Documenta Ophthalmologica 84: 309-333, 1993. © 1993 Kluwer Academic Publishers. Printed in the Netherlands.

Ocular myasthenia gravis

A critical review of clinical and pathophysiological aspects NORBERT SOMMER, ARTHUR MELMS, MICHAEL WELLER 1 & JOHANNES DICHGANS Department of Neurology, Tiibingen University, Tiibingen, Germany; 1Present address: Department of Clinical Immunology, University of Zurich, Switzerland Accepted 25 October 1993

Key words: Acetylcholine receptor, Autoimmunity, Extraocular muscles, Immunosuppressive therapy, Myasthenia gravis, Tensilon test Abstract. Myasthenia gravis (MG) is probably the best studied autoimmune disease caused by autoantibodies against the acetylcholine receptor (AChR) at the neuromuscular junction, subsequently leading to abnormal fatigability and weakness of skeletal muscle. Extraocular muscle weakness with droopy eyelids and double vision is present in about 90% of MG patients, being the initial complaint in about 50%. In approximately 20% of the patients the disease will always be confined to the extraocular muscles. The single most important diagnostic test is the detection of serum antibodies against AChR which is positive in 90% of patients with generalized MG, but only in 65% with purely ocular MG. Electromyographic studies and the Tensilon test are of diagnostic value in clear-cut cases, but may be equivocal in purely ocular myasthenia, especially the latter not rarely producing false-positive results. Treatment response to corticosteroids and anti-cholinesterase agents is satisfactory in many patients with ocular MG, however other immunosuppressive drugs may also be needed. Pathogenetically relevant steps of the underlying autoimmune process have been elucidated during the last few years; nevertheless a number of questions remain open, especially what starts off the autoimmune process, and why are eye muscles so frequently involved in MG? Abbreviations: AChR - acetylcholine receptor; EOM - extraocular muscles; MG - myasthenia

gravis

Introduction Myasthenia gravis ( M G ) is not a rare disease. In recent epidemiological studies the prevalence rate was estimated up to 1:10,000 with an annual incidence of 2 - 4 per million [1-4]. It m a y affect individuals of any race at any age with a preference for w o m e n under 40 and men over 40 years of age

[5, 61. The spontaneous course of the disease is now rarely fatal [7]. Before 1934, when anticholinesterase drugs were introduced into therapy, the mortality rate was 70% since only the most severely affected patients were recognized to have myasthenia. In the nineteen forties and fifties still a third of the patients died of the disease, most often in a crisis with respiratory

310

insufficiency within the first three years after onset [7, 8]. In the late sixties and early seventies corticosteroids and azathioprine were increasingly used [9] and further improved prognosis. The introduction of plasmapheresis, early thymectomy, and improved intensive care facilities have reduced the mortality to probably well below 3% since the mid-seventies [7]. Nowadays the diagnosis of myasthenia gravis does not normally implicate a reduced life span, if it is treated promptly and adequately. Two classifications have been applied to divide patients into subgroups and provide a basis for clinical management. They are used in parallel and complement each other. Osserman's classification [10] grades the patients according to clinical severity into groups I (purely ocular), IIa, IIb, III (mild, moderate, acute severe), and IV (late severe). Newsom-Davis' classification [5, 6, 11, 12] uses a pathophysiological approach and is based on the clinical heterogeneity due to age of onset, thymic abnormalities and other immune parameters (Table 1). Approximately 50% of the patients present with ocular symptoms and therefore ophthalmologists or neurologists will be the first specialists a myasthenic patient is referred to. Ocular MG has to be considered in every patient with diplopia or ptosis or both and physicians of either specialty

Table 1. Clinical heterogeneity in myasthenia gravis (based on references 5, 6, the relation of purely ocular MG (group D) to the other groups of generalized

Group Clinical features at onset

11, 12). MG.

A

B

C

D

'E

(young onset)

(old onset)

(thymoma)

(ocular)

(sero-negative)

Gener?

Gener.

Age

<45

>45

Gener. Any

Ocular Any

Gener. Any

Thymic pathology

Hyperplasia

Atrophy

Thymoma

?b

T-cell

AChRantibodies °

+ +

+

+ or -

-

Strong

Less strong

None

NA ~

NA

+

+ or -

+

HLA

association

(B8, D R 3 )

or

+ +

+

or

+ +

(B7, D R 2 )

Treatment response to

Anti-cholinesterase drugs Immunosuppression e Thymectomy Plasmapheresis

Note

+

+

+

+

+

+

+

+ +

-(?)

+ or -

?b

?

+

+

NA

+

" Generalized symptoms. b Preliminary data available [114]. c + + high, + low, - negative. d N A = not available.

° Corticosteroids and/or azathioprine.

areas

311 should recognize the clinical peculiarities of MG and apply the diagnostic steps that are required to confirm or exclude this diagnosis. The long-term medical care of an MG patient should always be guided by a specialized centre and the patient should regularly be seen there. Nevertheless the patient and his or her general practitioner must be wellinformed of the long-term problems due to the disease itself and the potential hazards brought about by the treatment performed. This paper essentially deals with ocular myasthenia gravis and its clinical features, treatment and pathophysiology. However, since ocular and generalized myasthenia gravis are closely related in many respects, the problems of generalized MG are also discussed where it seems necessary.

1. Diagnosis of myasthenia gravis (in general) The clinical diagnosis of MG is based on the findings of exertional muscle weakness and improvement of muscle strength after rest. Most commonly affected are the extraocular, facial, pharyngeal, and proximal limb muscles. Weakness is often asymmetrical, tendon stretch reflexes are normal, and a sensory deficit is not found. Muscle pain is atypical of MG, but some patients report pain or cramps in chronically fatigued muscles, particularly the neck muscles. The 'typical patient' is either a young female or an elderly male with (a) droopy eyelid(s) and/or double vision, (b) difficulty chewing and swallowing, or (c) exertional fatigability of proximal extremity muscles or a combination of (a) to (c). Nevertheless MG may occur at any age and any striated muscle may be affected. Symptoms often vary during the day and from one day to another and are normally more pronounced in the evening and after exercise. The remaining ophthalmologic and neurological examination is normal. The following diagnostic tests should be run in the given order when history and examination are suspect of MG. The suggested sequence is solely based on practical considerations and does not reflect their significance, e.g. the detection of AChR antibodies is far more specific than a positive Tensilon test, and the EMG findings obviously depend on the muscle group investigated. 1.1 Tensilon test

This procedure is easy to perform and has great diagnostic value in clear-cut cases [13-15]. However, problems may arise to the inexperienced physician, if the result is equivocal, or the patient shows side effects of the drug. The intravenous application of the cholinesterase blocking agent (edrophonium chloride, Tensilon R) leads to a temporary excess of acetylcholine in the neuromuscular synapse and thus increases muscle strength [16]. The effect starts seconds after the injection and lasts for 3-10 minutes. Possible

312 side effect result from the autonomic action of the drug: common but harmless are increased salivation, mild sweating, nausea, and perioral fasciculations; rare, but more severe side effects are hypotension or bradycardia. Tensilon R is available from Roche Products Ltd. (Welwyn Garden City, UK) with 10 mg in i ml ampoules. For better application we dilute 1 ml with 9 ml saline and give an initial dose of 2 ml (i.e. 2 mg) in patients with generalized MG (bulbar or extremity muscle weakness). If there is no response after 60 seconds we inject another 4 mg and observe the patient for another few minutes. A dose of more than 6 mg is unlikely to cause an improvement, if a lower dose does not [11]. In children we inject 0.02 mg/kg body weight initially, followed by twice that amount after 60 seconds, if necessary. We do not routinely administer atropin prior to the test, but always keep a syringe with 1 mg atropin ready in case of bradycardia or collapse. Relative contraindications for Tensilon testing are bronchial asthma or cardiac dysrhythmias. The special problems of the test in ocular MG are discussed below.

