Ocular myasthenia gravis

INTRODUCTION — Myasthenia gravis (MG) is an autoimmune disorder characterized by weakness and fatigability of skeletal muscles. Muscle weakness due to dysfunction of the neuromuscular junction (myasthenia) is usually an acquired disorder, and the vast majority of patients who develop generalized myasthenia in adolescence or adulthood have autoantibodies that play a pathogenetically important role. Such antibody-mediated disease is referred to as MG.

When the clinical manifestations of MG are isolated to the levator palpebrae superioris, orbicularis oculi, and extraocular muscles affecting eye movement, it is referred to as ocular MG (OMG). Over one-half of all patients with MG initially present with isolated ptosis, diplopia, or both, and no signs or symptoms of weakness elsewhere [1-4]. Approximately 15 percent of all patients with MG have isolated OMG as the only manifestation of their disease. Although the disease pathology at the neuromuscular junction is not known to be distinct from generalized MG (GMG), there are several important and unique differences between OMG and GMG with regard to diagnosis and treatment.

This topic will discuss OMG. Other aspects of MG are discussed separately. (See "Pathogenesis of myasthenia gravis" and "Clinical manifestations of myasthenia gravis" and "Differential diagnosis of myasthenia gravis" and "Diagnosis of myasthenia gravis" and "Overview of the treatment of myasthenia gravis".)

PATHOPHYSIOLOGY — The pathophysiology of OMG is believed to be identical to that for generalized MG (GMG). (See "Pathogenesis of myasthenia gravis".)

It is uncertain why ocular muscles are frequently involved in myasthenia and why the disease stays localized to the extraocular muscles in 15 percent of cases. It has been proposed that subtle alterations in the function of extraocular muscles are more likely to produce symptoms there than in limb muscles, and that the levator palpebrae superioris, under constant activation during eye opening, may be especially susceptible to fatigue. Other factors may contribute to this phenomenon [5-7]:

●Patients with OMG are more likely to be seronegative for acetylcholine receptor antibodies (AChR-Ab) than patients with GMG.

●Extraocular muscles (but not the levator palpebrae superioris) are unique in their expression of fetal acetylcholine receptors at the neuromuscular junction; however, there is no evidence for specific immunologic targeting of this receptor.

●The junctional folds of muscle endplates are sparse in the extraocular and levator muscles, perhaps producing a lower safety factor for neuromuscular transmission.

●Complement regulatory genes are expressed differently in extraocular muscles, perhaps reducing protective mechanisms to complement-mediated tissue injury.

CLINICAL FEATURES — OMG is characterized by ptosis and ophthalmoparesis (weakness of the muscles controlling eye movement) and extraocular muscle weakness. Some patients also have mild orbicularis oculi weakness and thus have difficulty closing the eye. This triad (ptosis, ophthalmoparesis, and orbicularis oculi weakness) should prompt an evaluation for MG.

As with generalized MG (GMG), signs and symptoms of OMG are characterized by fluctuating, fatigable weakness. Most patients note symptoms that worsen as the day progresses or with tasks such as driving. Patients may report that they have mild or no symptoms upon awakening. The examination may elicit signs of fatigable levator and extraocular muscle weakness.

Ptosis — Ptosis is often unilateral or asymmetric on presentation. A historical pattern of ptosis alternating from one side to the other is nearly always a sign of OMG. Old photos may help to determine if ptosis is new or longstanding. Measurements of the eyelid position and levator palpebrae superioris function using standard methods help to identify and quantify weakness as well as fluctuations associated with fatigue. (See "Overview of ptosis".)

Examination for eyelid fatigue — Fluctuation present during the examination suggests fatigability of the levator muscle. Fatigability can also be elicited with the following maneuvers:

●Sustained upgaze. The patient is instructed to maintain upgaze for one to two minutes. Enhancement of ptosis during prolonged upgaze or upon return to primary gaze suggests fatigability.

●Shifting from downgaze to primary gaze. The patient is asked to sustain downgaze briefly and then make a fast conjugate eye movement (saccade) to primary gaze. An affected eyelid will quickly rise and then fall (by as little as 1 mm or more), such that the lid appears to twitch (Cogan lid twitch sign) [8].

●Rest test. Improvement in baseline ptosis after a two- to five-minute period of restful lid closure suggests fatigability.

While each of these signs suggests fatigable levator function, none are specific for OMG.

