, et al. Relationship between thenia gravis. Neuromuscul Disord 2006;16(07):459–467
age, gender, and race in patients presenting with myasthenia 46 Fakiri MO, Tavy DL, Hama-Amin AD, Wirtz PW. Accuracy of the
gravis with only ocular manifestations. J Neuroophthalmol ice test in the diagnosis of myasthenia gravis in patients with
2016;36(01):29–32 ptosis. Muscle Nerve 2013;48(06):902–904

Seminars in Neurology Vol. 39 No. 6/2019

 Update on Ocular Myasthenia Gravis O’Hare, Doughty 759

47 Odel JG, Winterkorn JM, Behrens MM. The sleep test for myas- 70 Rivner MH, Liu S, Quarles B, et al. Agrin and low-density
thenia gravis. A safe alternative to Tensilon. J Clin Neurooph- lipoprotein-related receptor protein 4 antibodies in amyotrophic
thalmol 1991;11(04):288–292 lateral sclerosis patients. Muscle Nerve 2017;55(03):430–432
48 Cogan DG. Myasthenia gravis: a review of the disease and a 71 Meriggioli MN, Sanders DB. Myasthenia gravis: diagnosis. Semin
description of lid twitch as a characteristic sign. Arch Ophthal- Neurol 2004;24(01):31–39
mol 1965;74:217–221 72 Seybold ME. The office Tensilon test for ocular myasthenia
49 Van Stavern GP, Bhatt A, Haviland J, Black EH. A prospective study gravis. Arch Neurol 1986;43(08):842–843
assessing the utility of Cogan’s lid twitch sign in patients with isolated 73 Dirr LY, Donofrio PD, Patton JF, Troost BT. A false-positive
unilateral or bilateral ptosis. J Neurol Sci 2007;256(1,2):84–85 edrophonium test in a patient with a brainstem glioma. Neurol-
50 Moorthy G, Behrens MM, Drachman DB, et al. Ocular pseudo- ogy 1989;39(06):865–867
myasthenia or ocular myasthenia ‘plus’: a warning to clinicians. 74 Fierro B, Croce G, Filosto L, Carbone N, Lupo I. Ocular pseudo-
Neurology 1989;39(09):1150–1154 myasthenia: report of a case with a pineal region tumor. Ital J
51 Ragge NK, Hoyt WF. Midbrain myasthenia: fatigable ptosis, ‘lid Neurol Sci 1991;12(06):593–596
twitch’ sign, and ophthalmoparesis from a dorsal midbrain 75 AAEM Quality Assurance Committee. American Association of
glioma. Neurology 1992;42(04):917–919 Electrodiagnostic Medicine. Practice parameter for repetitive nerve
52 Apinyawasisuk S, Zhou X, Tian JJ, Garcia GA, Karanjia R, Sadun AA. stimulation and single fiber EMG evaluation of adults with sus-
Validity of forced eyelid closure test: a novel clinical screening test for pected myasthenia gravis or Lambert-Eaton myasthenic syndrome:
ocular myasthenia gravis. J Neuroophthalmol 2017;37(03):253–257 summary statement. Muscle Nerve 2001;24(09):1236–1238
53 Howard FM Jr, Lennon VA, Finley J, Matsumoto J, Elveback LR. 76 Preston DC, Shapiro BE. Electromyography and Neuromuscular

Downloaded by: Collections and Technical Services Department. Copyrighted material.