1.2 Acetylcholine receptor antibodies The detection of acetylcholine receptor antibodies is the single most important diagnostic investigation in myasthenia gravis [17]. Approximately 90% (87-99%) of patients with generalized disease and 65% (45-71%) with purely ocular MG have an increased serum antibody titer [17-22]. As a rule, patients with ocular myasthenia have relatively low antibody titers, whereas patients with generalized MG or thymoma have intermediate or high titers (Fig. 1). There is no correlation between disease severity and antibody titer in general; however, in an individual patient, the relative titer correlates very well with the development of clinical symptoms. Accordingly, there is a relative decrease in titer in patients with good clinical response to thymectomy and during immunosuppressive medication [23-26]. Regular checking of serum antibody levels is also helpful in patients that have recently come off immunosuppressants, since a rise in titer may precede clinical relapse [27]. Acetylcholine receptor antibodies are highly specific for MG, and are practically never detectable in healthy individuals. They may be found in patients with thymomas without muscle weakness [28]. False positive results have rarely been reported in tardive dyskinesia, motor neuron disease, and myotonic dystrophy [29-31]. This will hardly ever be relevant in the respective clinical situation. Antibodies are measured by a radioimmunoassay (RIA) [17] which should only be performed by a specialized laboratory where a continuous quality control is ensured [32]. Other methods, such as enzyme immonoassays are far less sensitive and specific. Human muscle from amputated legs has mostly been used as source of antigen [33]. AChR prepared from a human

313 I000

AChR-antibody

titer

O t O n m o l / l e 0 O 100

O

e O

• :.

.

O

7

: o

eo

y

:

:

o*

o o

•

°

I0

o

o0

o

o•

o

::

°• # O0 000 OO0 O~O

0.i ~j

~a

~4 C 0

u% C 0

E 0

r. C •~ 0

,-

E U 0

r0

Fig. 1. Acetyicholine receptor antibody titer in various patient groups (see also Table l) plotted on a log-scale. Note that 16 out of 43 (37%) patients with ocular MG have no detectable antibodies. The dashed line shows the cut-off for healthy controls (0.5 nmol/1). rhabdomyosarcoma cell line may be safer, more homogeneous and more readily available, but is only beginning to be used [34].

1.3 Electromyography Electrophysiological methods have been of great diagnostic importance in M G and have held their position despite the advances of immunological techniques. The commonly used method in routine laboratories is repetitive nerve stimulation with 3 Hz and recording the compound muscle action potential ( C M A P ) [35, 36]. In M G there is a pathological decrement of greater than 10% between the first and the fourth (or fifth) potential in at least 60% of the patients [37]. According to the clinical distribution pathological findings are significantly more common in proximal than in

314 distal muscles. We therefore routinely stimulate the accessory nerve and record from the trapezius muscle [38]. A more pronounced decrement may be seen after a period (60 sec) of maximal voluntary activation. Other provocative tests such as standardized electrical stimulation for several minutes, increasing temperature, or application of curare may also be used, but cause considerable discomfort and potential hazards. Pathological decrement is not specific for MG. It is also found in other disorders of neuromuscular transmission, such as Lambert-Eaton myasthenic syndrome, botulism, or non-immune mediated congenital myasthenia [39]. Single fiber electromyography (SF-EMG) is an advanced method of determining neuromuscular defects, disclosing an asynchronous activity ('jitter') of muscle fibers within a single motor unit. The method may be valuable in mild MG because of its high sensitivity (99% if two muscles are tested!), but is rather time consuming and requires more patient cooperation [40]. Pathological jitter may be seen in other disorders of neuromuscular transmission and sometimes in neuropathies, myopathies, and motor neuron disease. These must be excluded by other electrophysiologic and clinical methods. If jitter is normal in a muscle with definite weakness, the weakness is not due to myasthenia. When MG is mild, anti-cholinesterase drugs may mask pathological findings [41]. When abnormal neuromuscular transmission has been demonstrated by repetitive nerve stimulation SF-EMG does not add to the diagnosis [36, 40]. 1.4 Other investigations in MG If MG is certain or at least very likely then the following investigations must be performed as well: (a) CT scan of the chest to search for thymic enlargement or thymoma [42]; (b) striated muscle autoantibodies which are found in 80-90% of thymoma patients [43]; (c) thyroid function and autoimmune serology, since up to 10% of all MG patients have other autoimmune disease, most commonly autoimmune thyroiditis [44].

2. Diagnostic problems in ocular myasthenia Most typical of ocular MG is the asymmetrical and variable expression of eye muscle weakness, symptoms being more pronounced after exercise and in the evening. Normally, several extraocular muscles are affected in patterns that are not characteristic of lesions of one or more nerves. Apart from ptosis the most frequently affected muscles are probably the medial recti, followed by the superior recti [47]. In some patients eye movements are completely normal after rest and only provocative tests with looking upward or to one side for 60 seconds give the clue to the diagnosis (Simpson's sign) [45, 46]. Sometimes a kind of paretic nystagmus can be seen with upward or lateral gaze [47]. In more severe and longstanding cases

315 fluctuations can be almost absent, and only involvement of muscles innervated by multiple cranial nerves, especially if bilateral, make the physician consider a myopathic syndrome [48]. Symptoms often worsen while reading, watching television or driving, especially in bright sunlight. Many patients have had a spell of ptosis or diplopia in the past or have changed eyeglasses several times in an attempt to correct blurry vision, and the physician should specifically enquire about that. 'Enhanced' ptosis has been reported to be typical- though not specificof ocular MG and is demonstrated during upgaze in patients with bilateral ptosis. Following manual elevation of one eyelid, the contralateral lid closes. This phenomenon may be explained by Hering's law of equal innervation: manual elevation decreases the effort for eyelid elevation ipsilaterally and results in relaxation contralaterally thus leading to 'enhanced' ptosis [49, 50]. The 'ice pack' test has also been advocated for MG, and is based on the fact that cooling facilitates neuromuscular transmission: ptosis may improve by applying ice, wrapped in a towel over the closed eye for 2 minutes [51]. The 'sleep test' has also been recommended and is based on the observation that ptosis or ophthalmoparesis resolves regularly after a 30-minute period of sleep [52]. Cogan's sign ('twitch response') is frequently seen in myasthenic patients with ptosis and is characterized by a quick overshooting upward movement followed by a down-drift of the upper lid after change of fixation from infraversion to the primary position [53]. Nonetheless it has also been reported in various other syndromes [45]. These and other typical, allegedly specific clinical signs have been described in ocular MG, but in fact cannot prove or exclude the diagnosis. Myasthenic pseudo-internuclear ophthalmoplegia [54] or pseudo-one-and-a-half syndrome [55] may also be seen in MG. Abduction nystagmus in these patients suggests a medial longitudinal fasciculus lesion and may lead to the erroneous diagnosis of multiple sclerosis, if myasthenia is not specifically tested for. Exercise induced diplopia on one hand, and fatigable ptosis and upgaze paresis on the other hand, however, have also been reported in tumours of the posterior fossa [56, 57]. Pupils are normal in MG. Slower pupillary responses have been reported in patients being treated for myasthenia gravis [58], but this may be an effect of anticholinesterase agents or corticosteroids rather than reflecting autonomic dysfunction. Orbicularis oculi weakness should also be tested; it may be the first sign of generalization, but in other patients may persist in the context of purely ocular MG [46]. Symptoms in ocular myasthenia result from the combination of muscle weakness and central adaptive mechanisms [59,60]. Abnormalities of saccades have been documented a number of times [59, 61, 62]. Large saccades are often hypometric, whereas smaller saccades may be normal or hypermetric due to adaptive mechanisms. The latter phenomenon has been explained by the central nervous system having increased the size of the saccadic impulse in an attempt to overcome the myasthenic weakness [63]. It

316 has been observed that saccades of supernormal velocity may persist for at least 10 days after eye muscle pareses have become normal under treatment; after a year's treatment the adaptation mechanisms and the saccadic velocity had been decreased [64]. Intersaccadic variation was found in 18 of 22 patients with myasthenic eye muscle pareses using infrared reflection oculography and saccadic abnormalities were also documented in clinically uninvolved eyes [62].