Ice pack test — In patients with myasthenia, ptosis typically improves after the application of ice to the upper lid. The sensitivity of this test is approximately 80 percent in those with prominent ptosis [9,10]. The predictive value of this test has not been established. The test is not diagnostic but can help to raise suspicion for myasthenia.

To perform the test, a bag is filled with ice and placed on the closed eyelid for one minute. The ice is then removed and the extent of ptosis (and ocular misalignment) is assessed immediately; the duration of improvement is short (less than one minute). The change in muscle strength is based on the physiologic principle of improved neuromuscular transmission at lower muscle temperature.

The utility of the ice pack test to assess ophthalmoparesis has not been investigated in a controlled manner; some authors have noted detectable changes in ocular alignment [11].

Diplopia — Binocular diplopia (diplopia that is present only with both eyes open) is a prominent feature of OMG when ophthalmoparesis (weakness of the eye muscles controlling eye movement) is present. At presentation, OMG may involve individual or multiple extraocular motor muscles in one or both eyes. For patients with medial or lateral rectus muscle involvement, diplopia will be binocular and horizontal. If the superior or inferior recti or the oblique muscles are involved, there will be a vertical or diagonal component to the diplopia.

Any pattern of ocular misalignment may be observed in OMG, including those that are caused by isolated oculomotor cranial nerve palsies and central nervous system brainstem pathway lesions such as internuclear ophthalmoplegia.

The cardinal positions of gaze should be tested and observed for paresis. (See "Overview of diplopia", section on 'Examination'.)

As with ptosis, OMG can result in fatigable ocular motility. Fluctuation in either the degree of diplopia or the direction of gaze that elicits the diplopia suggests fatigable ophthalmoparesis. Prolonged or sustained gaze in the direction of action of an involved muscle may result in fatigue of the muscle with increasing paresis [12].

Similarly, repeated measurements using a Maddox rod or red glass during or after sustained gaze may reveal increasing degrees of strabismus (ocular misalignment). For the red glass test, a translucent red glass is placed over one eye of the patient while the patient is looking at a light. It is best to use a light source with a focused beam. The patient with ophthalmoparesis will see one red light and one white light, and the relationship of the two lights allows the examiner to determine the type of ocular misalignment that is causing the double vision. For example, when the red glass is placed over the right eye and the patient reports that the red light is to the right of the white light, then the eyes are directed toward each other, as noted with lateral rectus weakness of either eye.

An alternative to the red glass is a Maddox rod, in which the red glass is replaced by parallel translucent cylinder lines. The Maddox rod is more commonly used than the red glass. With the same technique as described in the red glass test, a patient places the Maddox rod over one eye and looks at a light source. The patient will see a red line with the eye with the Maddox rod over it and a white light with the fellow eye [13]. The findings help identify the misalignment but cannot identify whether both eyes or one eye is weak. However, after testing in multiple positions of gaze (ie, up, right, left, down), the examiner can pinpoint the weak extraocular muscles by noting the pattern of misalignment. (See "Overview of diplopia", section on 'Examination'.)

Other findings — Difficulty closing the eye fully (lagophthalmus) is rare in OMG. However, some weakness of the orbicularis oculi is often demonstrable when the examiner attempts to open the eyes against forced eyelid closure, nearly impossible in an unaffected patient. The "peek-a-boo" sign of lagophthalmus after prolonged forced eyelid closure suggests fatigue in this muscle.

When the more ptotic eyelid is lifted above the iris by the examiner, the contralateral eyelid drops ("curtaining"). This phenomenon results because of the law of equal innervation to both levator palpebrae superioris muscles (Hering's law) [14]. Eyelid curtaining has also been described in cases of third nerve palsies and in cases of narrowed palpebral fissures due to peripheral facial palsies [14,15].

Anisocoria and failure of pupil constriction to light are not observed in OMG. Subclinical pupil sphincter fatigue with repeated light stimulation as well as accommodative fatigue have been reported; however, we do not routinely evaluate for these [16,17].

The use of edrophonium (Tensilon) in the evaluation of myasthenia has fallen out of use, and the drug is no longer available in the United States and many other countries [18]. The sensitivity of the Tensilon test for OMG was 85 to 95 percent, similar to that for GMG, but it was associated with false-positive results [19-21]. (See "Diagnosis of myasthenia gravis", section on 'Pharmacologic testing'.)