Clinical correlations of antibodies that bind, block, or modulate Disorders: Clinical-Electrophysiologic Correlations. 3rd ed. New
human acetylcholine receptors in myasthenia gravis. Ann N Y York, NY: Saunders; 2012
Acad Sci 1987;505:526–538 77 Howard JF Jr. Electrodiagnosis of disorders of neuromuscular
54 Sommer N, Melms A, Weller M, Dichgans J. Ocular myasthenia transmission. Phys Med Rehabil Clin N Am 2013;24(01):
gravis. A critical review of clinical and pathophysiological 169–192
aspects. Doc Ophthalmol 1993;84(04):309–333 78 Zambelis T, Kokotis P, Karandreas N. Repetitive nerve stimula-
55 Vaphiades MS, Bhatti MT, Lesser RL. Ocular myasthenia gravis. tion of facial and hypothenar muscles: relative sensitivity in
Curr Opin Ophthalmol 2012;23(06):537–542 different myasthenia gravis subgroups. Eur Neurol 2011;65(04):
56 Galassi G, Mazzoli M, Ariatti A, Kaleci S, Valzania F, Nichelli PF. 203–207
Antibody profile may predict outcome in ocular myasthenia 79 Sanders DB, Howard JF Jr. AAEE minimonograph #25: Single-
gravis. Acta Neurol Belg 2018;118(03):435–443 fiber electromyography in myasthenia gravis. Muscle Nerve
57 Benatar M, Mcdermott MP, Sanders DB, et al; Muscle Study 1986;9(09):809–819
Group (MSG). Efficacy of prednisone for the treatment of ocular 80 Padua L, Caliandro P, Di Iasi G, Pazzaglia C, Ciaraffa F, Evoli A.
myasthenia (EPITOME): A randomized, controlled trial. Muscle Reliability of SFEMG in diagnosing myasthenia gravis: sensitivity
Nerve 2016;53(03):363–369 and specificity calculated on 100 prospective cases. Clin Neuro-
58 Evoli A, Alboini PE, Iorio R, Damato V, Bartoccioni E. Pattern of physiol 2014;125(06):1270–1273
ocular involvement in myasthenia gravis with MuSK antibodies. 81 Vincent A, Newsom-Davis J. Acetylcholine receptor antibody as a
J Neurol Neurosurg Psychiatry 2017;88(09):761–763 diagnostic test for myasthenia gravis: results in 153 validated
59 Bhanushali MJ, Wuu J, Benatar M. Treatment of ocular symptoms cases and 2967 diagnostic assays. J Neurol Neurosurg Psychiatry
in myasthenia gravis. Neurology 2008;71(17):1335–1341 1985;48(12):1246–1252
60 Sanders DB, Andrews PI, Howard JF, Massey JM. Seronegative 82 Priola AM, Priola SM. Imaging of thymus in myasthenia gravis:
myasthenia gravis. Neurology 1997;48(Suppl 5):40S–45S from thymic hyperplasia to thymic tumor. Clin Radiol 2014;69
61 McConville J, Farrugia ME, Beeson D, et al. Detection and (05):e230–e245
characterization of MuSK antibodies in seronegative myasthenia 83 Sommer N, Sigg B, Melms A, et al. Ocular myasthenia gravis:
gravis. Ann Neurol 2004;55(04):580–584 response to long-term immunosuppressive treatment. J Neurol
62 Evoli A, Tonali PA, Padua L, et al. Clinical correlates with anti- Neurosurg Psychiatry 1997;62(02):156–162
MuSK antibodies in generalized seronegative myasthenia gravis. 84 Grob D, Arsura EL, Brunner NG, Namba T. The course of myas-
Brain 2003;126(Pt 10):2304–2311 thenia gravis and therapies affecting outcome. Ann N Y Acad Sci
63 Guptill JT, Sanders DB, Evoli A. Anti-MuSK antibody myasthenia 1987;505:472–499
gravis: clinical findings and response to treatment in two large 85 Gilhus NE. Myasthenia Gravis. N Engl J Med 2016;375(26):
cohorts. Muscle Nerve 2011;44(01):36–40 2570–2581
64 Sanders DB, El-Salem K, Massey JM, McConville J, Vincent A. 86 Mazzoli M, Ariatti A, Valzania F, et al. Factors affecting outcome
Clinical aspects of MuSK antibody positive seronegative MG. in ocular myasthenia gravis. Int J Neurosci 2018;128(01):15–24
Neurology 2003;60(12):1978–1980 87 Wong SH, Petrie A, Plant GT. Ocular myasthenia gravis: toward a
65 Zisimopoulou P, Evangelakou P, Tzartos J, et al. A comprehensive risk of generalization score and sample size calculation for a
analysis of the epidemiology and clinical characteristics of anti- randomized controlled trial of disease modification. J Neuro-
LRP4 in myasthenia gravis. J Autoimmun 2014;52:139–145 ophthalmol 2016;36(03):252–258
66 Higuchi O, Hamuro J, Motomura M, Yamanashi Y. Autoantibodies 88 Weinberg DH, Rizzo JF III, Hayes MT, Kneeland MD, Kelly JJ Jr.
to low-density lipoprotein receptor-related protein 4 in myas- Ocular myasthenia gravis: predictive value of single-fiber elec-
thenia gravis. Ann Neurol 2011;69(02):418–422 tromyography. Muscle Nerve 1999;22(09):1222–1227
67 Zhang B, Tzartos JS, Belimezi M, et al. Autoantibodies to lipopro- 89 Rostedt A, Saders LL, Edards LJ, Massey JM, Sanders DB, Stålberg
tein-related protein 4 in patients with double-seronegative EV. Predictive value of single-fiber electromyography in the
myasthenia gravis. Arch Neurol 2012;69(04):445–451 extensor digitorum communis muscle of patients with ocular
68 Pevzner A, Schoser B, Peters K, et al. Anti-LRP4 autoantibodies in myasthenia gravis: a retrospective study. J Clin Neuromuscul Dis
AChR- and MuSK-antibody-negative myasthenia gravis. J Neurol 2000;2(01):6–9
2012;259(03):427–435 90 Guan YZ, Cui LY, Liu MS, Niu JW. Single-fiber electromyography
69 Tzartos JS, Zisimopoulou P, Rentzos M, et al. LRP4 antibodies in in the extensor digitorum communis for the predictive prognosis
serum and CSF from amyotrophic lateral sclerosis patients. Ann of ocular myasthenia gravis: a retrospective study of 102 cases.
Clin Transl Neurol 2014;1(02):80–87 Chin Med J (Engl) 2015;128(20):2783–2786