2.1 Tensilon test in ocular myasthenia gravis This is the simplest, but also the most subjective and controversial test for myasthenia gravis, especially when EOM weakness is concerned. Many neurologists now do not regard Tensilon responsiveness as specifically diagnostic of myasthenia gravis [65]. In ocular myasthenia the evaluation is particularly difficult, because a minimal change in muscle strength has to be interpreted. The test has never been standardized and there are great variations in practical performance and personal experience of the examiner. We therefore strongly discourage the widespread practice that a positive Tensilon test is taken as a certain diagnostic sign for ocular myasthenia gravis without careful consideration of other clinical and laboratory investigations. The Tensilon test has been reported to be positive in up to 95% of the patients with purely ocular myasthenia [66], though 70-80% positives [67] do more reflect our own experience. More importantly, a false positive test is not rare and is found in patients with parasetlar tumors, aneurysms and other causes of EOM weakness [68]. Moorthy et al. reported eight patients with eye muscle weakness and positive response to anti-cholinesterase treatment who later turned out to have intracranial mass lesions. Mild involvement of intraocular muscles was described in two, bilateral EOM weakness in two. In none of them anti-AChR antibodies were positive, and-atypical of m y a s t h e n i a - a t least four complained of headaches. The authors concluded retrospectively, that probably four patients had solely intracranial mass lesions, whereas the four others might have had both myasthenia gravis and an intracranial mass [68]. The Tensilon test should only be performed, if there is an obvious ptosis or eye muscle paresis. Without an objective clinical sign evaluation is not possible [15]. Moreover, the test should only be considered 'positive', if there is a definite improvement; a doubtful result, or too great a spontaneous fluctuation in muscle strength should be considered 'negative'. Many specialized neuroophthalmologic tests, above all the Lancaster red-green test, have been propagated to render Tensilon testing more objective, but naturally cannot make it more specific [14, 69]. The technique of the test should be similar as in patients with generalized MG (outlined above), but the required dose for improvement is clearly lower in ocular muscles. We usually start with 0.5 mg (i.e., 0.5 ml of a 1 in 10 dilution!) followed by another 0.5 mg after 2 minutes. If still no improvement occurs again 1-2 mg

317 are given after 2 more minutes. A total dose of more than 3-4rag is inefficient and will often worsen symptoms again due to a depolarisation effect, even if the 'low dose response' had been clearly positive. When muscle weakness is suspected to be a hysterical symptom [70], the effect of physiological saline may be tested first as negative control.

2.2 AChR antibodies and 'seronegative ocular myasthenia' The detection of anti-AChR antibodies is certainly diagnostic of MG. Problems therefore arise primarily in patients with suspected ocular myasthenia that are seronegative. They constitute 30-40% of the cases [17-22 and Fig. 1]. EMG studies, especially single-fiber EMG (see above), of clinically uninvolved limb muscles will often be valuable in these cases, but cannot distinguish between autoimmune myasthenia and other rarer disorders of the neuromuscular junction [39, 40]. Since MG patients with thymoma virtually always are antibody-positive [71, and Fig. 1], CT of the chest can add little to the diagnosis at this stage. 'Seronegative ocular myasthenia' is thus a mainly clinical diagnosis and requires a careful diagnostic work-up. We strongly recommend that cranial CT with contrast dye should be performed, if the slightest doubt arises as to the origin of the syndromes, even if Tensilon testing is clearly positive [65, 68]. This also applies, when an apparently established ocular MG worsens and becomes refractory to a so far sufficient therapy (Fig. 2). Approximately half the patients with newly diagnosed ocular MG have symptoms and signs compatible with a unilateral lesion (Table 2) and both MR and angiogram may be necessary to exclude a tumour of the cavernous sinus or an aneurysm. CSF studies should search for an inflammatory process such as tuberculous meningitis or neurosarcoidosis, both of which may present exclusively with oculomotor symptoms [72]. Table 3 outlines the differential diagnosis of ocular myasthenia gravis. A number of other diagnostic tests such as the regional ,curare test or measurement of the acoustic stapedial reflex [73] have been described in the diagnosis of mild myasthenia but are of minimal practical value in equivocal cases. None of them is used in our department. If ocular myasthenia gravis is not certain but very likely, it is reasonable to start a therapeutic trial with pyridostigmine and corticosteroids according to the principles outlined below. The effect of pyridostigmine will be evident after a few days; corticosteroids will have to be given for at least six to eight weeks before the diagnosis can be reevaluated.

3. Treatment of myasthenia gravis (in general) There is no single therapy that is best for all patients with MG. The clinician must choose among five treatment measures and decide the sequence and combination of these for each individual patient. Planning the treatment

318

Fig. 2. MR scan showing a contrast enhancing lesion in the region of the left cavernous sinus, probably meningioma (courtesy of Dr C. Thomas-Ziegler). This 57-year-old female patient presented with a 'partial third nerve palsy' on the left with minor daily fluctuations. On the grounds of a positive Tensilon test the diagnosis of ocular myasthenia gravis was made. AChR antibodies and EMG were negative. She was thymectomized, because thorax CT was suspect of an enlarged thymus. Histology showed thymic remnants, but no germinal centres. Symptoms then resolved completely with 120 mg Mestinon daily and 20 mg prednisolone on alternate days. Three years later ocular symptoms worsened again after tapering steroids and MR was performed. Since the tumor could not be operated on for obvious anatomic reasons, Mestinon and steroids were increased to the former dosage which led to remission again. This history is compatible with ocular myasthenia or meningioma in the left cavernous sinus or both. It illustrates well the problems of 'seronegative ocular myasthenia gravis' s c h e d u l e has to i n c l u d e aspects of severity a n d d u r a t i o n of the disease, as well as p o t e n t i a l side effects a n d a c c e p t a n c e of the p a t i e n t . A n t i - c h o l i n e s t e r ase agents, corticosteroids, i m m u n o s u p p r e s s i v e drugs, t h y m e c t o m y , a n d p l a s m a p h e r e s i s are briefly o u t l i n e d b e l o w , the p r o b l e m s arising in p a t i e n t s with o c u l a r m y a s t h e n i a are t h e n discussed separately. A t this p o i n t is m u s t b e e m p h a s i z e d that the principles of t h e r a p y h a v e hardly c h a n g e d d u r i n g the last fifteen years. D e s p i t e the r e m a r k a b l e progress in u n d e r s t a n d i n g the i m m u n o p a t h o l o g y of M G c o n s i d e r a b l e con-

319 Table 2. Unilateral versus bilateral symptoms in 68 patients with newly diagnosed ocular myasthenia gravis

Unilateral a Thymoma Non-thymoma (remaining ocular c) Non-thymoma (developing generalized disease within 2 years) Total

Bilateral b

Unclear

1

2

3

20

22

3

8 29

5 29

4 10

a 'Unilateral' means that symptoms and signs are compatible with unilateral eye muscles weakness. b'Bilateral' stands for a clearly bilateral involvement. CThe table only includes patients suffering from ocular symptoms for at least three months. Patients with earlier generalization are not considered, therefore the total percentage of generalization is comparably low (see Section 4). Table 3. Differential diagnosis of ocular myasthenia gravis

1. 2. 3. 4. 5. 6. 7. 8. 9.

Diabetic third nerve palsy Cranial neuritis Lambert-Eaton myasthenic syndrome Mitochondrial Myopathy (Kearns-Sayre syndrome) Ocular myositis Ocular muscle dystrophy Multiple sclerosis Brain stem tumour Tumour/aneurysm of the superior orbital fissure

troversies about the value of the various treatment modalities still exist [46, 74]. In fact, hardly one prospective double-blind trial has been p e r f o r m e d for any of them. This would seem superfluous for the obviously efficient anti-cholinesterase agents, but the experts' opinion on the use of immunosuppressive drugs and t h y m e c t o m y is not always uniform and often lacking formal statistical evidence. M e s t i n o n ® (pyridostigmine) is the anti-cholinesterase drug most c o m m o n l y

used (orally and intravenously) in M G [16, 75]. Tensilon ® (edrophoniumchloride) is only used for diagnostic purposes due to its short half life. The therapeutic effect of these drugs is based on their cholinesterase blocking effect, subsequently increasing the n u m b e r of acetylcholine molecules in the synaptic gap, so functionally compensating the reduced n u m b e r of acetylcholine receptors at the postsynaptic m e m b r a n e [16]. Consequently this is a purely symptomatic therapy and will not alter the underlying immunological mechanisms or the course of the disease. Anti-cholinesterase drugs have their greatest value in patients with mild M G with a benign course in w h o m they m a y be the only long-term therapy. T e m p o r a r y but often incomplete responses are found in most other patients with m o r e severe disease, but