DIFFERENTIAL DIAGNOSIS — The differential diagnosis of the clinical manifestations of OMG depends on the symptom cluster:

●Isolated ptosis

●Isolated ophthalmoparesis

●Ptosis and ophthalmoparesis

●Ptosis, ophthalmoparesis, and other signs or symptoms of bulbar weakness

The differential diagnosis of isolated ptosis and ophthalmoparesis is presented elsewhere. (See "Overview of ptosis" and "Overview of diplopia".)

Conditions that cause bilateral impairment of both eyelid and extraocular muscle function are most likely to be considered in the differential diagnosis of OMG.

Thyroid ophthalmopathy — Graves' disease produces restricted eye movements due to a constrictive ophthalmopathy. It can usually be differentiated from MG by the lack of ptosis and the presence of proptosis, lid retraction, lid lag, and periorbital edema. However, thyroid disease can coexist with MG. (See "Clinical features and diagnosis of thyroid eye disease".)

Restricted eye movement in Graves' ophthalmopathy can be confirmed by forced duction testing, which involves grasping the extraocular muscle insertion with a forceps or placing a suction cup on the eye and rotating the eye manually in the direction of the reduced movement. If the reduced excursion cannot be overcome with these maneuvers, the diagnosis is a shortened muscle rather than weakness.

In addition, the enlarged extraocular muscles of Graves' disease can usually be seen on computed tomography (CT) images of the orbits.

Because MG and Graves' disease can be comorbid, we recommend thyroid function studies before treatment of OMG is instituted, even when the diagnosis of OMG is clear.

Chronic progressive external ophthalmoplegia — Chronic progressive external ophthalmoplegia (CPEO) and Kearns-Sayre syndrome (KSS) are mitochondrial disorders that produce progressive and generally symmetric ophthalmoparesis and ptosis. In KSS and CPEO, saccades are generally slow in all directions, whereas in OMG, saccades are of normal velocity even if they are of limited amplitude.

CPEO generally remains isolated to the extraocular muscles, but some CPEO syndromes and KSS involve other findings, including generalized muscle weakness, cerebellar ataxia, and reduced vision from retinal degeneration, abnormalities not found in OMG [22]. Most patients with KSS present with ptosis months to years before the onset of ophthalmoparesis. Because of the symmetric nature of the ophthalmoparesis, patients with KSS and CPEO typically do not complain of diplopia except at reading distance, where reduced convergence creates ocular misalignment. (See "Myopathies affecting the extraocular muscles in children", section on 'Kearns-Sayre syndrome'.)

Muscular dystrophy — Myotonic dystrophy and oculopharyngeal dystrophy may produce ptosis and ophthalmoparesis.

Myotonic dystrophy is an autosomal dominant disorder characterized by variable ptosis and weakness of the face, jaw, and neck as well as weakness of the extremities. Associated abnormalities include cataracts, cardiac conduction defects, characteristic facies (long face with atrophy of temporalis and masseter), frontal balding, and variable intellectual impairment. Patients may also develop dysphagia and ophthalmoparesis in some cases. (See "Myotonic dystrophy: Etiology, clinical features, and diagnosis".)

Oculopharyngeal dystrophy is characterized by slowly progressive ptosis and dysphagia with onset in the fourth or fifth decade of life. Weakness of the proximal muscles is commonly present upon initial presentation, but ophthalmoparesis typically develops after the onset of ptosis and dysphagia. (See "Oculopharyngeal, distal, and congenital muscular dystrophies", section on 'Oculopharyngeal muscular dystrophy'.)

Brainstem and motor cranial nerve pathologies — Structural disease of the brainstem can cause manifestations that resemble OMG. As examples, parasellar tumors and aneurysms can impair function of the third, fourth, and sixth cranial nerves [23-25]. The presence of trigeminal dysfunction and/or pupillary abnormalities is inconsistent with OMG and may point to a structural lesion. Brain magnetic resonance imaging (MRI) to exclude these disorders is warranted in unconfirmed cases of OMG.

Multiple motor cranial neuropathies, such as those produced by carcinomatous or lymphomatous meningitis, may also produce eye movement abnormalities that may be confused with MG. If the diagnosis of MG is not firmly established in cases of possible multiple cranial nerve abnormalities, examination of the cerebrospinal fluid for abnormal cells and cytology is usually necessary.

In one case series of 979 patients with multiple cranial nerve palsies (not necessarily including or isolated to the oculomotor cranial nerves), tumor was the most prevalent cause (30 percent) [26]. Other causes included vascular disease, trauma, infection, Guillain-Barré syndrome, cavernous sinusitis, diabetes, and multiple sclerosis.