Seminars in Neurology Vol. 39 No. 6/2019

760 Update on Ocular Myasthenia Gravis O’Hare, Doughty

91 Suzuki S, Murai H, Imai T, et al. Quality of life in purely ocular 111 Wolfe GI, Kaminski HJ, Aban IB, et al; MGTX Study Group. Ran-
myasthenia in Japan. BMC Neurol 2014;14(01):142 domized trial of thymectomy in myasthenia gravis. N Engl J Med
92 Kupersmith MJ. Ocular myasthenia gravis: treatment successes 2016;375(06):511–522 (Erratum in: Randomized Trial of Thymec-
and failures in patients with long-term follow-up. J Neurol 2009; tomy in Myasthenia Gravis. [N Engl J Med 2017])
256(08):1314–1320 112 Wolfe GI, Kaminski HJ, Aban IB, et al; MGTX Study Group. Long-
93 Mee J, Paine M, Byrne E, King J, Reardon K, O’Day J. Immunotherapy term effect of thymectomy plus prednisone versus prednisone
of ocular myasthenia gravis reduces conversion to generalized alone in patients with non-thymomatous myasthenia gravis: 2-
myasthenia gravis. J Neuroophthalmol 2003;23(04):251–255 year extension of the MGTX randomised trial. Lancet Neurol
94 Monsul NT, Patwa HS, Knorr AM, Lesser RL, Goldstein JM. The 2019;18(03):259–268
effect of prednisone on the progression from ocular to 113 Gilbert ME, De Sousa EA, Savino PJ. Ocular Myasthenia Gravis
generalized myasthenia gravis. J Neurol Sci 2004;217(02): treatment: the case against prednisone therapy and thymecto-
131–133 my. Arch Neurol 2007;64(12):1790–1792
95 Kupersmith MJ, Moster M, Bhuiyan S, Warren F, Weinberg H. 114 Chavis PS, Stickler DE, Walker A. Immunosuppressive or surgical
Beneficial effects of corticosteroids on ocular myasthenia gravis. treatment for ocular myasthenia gravis. Arch Neurol 2007;64
Arch Neurol 1996;53(08):802–804 (12):1792–1794
96 Europa TA, Nel M, Heckmann JM. Myasthenic ophthalmoparesis: 115 Roberts PF, Venuta F, Rendina E, et al. Thymectomy in the
Time To resolution after initiating immune therapies. Muscle treatment of ocular myasthenia gravis. J Thorac Cardiovasc
Nerve 2018;58(04):542–549 Surg 2001;122(03):562–568
97 Guidon AC, Hobson-Webb LD. On the double: early immuno- 116 Liu Z, Feng H, Yeung SC, et al. Extended transsternal thymectomy

Downloaded by: Collections and Technical Services Department. Copyrighted material.