320 these will have to rely on additional treatment measures. Pyridostigmine, as a rule, is given orally in three to five single doses of 30 - 90 mg. Its clinical effect lasts from approximately 30 minutes to 5 hours after ingestion [76]. Its intestinal absorption may greatly vary and plasma levels do not too closely correlate with the clinical effect [77]. Side effects are normally harmless and the exact dosage is not too critical; out-patients quickly learn to adapt their medication to every day needs increasing it in situations of physical stress. The long-acting pyridostigmine variant (Mestinon retard ®, Mestinon Timespan ®) is less often used by us (and others [37, 78]) because its pharmacokinetics is not well established. An overdosage may lead to cholinergic side effects such as sweating, lacrimation, increased production of mucus and saliva, abdominal cramping, diarrhea, and fasciculations, which in fact rarely occur in patients treated for MG with average doses (i.e., up to 500 mg/d). An acute myasthenic crisis (due to decompensation of the disease) may not easily be distinguished from a cholinergic crisis (as a result of an overdosage of the drug). However, this used to be a problem in the era before immunosuppressants were in use, when some patients received daily doses of up to 2000 mg of pyridostigmine. Nowadays, the danger of a cholinergic crisis appears to be an overestimated problem, and in our and others' [74] experience a clinically relevant overdosage is nearly exclusively found in patients with severe long lasting myasthenia and a known hypersensitivity to the drug. Most patients will require corticosteroids at some stage of the disease. Corticosteroids suppress autoantibody production [26] and specific T lymphocyte responses [79] and lead to improvement of muscle strength within several days or a few weeks, normally reaching its maximum effect within three months [80]. Since a positive treatment response is seen in 70-90% of the patients [74, 81] corticosteroids are the most reliable drugs for myasthenia gravis, prednisolone or prednisone being used most often. Nevertheless a permanent remission can be induced in only about 10% [37, 80]. The remaining patients require long-term steroid medication or additional immunosuppressants. Considering the potentially severe steroid side effects the treatment schedule should thus be carefully planned and regularly reevaluated. The maximum dose in generalized MG may be up to 75 mg prednisolone (in children 1.5 mg/kg). In milder cases 40 to 50 mg may be sufficient. To begin with, one should not give the full dose, but rather gradually increase by 10 to 15 mg every other day, especially in patients with bulbar and respiratory symptoms. This is necessary, since corticosteroids by themselves impair neuromuscular transmission by decreasing the presynaptic release of acetylcholine, possibly leading to severe exacerbations of symptoms [46, 82]. The peak dose is given for several weeks until improvement is stable and may then be slowly tapered off. Dose reduction had better not be faster than 5 mg in eight weeks. Too fast a reduction of steroids is the most

321 common mistake leading to a relapse [11, 83]. At a daily dose of 40-50 mg an alternate day regimen may be introduced, to limit side effects and adrenal insufficiency [84, 85]. In severe cases we institute other immunosuppressive drugs from the beginning, in milder disease it is started, when steroid treatment cannot induce clinical remission within a few months [82].

Azathioprine was first used in MG in the late sixties [86, 87] and is now the commonly used long-term immunosuppressive agent [88]. Azathioprine and its metabolic products inhibit T and B lymphocyte proliferation by blocking intracellular enzymes of purine metabolisms [27, 89]. It is given at a dose of 2.0-2.5 mg/kg and needs to be taken for several (three to six) months until a clinical response can be expected. Therefore therapy should last for a minimum of two years, to show its full potency. The reliable long term efficiency and the relatively rare toxic effects have made it one of the cornerstones of current therapy in myasthenia. In more severe MG it may help to reduce the steroid dose on the long run; in patients with comparably mild MG and contraindications for steroids it may be used as single therapy. Severe: side effects are leucopenia and impairment of liver function, both of which may require dose reduction or interruption of the medication [90]. A small percentage (probably 5%) of patients show acute gastrointestinal side effects and do not tolerate the drug. Other immunosuppressive agents, above all Cyclosporin A [25, 91, 92], are also of proven benefit in MG; however, due to their greater toxicity they are often restricted to patients that do not tolerate azathioprine.

Thymectomy was one of the first successful therapies in MG [93, 94], though a prospective randomized study formally proving its benefit has never been performed. From retrospective [95, 96] and case-control studies [97] it is clear that thymectomy leads to clinical improvement in 70-80% of the patients, with approximately 35% reaching complete remission. Success rate is higher if the operation is performed within the first year after disease onset [98, 99], in younger patients, and if there is no thymoma, but rather thymic hyperplasia [ll, 74]. It may take up to three years until the full effect is clinically apparent [99]. Indeed 75% of patients with generalized MG have thymic alterations: 10% have thymoma, and 65% have follicular thymic hyperplasia with germinal centres, the remaining 25% showing thymic atrophy [5, 100, 101]. Therefore the indication to thymectomy should rely on the pathological changes to be expected (see Table 1). In our centre thymectomy is performed in patients with generalized MG with disease onset before the age of 45(-50) (group A in Table 1) and in thymoma patients (group C). By contrast, we (and most other centres) do not thymectomize patients with a disease onset later in life (group B) unless there is evidence of thymoma, and patients with purely ocular MG (group D) [102]. (For discussion of

322 thymectomy in ocular MG see below.) Perioperative mortality is less than 1%, late sequelae of the operation are negligible.

Plasmapheresis is restricted to severely ill patients, often with respiratory insufficiency, to overcome a critical period until immunosuppressants act or before thymectomy [103, 104]. Normally, a series of three or six plasmaexchange procedures is performed, the clinical effect starting within two days lasting for about three weeks. Severe side effects, such as thromboembolic complications, bleedings, or anaphylactoid reactions are below 1% [1051. Other therapies, such as high-dose immunoglobins [106, 107], or splenectomy [108], have been advocated or are currently under study, but have not been able to replace the well established treatment measures outlined above. The hope for a specific immunotherapy- either by blocking autoantibody producing B lymphocytes or suppressing autoreactive T lymphocytes - has been maintained on the basis of promising experimental results, but such therapy is not available so far in humans. Apart from drug therapy simple aids will help the patients to cope with everyday problems. Many will find that wearing dark glasses reduces the discomfort of diplopia. In longstanding stable eye muscle weakness prismatic correction is also possible. Emotional stress, and systemic illness like viral infections should be avoided as far as possible. Drugs that impair neuromuscular transmission may also have adverse effects in MG. The physician as well as the patient must be aware that any new medication may worsen myasthenic symptoms and every patient should be regularly observed when a new medication is begun. Large charts exist in nearly every textbook listing drugs with known negative effects and giving possible alternatives, but they are based on few examples and soon outdated. We do not recommend to stick too closely to those recommendations, but rather follow the suggestion of Sanders and Howard's [37] that a notice should be placed on medical charts of MG patients that warn of the most deleterious drugs (Table 4). All other drugs will cause tolerable or no adverse effect in adequately treated and stable myasthenia gravis. Table 4. Drugs that should be strictly avoided in patients with myasthenia gravis*

1. 2. 3. 4. 5.

Muscle relaxants (pancuronium, succinylcholine etc.) Aminoglycosides Certain antiarrythmics (quinine, quinidine, procainamid) Magnesium D-penicillamin

* D-peniciUamin should never be used in MG, the others only if absolutely necessary. Any other new medication may theoretically worsen myasthenic symptoms and every patient should be regularly observed when a new medication is begun.

323 D-penicillamine is the only drug that should be absolutely avoided, since it may induce an autoimmune form of MG [109]. During pregnancy symptoms may improve, worsen or not change [110]. First trimester worsening is often observed in primagravida pregnancies, third trimester or postpartum exacerbations often in subsequent pregnancies. Due to the good prognosis of MG 'therapeutic' abortion is definitely not indicated. Anti-cholinesterase agents should not be given intravenously, because they may cause uterine contraction. Steroids have no reported side effects in pregnancy, except for a mildly increased incidence of cleft lippalate. Cytotoxic drugs should be avoided and especially in the first trimester be restricted to crises unresponsive to other therapies. Labor and delivery are usually normal, Cesarean section is indicated only for obstetric reasons. Drugs containing magnesium should be avoided, if possible (Table 4). One in six to eight newborn babies suffers from transient neonatal myasthenia due to transferred IgG autoantibodies and might sometimes require temporary respiratory assistance. This syndrome responds to anticholinesterase drugs and its incidence and severity seem to correlate with the antibody titer in the umbilical cord. [111].