EVALUATION AND DIAGNOSIS — The diagnosis of OMG should be suspected in a patient with typical history and examination findings and a positive result on ice test. Further confirmation should be obtained by a positive acetylcholine receptor antibody (AChR-Ab) titer. If this is negative (as it will be in approximately half of patients with OMG), then electromyography (EMG) with repetitive nerve stimulation (RNS) or single-fiber EMG (SFEMG) should be performed. If the titer is positive, then electrophysiologic testing is generally unnecessary. The sensitivity and specificity of tests are different for generalized MG (GMG) versus OMG (table 1).

Serum antibody studies — While more than 85 percent of patients with GMG have serum AChR-Ab [7,27-30], the sensitivity of AChR-Ab testing in OMG may be as low as 45 to 60 percent [27,31,32]. However, this is the most specific test for MG; no false positives have been reported (100 percent specificity). (See "Diagnosis of myasthenia gravis", section on 'Acetylcholine receptor antibodies'.)

Some patients with GMG who are seronegative for AChR-Ab have antibodies against the muscle-specific tyrosine kinase (MuSK) [33-35]. Once thought to be unassociated with OMG, MuSK antibodies have been detected in patients with OMG in a few case reports [36-38]. (See "Diagnosis of myasthenia gravis", section on 'MuSK antibodies'.)

A number of reports have linked low-density lipoprotein receptor-related protein 4 (LRP4) antibodies with GMG [39]. In three reported cases of OMG, patients who were found to be seronegative for both AChR-Ab and MuSK were found to have positive LRP4 antibodies [40]. Other antibodies with a potential role in the pathogenesis and diagnosis of GMG are discussed separately. (See "Diagnosis of myasthenia gravis", section on 'LRP4 antibodies'.)

Electrophysiology — Electrodiagnostic studies are an important supplement to the immunologic studies and may also provide confirmation of the diagnosis of myasthenia. These tests may not be necessary if serum antibody tests are diagnostic.

Repetitive nerve stimulation — Repetitive nerve stimulation (RNS) studies demonstrate decrement in the amplitude of the compound muscle action potential after repetitive stimulation of the motor nerve to that muscle [41] (see "Diagnosis of myasthenia gravis", section on 'Nerve conduction testing with repetitive nerve stimulation'). The orbicularis oculi may be studied in patients with OMG.

Among patients with GMG, the sensitivity of RNS is greater than 70 percent [42]. By contrast, among patients with OMG, the sensitivity was as low as 15 percent in some series, and only somewhat higher, 33 to 45 percent, in series where stimulation at multiple sites was performed [43,44]. The specificity is 89 percent; both false negatives and false positives do occur [21].

Single-fiber EMG — SFEMG identifies abnormal neuromuscular transmission by measuring temporal variability in the firing of adjacent motor nerve fibers from a single motor unit, a phenomenon called "jitter." SFEMG is more sensitive than RNS and may identify electrophysiologic abnormalities in clinically strong muscles. (See "Diagnosis of myasthenia gravis", section on 'Single-fiber electromyography'.)

Depending on which and how many muscles are examined, the sensitivity of SFEMG for OMG is between 63 and 100 percent [45-47]. Evaluation of the orbicularis oculi and frontalis muscles increases the sensitivity in patients with OMG [47,48]. However, this testing is available only at specialized centers with neuromuscular expertise.

Abnormal jitter is not specific for myasthenia. However, while patients with MG generally have an otherwise normal standard EMG examination, other neuromuscular diseases that increase jitter do not. An exception may be in differentiating patients suspected of OMG from those with chronic progressive external ophthalmoplegia (CPEO). CPEO is often, but not always, associated with a myopathic pattern on EMG, and SFEMG does not clearly help distinguish these two entities. Two reports give conflicting data: in the presence of a normal needle EMG examination, one showed abnormal SFEMG in CPEO [49], while the other did not [50].

Additional testing — If the diagnosis remains unclear after serum antibody testing and/or EMG, brain magnetic resonance imaging (MRI) and lumbar puncture (LP) with spinal fluid examination may be needed to exclude treatable inflammatory or structural disease of the brainstem or cranial nerve roots [23-25]. (See 'Differential diagnosis' above.)

Patients with suspected OMG should also have a chest computed tomography (CT) scan to rule out thymoma and thyroid function tests to rule out associated thyroid dysfunction.