therapy speeds recovery of ocular myasthenic weakness. Muscle for the treatment of ocular myasthenia gravis. Ann Thorac Surg
Nerve 2018;58(06):743–744 2011;92(06):1993–1999
98 Haines SR, Thurtell MJ. Treatment of ocular myasthenia gravis. 117 Zhu K, Li J, Huang X, et al. Thymectomy is a beneficial therapy for
Curr Treat Options Neurol 2012;14(01):103–112 patients with non-thymomatous ocular myasthenia gravis: a
99 Morris OC, O’Day J. Strabismus surgery in the management of systematic review and meta-analysis. Neurol Sci 2017;38(10):
diplopia caused by myasthenia gravis. Br J Ophthalmol 2004;88 1753–1760
(06):832 118 Oosterhuis HJ. The natural course of myasthenia gravis: a long
100 Peragallo JH, Velez FG, Demer JL, Pineles SL. Long-term follow-up term follow up study. J Neurol Neurosurg Psychiatry 1989;52
of strabismus surgery for patients with ocular myasthenia (10):1121–1127
gravis. J Neuroophthalmol 2013;33(01):40–44 119 Wong SH, Plant GT, Cornblath W. Does treatment of ocular
101 Evoli A, Batocchi AP, Minisci C, Di Schino C, Tonali P. Therapeutic myasthenia gravis with early immunosuppressive therapy
options in ocular myasthenia gravis. Neuromuscul Disord 2001; prevent secondarily generalization and should it be offered to
11(02):208–216 all such patients? J Neuroophthalmol 2016;36(01):98–102
102 Carter SR, Meecham WJ, Seiff SR. Silicone frontalis slings for the 120 Bartley GB, Gorman CA. Diagnostic criteria for Graves’ ophthalm-
correction of blepharoptosis: indications and efficacy. Ophthal- opathy. Am J Ophthalmol 1995;119(06):792–795
mology 1996;103(04):623–630 121 Richardson C, Smith T, Schaefer A, Turnbull D, Griffiths P. Ocular
103 Kerty E, Elsais A, Argov Z, Evoli A, Gilhus NE. EFNS/ENS Guide- motility findings in chronic progressive external ophthalmople-
lines for the treatment of ocular myasthenia. Eur J Neurol 2014; gia. Eye (Lond) 2005;19(03):258–263
21(05):687–693 122 Butler IJ, Gadoth N. Kearns-Sayre syndrome. A review of a
104 Sanders DB, Wolfe GI, Benatar M, et al. International consensus multisystem disorder of children and young adults. Arch Intern
guidance for management of myasthenia gravis: executive sum- Med 1976;136(11):1290–1293
mary. Neurology 2016;87(04):419–425 123 Emery AE. The muscular dystrophies. Lancet 2002;359(9307):
105 Palace J, Newsom-Davis J, Lecky B; Myasthenia Gravis Study 687–695
Group. A randomized double-blind trial of prednisolone alone or 124 Machuca-Tzili L, Brook D, Hilton-Jones D. Clinical and molecular
with azathioprine in myasthenia gravis. Neurology 1998;50(06): aspects of the myotonic dystrophies: a review. Muscle Nerve
1778–1783 2005;32(01):1–18
106 Pedersen EG, Pottegård A, Hallas J, et al. Risk of non-melanoma 125 Garg N, Yiannikas C, Hardy TA, et al. Late presentations of
skin cancer in myasthenia patients treated with azathioprine. congenital myasthenic syndromes: How many do we miss?
Eur J Neurol 2014;21(03):454–458 Muscle Nerve 2016;54(04):721–727
107 Chan JW. Mycophenolate mofetil for ocular myasthenia. J Neurol 126 North KN, Wang CH, Clarke N, et al; International Standard of Care
2008;255(04):510–513 Committee for Congenital Myopathies. Approach to the diagnosis of
108 Sanders DB, Hart IK, Mantegazza R, et al. An international, phase congenital myopathies. Neuromuscul Disord 2014;24(02):97–116
III, randomized trial of mycophenolate mofetil in myasthenia 127 Evans BJ. Optometric prescribing in decompensated heteropho-
gravis. Neurology 2008;71(06):400–406 ria. Optometry in Practice 2008;9:63–78
109 Heckmann JM, Rawoot A, Bateman K, Renison R, Badri M. A 128 Arnoldi K, Reynolds JD. A review of convergence insufficiency:
single-blinded trial of methotrexate versus azathioprine as what are we really accomplishing with exercises? Am Orthopt J
steroid-sparing agents in generalized myasthenia gravis. BMC 2007;57:123–130
Neurol 2011;11:97 129 Engel AG. Morphologic and immunopathologic findings in
110 Gilhus NE. Eculizumab: a treatment option for myasthenia myasthenia gravis and in congenital myasthenic syndromes.
gravis? Lancet Neurol 2017;16(12):947–948 J Neurol Neurosurg Psychiatry 1980;43(07):577–589

Seminars in Neurology Vol. 39 No. 6/2019