4. Management of ocular myasthenia The management of a chronic disease has to be judged by its spontaneous course. In patients with purely ocular myasthenia side effects of invasive therapies might outweigh a relatively mild discomfort; on the other hand, persistent diplopia does cause considerable disturbance of every day life, and mild myastenia symptoms may progress to severe generalized weakness, if not treated in time. Several studies have shown that at least half of the patients with ocular myasthenia will develop generalized symptoms; the risk of doing so, however, is greatest within the first two years and decreases markedly with time. In a study of 108 patients with ocular MG, 53 patients (49%) later generalized, but only 9 did so more than 2 years after onset [112]; the median time to generalization was less than 1 year. In an older series of patients that were managed without steroid or immunosuppressive treatment [7], 11 of 35 (31%) ocular patients remained purely ocular, whereas 21 generalized within 2 years after onset. Only 3 patients developed generalized symptoms between 10 and 22 years after onset [7]. Grob reported that approximately 40% of his patients were initially ocular, of those as many as 40% (i.e, 16% of all) would always remain ocular [113]. The maximum severity of the disease will be reached within the first three years. After one year of ocular myasthenia there is an 84% likelihood to remain ocular [113]. Drug treatment of ocular myasthenia relies on the basic principles outlined above, but certain modifications have to be made. Practically, pyridostigmine monotherapy is started at 20-30 mg three to four times daily,

324 and gradually increased until the symptoms disappear, but normally not above 240 mg. If a complete or satisfactory response is achieved additional treatment may not be necessary for the time being. Remarkably, eye muscles do not respond so well to pyridostigmine as other muscles do for reasons that have not fully been understood, and most patients will require steroids at some point. A good but not absolute prediction of the pyridostigmine response can be made from the Tensilon test. For mild symptoms, like a droopy eyelid or occasional double vision under physical exercise we recommend a dose of 20-30 mg prednisolone on alternate days, either alone or in combination with pyridostigmine. This dose will often lead to considerable improvement after four to six weeks. For more severe ocular symptoms 50-60 mg prednisolone daily for up to three months may be required until improvement is detectable [74]. The dose reduction scheme should be similar as outlined in Section 3. If symptoms do not remit completely under steroid treatment further immunosuppressive drugs are required. Indeed a few patients of ours needed to be treated with azathioprine (2-2.5 mg per kg, see above) for five to six months plus steroids until improvement of diplopia occurred. In most neurologists' opinion thymectomy is not a standard treatment for ocular myasthenia [102]. Despite some preliminary evidence for a positive effect [114], thymectomy does not seem to be appropriate considering the relatively good prognosis of medically treated ocular MG. However, this issue has not been settled and further investigations might change our attitude. At present we recommend that thymectomy in ocular MG should be restricted to patients with suspected thymoma. Plasmapheresis is only done in severe generalized MG and is no standard treatment for ocular myasthenia.

5. Pathogenetic concepts 5.1 Cornerstones of myasthenia research Myasthenia gravis (MG) was already recognized as a disease entity in the last century [115], when Jolly first used the term 'Myasthenia' and applied faradic nerve stimulation to study muscle fatigability. Only in 1960, however, did Simpson propose the autoimmune origin of the disease [116]. His assumption was based on purely clinical and empirical grounds, namely the association with other known autoimmune diseases (e.g., systemic lupus erythematosus), abnormalities of the immune system including thymic changes, and the possible transfer of transient myasthenia from mothers to newborn babies. A chance finding, as it were, brought the acetylcholine receptor onto the playing-field. Patrick and Lindstrom immunised rabbits with AChR in order to produce antisera for further biochemical characterization of the protein,

325 and discovered that the animals developed myasthenia-like muscle weakness that was reversible with anti-cholinesterase drugs [117]. Two 'facts of life' were crucial in becoming the AChR a rewarding research subject. First, snake toxins, such as alpha-bungarotoxin from the Asian krait Bungarus multicinctus, bind specifically to the AChR and are used to purify and quantify receptor protein by relatively simple biochemical procedures [118]. Secondly, large amounts of AChRs are present in electric organs of some fish, e.g. Torpedo marmorata, and provide a rich source of the protein, whereas other tissues like mammalian muscle contain only small amounts of AChR [119]. Today the fine structure of the AChR is known to consist of five structurally similar subunits (2 a, fl, r, 6) which form a transmembrane ion channel; there is great interspecies homology, e.g. the human and Torpedo receptor share up to 80% of their amino acid sequence in certain regions [120]. Based on the growing biochemical knowledge it was recognized that the majority of patients with myasthenia gravis have circulating serum antibodies against the AChR [121]. Definite proof of the autoimmune hypothesis was provided by Toyka and Drachman who showed that passive transfer to IgG fractions containing these human autoantibodies could induce disease symptoms in mice [122]. At the same time in the mid-seventies it was demonstrated by NewsomDavis and coworkers, that removal of the pathogenic serum antibodies from patients by plasmapheresis was an efficient therapy [123].

5.2 Antibody action at the neuromuscular junction The action of autoantibodies at the neuromuscular junction is now better understood [124-127] and several different mechanisms are likely to lead to the reduction of functional receptors: (a) complement-mediated lysis and destruction of the postsynaptic membrane, (b) antibody-driven cross-linking and accelerated degradation of acetylcholine receptors, and (c) direct blocking of the transmitter binding site, of which the latter is probably the least important. A yet unsolved problem is the proportion of 'seronegative' patients, i.e. myasthenics without detectable anti-AChR antibodies. In seronegative ocular myasthenia the amount of circulating antibodies may be too small to be detected by the highly sensitive assay; or else all of the autoantibodies produced may have bound to endplate receptors [17]. It may also be relevant that patients with ocular myasthenia have slightly different antibody fine specificities and their sera bind better to ocular muscle than to the routinely used leg muscle [128]. Also, sera from long-standing ocular cases that were antibody-negative in the routine assay showed motor endplate staining of extraocular muscle with an immunofluorescence method [129]. In generalized MG approximately 10% of the patients are seronegative. It has been speculated tfiat other antigens structurally linked to the AChR may be involved. In fact, passive transfer of serum from these patients induces

326 impaired neuromuscular transmission in mice and AChR-antibody negative patients may respond clinically to plasmapheresis [130[. Moreover, there is recent experimental evidence that an IgM serum factor in these patients may contribute to muscle weakness [131].

5.3 Antibodies, T lymphocytes and the thymus The interaction of basic immunological mechanisms has been the most recent issue in myasthenia research. Since the circulating autoantibodies are of IgG type and therefore T cell dependent, AChR-reactive T cells have been thoroughly investigated [132]. It has been demonstrated the T cell fine specificities to AChR epitopes are similarly heterogeneous as the patients' antibody specificities [133-136]. In fact, AChR reactive T cells could not only be found in the blood from myasthenics, but also from healthy individuals- though not quite as common [135-136[. It was concluded that the mere presence of autoreactive T lymphocytes does not cause the autoimmune disease, but other regulatory mechanisms must also be involved. Studying thymic changes in MG has given some important clues that may eventually help to clarify the complex pathogenesis. There are two different types of thymic pathology found in MG: thymic hyperplasia and thymoma. Thymic hyperplasia is in fact a misleading term and does not necessarily stand for a macroscopic enlargement of the gland, but rather for a typical histological pattern. It is more correctly described as lympho-follicular hyperplasia, meaning the presence of secondary lymph node-like follicles (so-called germinal centres) in the thymic medulla surrounded by T cell areas. These peculiarities are to a much lesser degree also found in patients with other autoimmune diseases and healthy individuals; however, they are present in virtually all patients with generalized MG, onset of symptoms under 45 years, and without previous immunosuppressive treatment [5,137]. Thymoma is found in about 10% of MG patients, but from all thymoma patients as many as 30-50% develop MG. Extensive immunohistological studies have led to the assumption that most thymoma patients are at risk of developing MG, because of a common phenotypical cell pattern in MG and non-MG associated thymoma [101]. The myasthenic thymus is enriched in AChR-reactive lymphocytes, no matter whether it is hyperplastic or a thymoma [138, 139]. In thymoma patients the origin of misdirected T lymphocyte activation seems to happen in the tumour itself; in the hyperplastic thymus, however, a more unspecific thymitis - involving T and B lymphocytes of various specificities - probably leads to a local selection of AChR-reactive cells subsequently causing the autoimmune reaction [139[. Clearly, these recent findings may apply to the majority of patients with thymic pathology. Those with thymic atrophy, or other rarer (e.g., Dpenicillamine induced) forms of autoimmune-mediated myasthenia are not

327 e x p l a i n e d by this model. W h e t h e r p a t i e n t s with p u r e l y o c u l a r m y a s t h e n i a gravis have to be c o n s i d e r e d a s e p a r a t e s u b g r o u p for p a t h o g e n i c r e a s o n s , or n e e d to be d i s t r i b u t e d into o n e of the o t h e r categories (according to T a b l e 1) r e m a i n s to b e established.