PROGNOSIS — Of patients presenting with OMG, two-thirds will go on to develop signs and symptoms of extremity weakness and other bulbar muscle weakness, while one-third will continue to have pure OMG [2,3]. Most (78 percent) of those who will develop generalized MG (GMG) will do so within the first year, and 94 percent will do so within three years.

Neither age nor sex has been consistently shown to alter the course of disease. A normal single-fiber electromyography (SFEMG) study may be helpful in stratifying risk of generalization. In one study of 37 patients with OMG followed for two years, 82 percent of those with normal SFEMG in the extensor digitorum communis persisted with isolated OMG, whereas 58 percent of those with an abnormal study developed GMG [51].

TREATMENT — Treatment considerations in the management of OMG include symptomatic and immunomodulatory treatment of myasthenia, thymectomy, and corrective treatments of ptosis and strabismus.

In patients with MG, certain drugs prescribed for coexisting medical conditions may exacerbate muscle weakness. A few medications such as fluoroquinolones and telithromycin carry US Food and Drug Administration (FDA) "black box warnings" for patients with MG (table 2). Patients should be advised to discuss all new medications with their physicians.

Symptomatic management of ptosis and diplopia — Assistive devices to keep the ptotic lid(s) elevated can be used pending a response to acetylcholinesterase inhibitors or immunosuppressive therapy. Devices consist of a ptosis "crutch" or lid adhesive tapes, but these are problematic due to discomfort and corneal dryness, and crutches are difficult to find.

An eye patch, opaque contact lens, or occlusion of an eyeglass lens are simple ways to eliminate diplopia. A patch that is concave (to avoid injury to the cornea) can be worn over either eye; it is not necessary to alternate the patch between eyes in adults. Occlusion of an eyeglass lens with opaque tape works as an eye patch does. For cosmetic reasons, an opaque contact lens may be more acceptable than patching or an opaque eyeglass lens. These interventions also have the disadvantage of eliminating depth perception, which usually requires unoccluded vision in two eyes. Some patients require stereoscopic vision in their occupation. Other activities of daily living, such as driving and watching television, are also impaired with a loss of depth perception; however, many patients are willing to sacrifice depth perception in favor of eliminating diplopia.

Prism lenses offer another nonpharmacologic alternative for patients with diplopia. If the ophthalmoparesis is stable over several weeks or months, prism lenses may minimize diplopia. However, fatigability and significant variability of eye movements and ocular alignment will limit the benefit of prisms in patients with OMG until stabilization occurs. Press-on, flexible prism lenses are a less expensive and more adaptable alternative than ground-in prism lenses, and they can be removed if there is considerable fluctuation of the ophthalmoparesis. A disadvantage is that they can distort vision when the prism strength is high.

Anticholinesterase agents — Pyridostigmine (Mestinon) is the most commonly used anticholinesterase agent for symptomatic treatment of myasthenia. No randomized trials have studied the efficacy of anticholinesterase agents in OMG [52]. Unfortunately, in our clinical experience, pyridostigmine treatment alone rarely results in resolution of ocular symptoms, particularly diplopia. In one case series, pyridostigmine treatment alone produced resolution of primary gaze diplopia in only 6.9 percent [53]. It is best used for very mild cases of OMG or as adjunctive symptomatic treatment for moderate or severe OMG. The dosing regimen and side-effect profile for pyridostigmine are the same for patients with generalized MG (GMG) and OMG. (See "Overview of the treatment of myasthenia gravis", section on 'Initial symptomatic therapy'.)

Immunosuppressive agents — Some clinicians prefer to reserve immunosuppressive treatment for patients with more severe disease than that manifested by pure OMG. However, in order to achieve resolution of ptosis or ophthalmoparesis, patients with OMG usually require immunosuppression. Many patients are not satisfied with eye patching, prism lenses, or nonpharmacologic therapy and prefer immunosuppression despite the known risks.

Retrospective data indicate that immunosuppressive treatment of OMG may decrease the likelihood of developing GMG [52,54-56]. It is not clear from these studies whether treatment actually reduces the incidence of GMG, delays its onset, or just masks its symptoms. Prospective studies are needed to confirm this finding [52].

Plasmapheresis and intravenous immune globulin are used for the short-term management of severe GMG and are used only in OMG when bilateral ptosis is debilitating. (See "Overview of the treatment of myasthenia gravis", section on 'Plasma exchange and IVIG as rescue or bridge therapies'.)