Acknowledgements We t h a n k Prof. V. H e r z a u for critical review of the m a n u s c r i p t , a n d Mrs G. Sch6nw/ilder a n d Mrs K.-S. Thies for c o n t i n u o u s assistance a n d technical help.

References 1. Somnier FE, Keiding N, Paulson OB. Epidemiology of myasthenia gravis in Denmark: A longitudinal and comprehensive population survey. Arch Neurol 1991; 48: 733-39. 2. Tola MR, Granieri E, Paolino E, Caniatti L, Quatrale R, Mazzanti B, D'Allessandro RD. Epidemiological study of myasthenia gravis in the province of Ferrara, Italy. J Neurol 1989; 236: 388-90. 3. Storm-Mathisen A. Epidemiology of myasthenia gravis in Norway. Acta Neurol Scand 1984; 70: 274-84. 4. Yu YL, Hawkins BR, Ip MSM, Wong V, Woo E. Myasthenia gravis in Hong Kong Chinese, 1: Epidemiology and adult disease. Acta Neurol Scand 1992; 86: 113-19. 5. Compston DAS, Vincent A, Newsom-Davis J, Batchelor JR. Clinical, pathological HLA antigen and immunological evidence for disease heterogeneity in myasthenia gravis. Brain 1980; 103: 579-601. 6. Newsom-Davis J. Autoimmunity in neuromuscular disease. Ann NY Acad Sci 1988; 540: 25-38. 7. Oosterhuis HJGH. The natural course of myasthenia gravis: A long term follow up study. J Neurol Neurosurg Psychiatry 1989; 52: 1121-27. 8. Grob D, Arsura EL, Brunner NG, Namba T. The course of myasthenia gravis and therapies affecting outcome Ann NY Acad Sci 1987; 505: 472-99. 9. Mertens HG, Balzereit F, Leipert M. The treatment of severe myasthenia gravis with immunosuppressive agents. Eur Neurol 1969; 2: 321-39. 10. Osserman KE, Genkins G. Studies in myasthenia gravis: Review of a twenty years experience in over 1200 patients. Mt Sinai J Med 1971; 38: 497-537. 11. Newsom-Davis J. Diseases of the neuromuscular junction. In: Asbury AK, McKhann GM, McDonald WI, eds. Diseases of the nervous system. London: Heinemann, 1992: 197-212. 12. Willcox N, Schluep M, Ritter MA, Newsom-Davis J. The thymus in seronegative myasthenia gravis patients. J Neurol 1991; 238: 256-61. 13. Rowland LP. Prostigmine-responsiveness and the diagnosis of myasthenis gravis Neurology 1955; 5: 612-24. 14. Seybold ME. The office Tensilon test for ocular myasthenia gravis. Arch Neurol 1986; 43: 842-43. 15. Daroff RB. The office Tensilon test for ocular myasthenia gravis. Arch Neurol 1986; 43: 843-44. 16, Taylor P. Anticholinesterase agents. In: Gilman AG, Rail TW, Nies AS, Taylor P, eds. Goodman and Gilman's the pharmacological basis of therapeutics. New York: Pergamon Press, 1990: 131-49.

328 17. Vincent A, Newsom-Davis J. Acetylcholine receptor antibody as a diagnostic test for myasthenia gravis: Results in 153 validated cases and 2967 diagnostic assays. J Neurol Neurosurg Psychiatry 1985; 48: 1246-52. 18. Lefvert AK, Bergstrom K, Matell G, Osterman PO, Pirskanen R. Determination of acetylcholine receptor antibody in myasthenia gravis: Clinical usefulness and pathogenic implications. J Neurol Neurosurg Psychiatry 1978; 41: 394-403. 19. Oda K, Goto I, Kuriowa Y, Onoue K, Ito Y. Myasthenia gravis: Antibodies to acetylcholine receptor with human and rat antigens. Neurology 1980; 30: 543-46. 20. Tindall RSA. Humoral immunity in myasthenia gravis: Biochemical characterization of acquired antireceptor antibodies and clinical correlations. Ann Neurol 1981; 10: 437-47. 21. Limburg PC, The TH, Hummel-Tappel E, Oosterhuis HJGH. Anti-acetylcholine receptor antibodies in myasthenia gravis, Part 1: Relation to clinical parameters in 250 patients. J Neurol Sci 1983; 58: 357-70. 22. Toyka KV, Heininger K. Acetylcholine-receptor antibodies in the diagnosis of myasthenia gravis. Dtsch Med Wochenschr 1986; 111: 1435-39. 23. Vincent A, Newsom-Davis J, Newton P, Beck N. Acetylcholine receptor antibody and clinical response to thymectomy in myasthenis gravis. Neurology 1983; 33: 1276-82. 24. Schumm F, Fateh-Moghadam A, Dichgans J. Correlation between serum-anti-acetylcholine receptor antibody levels and clinical state under immunosuppressive therapy in myasthenia gravis. Eur Arch Psychiat Neurol Sci 1984; 234: 224-30. 25. Tindall RSA, Rollins JA, Phillips JT, Greenlee RG, Wells L, Belendiuk G, Preliminary results of a double-blind, randomized, placebo-controlled trial of cyclosporine in myasthenia gravis. N Engl J Med 1987; 316: 719-24. 26. Tindall RSA. Humoral immunity in myasthenia gravis: Effect of steroids and thymectomy. Neurology 1980; 30: 554-57. 27. Hohlfeld R, Toyka KV, Besinger UA, Gerhold B, Heininger K. Myasthenia gravis: Reactivation of clinical disease and of autoimmune factors after discontinuation of long-term azathioprine. Ann Neurol 1985; 17: 238-42. 28. Cu6noud S, Feltkamp TEW, Fulpius BE, Oosterhuis HJGH. Antibodies to acetylcholine receptor in patients with thymoma but without myasthenia gravis. Neurology 1980; 30: 201-3. 29. Mittag TW, Caroscio J. False-positive immunoassay for acetylcholine-receptor antibody in amyotrophic lateral sclerosis. N Engl J Med 1980; 302: 868. 30. Liebermann JA, Bradley RJ, Rubinstein M, Kane JM. Antibodies to acetylcholine receptors in tardive dyskinesia. Lancet 1984 (i): 1066. 31. Behan PO, Behan WMH, Grant R. Immunological aspects of myotonic dystrophy. In: Aarli JA, Behan WMH, Behan PO, eds. Clinical neuroimmunology. Oxford: Blackwell, 1987: 360-69. 32. Vincent A. Conclusions of anti-AChR workshop at Maasticht, June 1987. Monogr Allergy (Basel: Karger) 1988; 25: 150-53. 33. Lindstrom J, Campbell M, Nave B. Specificities of antibodies to acetylcholine receptors. Muscle Nerve 1978; 1: 140-45. 34. Voltz R, Hohlfeld R, Fateh-Moghadam A, Witt TN, Wick M, Reimers C, Siegele B, Wekerle H. Myasthenia gravis: Measurement of anti-AChR autoantibodies using cell line TE671. Neurology 1991; 41: 1836-38. 35. Stalberg E. Clinical electrophysiology in myasthenia gravis. J Neurol Neurosurg Psychiatry 1980; 43: 622-33. 36. Sanders DB. The electrodiagnosis of myasthenia gravis. Ann NY Acad Sci 1987; 505: 539-56. 37. Sanders DB, Howard JF. Disorders of neuromuscular transmission. In: Bradley WG, Daroff RB, Fenichel GM, Marsden CD, eds. Neurology in clinical practice. Boston: Butterworth-Heinemann, 1991: 1819-42. 38. Schumm F, St6hr M. Accessory nerve stimulation in the assessment of myasthenia gravis. Muscle Nerve 1984; 7: 147-51.

329 39. Vial C, Charles N, Chauplannaz G, Bady B. Myasthenia gravis in childhood and infancy: Usefulness of electrophysiologic studies. Arch Neurol 1991; 48: 847-49. 40. Sanders DB, Howard JF. Single-fiber electromyography in myasthenia gravis. Muscle Nerve 1986; 9: 809-19. 41. Massey JM, Sanders DB, Howard JF. The effect of cholinesterase inhibitors of SFEMG in myasthenia gravis. Muscle Nerve 1989; 12: 154-55. 42. Newsom-Davis J. Autoimmune neurological disorders. In: Brostoff J, Scadding GK, Male D, Roitt IM, eds. Clinical immunology. London: Gower Medical Publishing, 1991: 9.10-9.16. 43. Lanska DJ. Diagnosis of thymoma in myasthenics using anti-striated muscle antibodies: Predictive value and gain in diagnostic certainty. Neurology 1991; 41: 520-24. 44. Monden Y, Fujii Y, Masaoka A. Clinical characteristics of myasthenia gravis with other autoimmune diseases. Ann NY Acad Sci 1987; 505: 876-78. 45. Keane JR. Vertical diplopia. Seminars in Neurology 1986; 6: 147-54. 46. Oosterhuis HJGH. Myasthenia gravis: Clinical neurology and neurosurgery monographs. Edinburgh: Churchill Livingstone, 1984. 47. Finalli PF, Hoyt WF. Myasthenic abduction nyastagmus in a patient with hyperthyroidism. Neurology 1976; 26: 589-90. 48. Berlit E Isolated and combined pareses of cranial nerves, 1II, IV, and VI: A retrospective study of 412 patients. J Neurol Sci 1991; 103: 10-15. 49. Gorelick PB, Rosenberg M, Pagano RJ. Enhanced ptosis in myasthenia gravis. Arch Neurol 1981; 38: 531. 50. Gorelick PB. Office Tensilon test. Arch Neurol 1987; 44: 689-90. 51. Sethi KD, Rivner MH, Seift TR. Ice pack test for myasthenia gravis. Neurology 1987; 37: 1383-85. 52. Odel JG, Winterkorn JMS, Behrens MM. The sleep test for myasthenia gravis: A safe alternative to Tensilon. J Clin Neuroophthalmol 1991; 11: 288-92. 53. Cogan DG. Myastenia gravis: A review of the disease and a description of lid twitch as a characteristic sign. Arch Ophthalmol 1965; 74: 217-21. 54. Glaser JS. Myasthenic pseudo-internuclear ophthatmoplegia. Arch Ophthalmol 1966; 75: 363-66. 55. Davis TL, Lavin PJ. Pseudo one-and-a-half syndrome with ocular myasthenia. Neurology 1989; 39: 1553. 56. Ragge NK, Hoyt WF. Midbrain myasthenia: Fatigable ptosis, 'lid twitch' sign, and ophthalmoparesis from a dorsal midbrain glioma. Neurology 1992; 42: 917-19. 57~ Shinton RA, Jamieson DG. Exercise induced diplopia as a presentation of midline cerebral tumour. J Neurol Neurosurg Psychiatry 1989; 52: 916-17. 58. Lepore FE, Sanborn GE, Slevin JT. Pupillary dysfunction in myasthenia gravis. Ann Neurol 1979; 6: 29-33. 59. Schmidt D, Dell'Osso LF, Abel LA, Daroff RB. Myasthenia gravis: Saccadic eye movement waveforms. Exp Neurol 1980; 68: 346-64. 60. Kaminski H J, Maas E, Spiegel P, Ruff RL. Why are eye muscles frequently involved in myasthenia gravis? Neurology 1990; 40: 1663-69. 61. Yee RD, Cogan DG, Zee DS, Baloh RW, Honrubia V. Rapid eye movements in myasthania gravis. Arch Ophthalmol 1976; 94: 1465-72. 62. Sollberger CE, Meienberg O, Ludin HP. The contribution of oculography to early diagnosis of myasthenia gravis: A study of saccadic eye movements using the infrared reflection method in 22 cases. Eur Arch Psychiatry Neurol Sci 1986; 236: 102-8. 63. Leigh RJ, Zee DS. The neurology of eye movements, 2nd ed Philadelphia: Davis, 1991: 338-77. 64. Oohira A, Goto K, Sato Y. Saccades of supernormal velocity: Adaptive response to ophthalmoplegia in a patient with myasthenia gravis. Neuro-Ophthalmol 1987; 7: 203-9. 65. Newsom-Davis J. Edrophonium responsiveness not necessarily diagnostic of myasthenia gravis. Muscle Nerve 1990; 13: 1186.

330 66. Evoli A, Tonali P, Bartoccioni E, Lo Monaco. Ocular myasthenia: Diagnostic and therapeutic problems. Acta Neurol Scand 1988; 77: 31-35. 67. Kelly JJ, Daube JR, Lennon VA, Howard FM; Younge BR. The laboratory diagnosis of mild myasthenia gravis. Ann Neurol 1982; 12: 238-42. 68. Moorthy G, Behrens MM, Drachman DB, Kirkham TH, Knox DL, Miller NR, Slamovitz TL, Zinreich SJ. Ocular pseudomyasthenia or ocular myasthenia 'plus': A warning to clinicians. Neurology 1989; 39: 1150-54. 69. Retzlaff JA, Kearns TP, Howard FM. Lancaster red-green test in evaluation of edrophonium effect in myasthenia gravis. Am J Ophthalmol 1969; 67: 13-21. 70. Weller M, Wiedemann P. Hysterical symptoms in ophthalmology. Doc Ophthalmol 1989; 73: 1-33. 71. Soliven BC, Lange DJ, Penn AS, Younger D, Jaretzki A, Lovelace RE, Rowland LP. Seronegative myasthenia gravis. Neurology 1988; 38: 514-17. 72. Sommer N, Weller M, Petersen D, Wieth61ter H, Dichgans J. Neurosarcoidosis without systemic sarcoidosis. Eur Arch Psychiatry Clin Neurosci 1991; 240: 334-38. 73. Bischoff C, Klingelh6fer J, Conrad B. Decay and recovery of the stapedial reflex by prolonged stimulation in the diagnosis of myasthenia gravis. J Neurol 1989; 236: 343-48. 74. Rowland LP. Controversies about the treatment of myasthenia gravis. J Neurol Neurosurg Psychiatry 1980; 43: 644-59. 75. Breyer-Pfaff U, Schmezer A, Maier U, Brinkmann A, Schumm F. Neuromuscular function and plasma drug levels in pyridostigmine treatment of myasthenia gravis. J Neurol Neurosurg Psychiatry 1990; 53: 502-6. 76. Schumm F, Gaertner HJ, Wiatr G, Dichgans J. Plasma levels of pyridostigmine in myasthenia gravis: Method and clinical significance. Fortschr Neurol Psychiat 1985; 53: 201-11. 77. White MC, DeSilva P, Harvard CWH. Plasma pyridostigmine levels in myasthenia gravis. Neurology 1981; 31: 145-50. 78. Genkins G, Kornfeld P. Papatestas AE, Bender AN, Matta RJ. Clinical experience in more than 2000 patients with myasthenia gravis. Ann NY Acad Sci 1987; 505: 500-514. 79. Abramsky O, Aharonov A, Teitelbaum D, Fuchs S. Myasthenia gravis and acetylcholine receptor. Arch Neurol 1975; 32: 684-87. 80. Johns TR. Long-term corticosteroid treatment of myasthenia gravis. Ann NY Acad Sci 1987; 505: 568-83. 81. Cornelio F, Peluchetti D, Mantegazza R, Sghirlanzoni A, Collarile C. The course of myasthenia gravis in patients treated with costicosteroids, azathioprine, and plasmapheresis. Ann NY Acad Sci 1987; 505: 517-25. 82. Argov Z, Mastaglia FL. Disorders of neuromuscular transmission caused by drugs. N Engl J Med 1979; 301: 409-13. 83. Miano MA, Bosley TM, Heiman-Patterson TD, Reed J, Sergott RC, Savino PJ, Schatz NJ. Factors influencing outcome of prednisone dose reduction in myasthenia gravis. Neurology 1991; 41: 919-21. 84. Ackerman GL, Nolan CM. Adrenocortical responsiveness after alternate-day corticosteroids therapy. N Engl J Med 1968; 278: 405-9. 85. Byyny RL. Withdrawal from corticosteroid therapy. N Engl J Med 1976; 295: 30-32. 86. Delwaide PJ, Salomon J, van Cauwenberge H, Premiere essai de traitement de la myasthenie par azathioprine. Acta Neurol Belg 1967; 67: 701-10. 87. Mertens HG, Balzereit F, Leipert M. The treatment of severe myasthenia gravis with immunosuppressive agents. Eur Neurol 1969; 2: 321-39. 88. Matell G. Immunosuppressive drugs: Azathioprine in the treatment of myasthenia gravis. Ann NY Acad Sci 1987; 505: 588-94. 89. Ten Berge RJM, Schellekens PTA, Surachno S, The TH, ten Veen JH, Wilmink JM. The influence of therapy with azathioprine and prednisone on the immune system of kidney transplant recipients. Clin Immunol Immunopathol 1981; 21: 20-32. 90. Hohlfeld R, Michels M, Heininger K, Besinger U, Toyka KV. Azathioprine toxicity