Prednisone — The most commonly used immunosuppressive agent for the treatment of myasthenia is prednisone. However, the use of corticosteroids in OMG is somewhat controversial [57,58]. In making a decision to use corticosteroids in OMG, clinicians and patients must balance the severity of their symptoms and the efficacy of nonpharmacologic measures (eg, ptosis tape to elevate the eyelid and patching to eliminate diplopia) with the difficulty in weaning prednisone in some patients and the side effects associated with corticosteroid therapy. Also, 10 percent of patients with OMG will have spontaneous remission without corticosteroids [58].

The efficacy of corticosteroids was demonstrated in a case series of 55 patients with OMG [53]. Prednisone produced symptom resolution at one month in approximately 70 percent that was sustained over a two-year follow-up period. Another observational study of 35 patients found that ocular symptoms were more likely to resolve with corticosteroids than with anticholinesterase treatment (70 versus 29 percent) [59].

In addition to symptomatic benefit, there are some studies that suggest that treatment with corticosteroids may reduce the progression to GMG [54,55,60-62]. As an example, one retrospective study found that among 56 patients with OMG assessed at two years after onset of treatment, fewer prednisone-treated patients (3 of 27) had developed generalized disease compared with 10 of 29 pyridostigmine-treated patients [60].

Neurologic worsening with initiation of high-dose prednisone is less common in patients with OMG than with GMG. Nonetheless, starting with lower doses of 10 to 15 mg of prednisone daily and gradually increasing over three to four weeks is recommended, as some patients may have unrecognized GMG. Maximum daily doses of 0.5 to 1 mg/kg of prednisone are often needed for several weeks (occasionally months) in order to maintain benefit.

There is no single recommended dosing regimen. Because of the problematic side effects of chronic corticosteroid use, tapering to the lowest effective dose is essential after initial resolution and stabilization of symptoms. Low maintenance doses (5 to 10 mg every other day) are often effective in OMG, but response is highly variable. Alternate-day dosing is preferred. (See "Chronic immunotherapy for myasthenia gravis", section on 'Glucocorticoids'.)

Steroid-sparing agents and other immunosuppressive agents — Azathioprine, mycophenolate mofetil, cyclosporine, and rituximab are second-line immunosuppressants for patients who do not respond to or tolerate prednisone. These agents can also be used with prednisone in order to wean prednisone. The side effects of these medications often limit their use for patients with OMG, but they may be necessary to control symptoms and may be preferred by patients if they do not experience major side effects [63]. While observational studies suggest that azathioprine reduces the risk of progression to GMG, further studies are required to confirm this finding [55,62,64]. Serious risks, such as lymphoma with azathioprine and renal toxicity with cyclosporine, must be weighed against symptom severity in individual patients. The role of eculizumab for the treatment of MG is evolving, but response to and appropriateness for OMG is not known [63]. (See "Chronic immunotherapy for myasthenia gravis", section on 'Glucocorticoid-sparing therapy'.)

Thymectomy — Thymectomy is required for patients who have a thymoma and myasthenia. Thymectomy is also recommended for certain patients with nonthymomatous GMG. This is discussed separately. (See "Role of thymectomy in patients with myasthenia gravis".)

For nonthymomatous OMG, practice varies. Because prospective studies have not been performed in patients with nonthymomatous OMG, data regarding the benefit of thymectomy in these patients are limited. Several case series have demonstrated a similar response to thymectomy in OMG as with GMG [65-68]. Remission rates of 57 to 71 percent over more than five years have been reported after thymectomy in OMG patients [68,69]. In addition, no patients in two series of 96 total patients [65,69], and only 1 in a series of 61 patients [67], developed GMG after thymectomy. This contrasts with the expected 50 percent generalization rate over this timeframe. In a systematic review and meta-analysis of 26 studies, thymectomy was beneficial for nonthymomatous OMG, with complete stable remission of 51 percent, although substantial heterogeneity among studies limited certainty of this estimate [70].

Prospective, controlled trials are needed to determine if there is a reproducible beneficial effect of thymectomy in nonthymomatous OMG. In the meantime, while some centers do not advise thymectomy for patients with OMG in the absence of thymoma [55,71], others offer this as a therapeutic option, particularly if the less invasive extended-cervical procedure is performed [72]. Current guidelines note that thymectomy may be offered to patients with acetylcholine receptor antibody (AChR-Ab) positive nonthymomatous OMG who do not respond to acetylcholinesterase inhibitors or immunosuppressive agents or who have contraindications or cannot tolerate immunosuppressants [63].

Surgery for ptosis and diplopia — Surgical correction can be performed on patients with stable ptosis. For patients who do not demonstrate stability prior to surgery, recurrence of ptosis is common due to the nature of myasthenia. The optimal duration of stability prior to surgery is unknown, although there are some indications that three to four years of stability prior to ptosis repair is appropriate [73]. (See "Overview of ptosis".)

Extraocular muscle (strabismus) surgery can be performed on patients with stable symptomatic ophthalmoparesis that persists despite medication. Similar to the situation for ptosis, recurrence of diplopia is common for patients who do not demonstrate stability prior to surgery. The optimal duration of stability prior to surgery is unknown; shorter durations of stability (five to six months) prior to strabismus surgery are advocated compared with those recommended for ptosis surgery, but follow-up data on these patients are limited [74]. Based on natural history data, which indicate that myasthenia has a more stable course three years after onset [2,75], a conservative approach is to consider strabismus surgery three years after onset of OMG if deficits have been stable for at least one year and if the patient desires surgery and has either failed to respond to medical therapy or has contraindications for continued medical therapy.

SUMMARY AND RECOMMENDATIONS

●Clinical features – The diagnosis of ocular myasthenia gravis (OMG) is suggested by the triad of ophthalmoparesis, ptosis, and orbicularis oculi weakness; however, isolated ptosis and ophthalmoparesis are also common presentations. Demonstrating fatigue in the ocular muscles is a helpful but not infallible diagnostic sign for OMG. Pupillary function is spared. (See 'Clinical features' above.)

●Differential diagnosis – Clinicians should consider thyroid disease, muscle disease, and other brainstem and cranial nerve lesions in the differential diagnosis of OMG. (See 'Differential diagnosis' above.)

●Diagnostic evaluation – The diagnosis of OMG should be suspected in a patient with ophthalmoparesis, ptosis, and/or orbicularis oculi weakness, especially with a positive result on the ice test or evidence of fatiguability.

A positive acetylcholine receptor antibody (AChR-Ab) titer confirms the diagnosis. If this is negative, then an electrophysiologic test should be performed.

If the diagnosis remains unclear, then brain magnetic resonance imaging (MRI) and lumbar puncture (LP) should be performed to rule out other diagnoses. (See 'Evaluation and diagnosis' above.)

●Additional testing – Patients with suspected OMG should have thyroid function tests and a chest computed tomography (CT) to exclude thymoma. (See 'Additional testing' above.)

●Progression to generalized MG (GMG) – Two-thirds of patients with OMG will develop signs and symptoms of extremity weakness and other bulbar muscle weakness, while one-third will have pure OMG. Most of those who will develop GMG do so within the first year, and virtually all will do so within three years. (See 'Prognosis' above.)

●Medical treatment – We suggest anticholinesterase agents as a first treatment in OMG (Grade 2B). Most patients have some benefit from anticholinesterase agents, although many will not fully respond. (See 'Anticholinesterase agents' above.)

For patients who continue to have disabling symptoms despite treatment with an anticholinesterase agent, we suggest immunosuppressive therapy (Grade 2B). Prednisone is most commonly used. In deciding to use prednisone, clinicians and patients must balance the severity of symptoms and the efficacy of nonpharmacologic measures with the long-term side effects and the difficulty in weaning. (See 'Immunosuppressive agents' above.)

●Role of thymectomy – Thymectomy is recommended for individuals with thymoma. (See "Clinical presentation and management of thymoma and thymic carcinoma", section on 'Management of localized disease'.)

For most patients with nonthymomatous OMG, we suggest not performing thymectomy (Grade 2C). However, surgery may be considered in select patients with AChR-Ab and disabling, medically refractory disease or significant contraindications to medical treatment. This decision must consider the risks of progression to GMG and the risk of surgery. Less invasive (ie, thorascopic) surgeries at experienced medical centers can decrease the risk associated with thymectomy. (See 'Treatment' above and "Role of thymectomy in patients with myasthenia gravis".)

●Persistent diplopia and ptosis – Patients with stable diplopia or ptosis despite maximal medical therapy may benefit from surgical interventions. A period of stability of several months to a few years is suggested prior to surgical intervention. (See 'Surgery for ptosis and diplopia' above.)