331

91.

92. 93.

94.

95.

96. 97. 98.

99.

100. 101.

102. 103. 104.

105. 106.

107. 108. 109.

110. 111.

during long-term immunosuppression of generalized myasthenia gravis. Neurology 1988; 38: 258-61. Schalke B, Kappos L, Dommasch D, Rohrbach E, Mertens HG. Cyclosporin-A treatment of myasthenia gravis: Initial results of a double-blind trial of cyclosporin-A versus azathioprine. Ann NY Acad Sci 1987; 505: 872-75. Perez MC, Buot WL, Mercado-Danguilan C, Bagabaldo ZG, Renales LD. Stable remissions in myasthenia gravis. Neurology 1981; 31: 32-37. Schumacher CH, Roth P. Thymektomie bei einem Fall von Morbus Basedowi mit Myasthenie. Mitteilungen aus den Grenzgebieten der Medizin und Chirurgie 1912; 25: 746-65. Blalock A, Mason MF, Morgan HJ, Riven SS. Myasthenia gravis and tumors of the thymic region: Report of a case in which the tumor was removed. Ann Surg 1939; 100: 544-60. Buckingham JM, Howard FM, Bernatz PE, Payne WS, Harrison EG, O'Brien PC, Weiland LH. The value of thymectomy in myasthenia gravis: A computer-assisted matched study. Ann Surg 1976; 184: 453-57. Rodriquez M, Gomez MR, Howard FM, Taylor WF. Myasthenia gravis in children: Long-term follow-up. Ann Neurol 1983; 13: 504-10. Olanow CW, Wechsler AS, Roses AD. A prospective study of thymectomy and serum acetylcholine receptor antibodies in myasthenia gravis. Ann Surg 1982; 196: 113-21. Hankins JR, Mayer RF, Satterfield JR, Turney SZ, Attar S, Sequeira AJ, Thompson BW, McLaughlin JS. Thymectomy for myasthenia gravis: A 14-year experience. Ann Surg 1985; 201: 618-25. Monden Y, Nakahara K, Kagotani K, Fujii Y, Nanjo S, Masaoka A, Kawashima A. Effects of preoperative duration of symptoms on patients with myasthenia gravis. Ann Thorac Surg 1984; 38: 287-91. Rosai J, Levine GD. Tumors of the thymus. Atlas of tumor pathology, 2nd series, fascicle 13. Washington DC: Armed Forces Institute of Pathology, 1976. Willcox N, Schluep M, Ritter MA, Schuurman HJ, Newsom-Davis J, Christensson B. Myasthenic and nonmyasthenic thymoma. An expansion of a minor cortical epithelial subset? Am J Pathol 1987; 127: 447-60. Lanska DJ. Indications for thymectomy in myasthenia gravis. Neurology 1990; 40: 1828-29. Newsom-Davis J, Wilson SG, Vincent A, Ward CD. Long-term effects of repeated plasma exchange in myasthenia graves. Lancet 1979 (i): 464-468. Newsom-Davis J, Pinching AJ, Vincent A, Wilson SG. Function of circulating antibody to acetylcholine receptor in myasthenia gravis: Investigation by plasma exchange. Neurology 1978; 28: 266-72. Toyka KV. Neurologische Indikationen zur Plasmapherese. Akt Neurol 1984; 11: 114-17. Gajdos P, Outin HD, Morel E, Raphael JC, Goulon M. High-dose intravenous gamma globulin for myasthenia gravis: An alternative for plasma exchange? Ann NY Acad Sci 1987; 505: 842-44. Cosi V, Lombardi M, Piccolo G, Erbetta A. Treatment of myasthenia gravis with high-dose intravenous immunoglobulin. Acta Neurol Scand 1991; 84: 81-84. Hofmann WE, Reuther P, Schalke B, Mertens HG. Splenectomy in myasthenia gravis: A therapeutic concept? J Neurol 1985; 232: 215-18. Vincent A, Newsom-Davis J. Acetylcholine receptor antibody characteristics in myasthenia gravis, II: Patients with penicillamine-induced myasthenia or idiopathic myasthenia of recent onset. Clin exp Immunol I982; 49: 266-72. Plauch6 WC. Myasthenia gravis in pregnancy: An update. Am J Obstet Gynecol 1979; 135: 691-97. Morel E, Eymard B. Immunological and clinical aspects of neonatal myasthenia gravis. Ann NY Acad Sci 1987; 505: 879-80.

332 112. Bever CT, Aquino AV, Penn AS, Lovelace RE, Rowland LIP. Prognosis of ocular myasthenia. Ann Neurol 1983; 14: 516-19. 113. Grob D. Clinical manifestation of myasthenia gravis. In: Albuquerque EX, Eldefravi AT, eds. Myasthenia gravis. London: Chapmann and Hall, 1983: 319-45. 114. Schumm F, WiethSlter H, Fateh-Moghadam A, Dichgans J. Thymectomy in myasthenia with pure ocular symptoms. J Neurol Neurosurg Psychiatry 1985; 48: 332-37. 115. Jolly F. Ueber Myasthenia gravis pseudoparalytica. Berl Klin Wschr 1895; 32: 1-7. 116. Simpson JA. Myasthenia gravis: A new hypothesis. Scot Med J 1960; 5: 419-36. 117. Patrick J, Lindstrom J. Autoimmune response to acetylcholine receptor. Science 1973; 180: 871-72. 118. Lindstrom J, Einarson B, Tzartos S. Production and assay of antibodies to acetylcholine receptors. Meth Enzymol 1981; 74: 432-60. 119. Vincent A. Immunology of acetylcholine receptors in relation to myasthenia gravis. Physiological Reviews 1980; 60: 756-824. 120. Noda M, Furutani Y, Takahashi H, Toyosato M, Tanabe T, Shimizu S, Kikyotani S, Kayano T, Hirose T, Inayama S, Numa S. Cloning and sequence analysis of calf cDNA and human genomic DNA encoding a-subunit precursor of muscle acetyleholine receptor. Nature 1983; 305: 818-23. 121. Lindstrom JM, Seybold ME, Lennon VA, Whittingham S, Duane DD. Antibody to acetylcholine receptor in myasthenia gravis. Neurology 1976; 26: 1054-59. 122. Toyka KV, Drachman DB, Griffin DE, Pestronk A, Winkelstein JA, Fischbeck KH, Kao I. Myasthenia gravis: Study of humoral immune mechanisms by passive transfer to mice. N Engl J Med 1977; 296: 125-31. 123. Pinching AJ, Peters DK, Newsom-Davis J. Remission of myasthenia gravis following plasma-exchange. Lancet 1976(ii): 1373-76. 124. Engel AG. Morphologic and immunopathologic findings in myasthenia gravis and in congenital myasthenic syndromes. J Neurol Neurosurg Psychiat 1980; 43: 577-89. 125. Engel AG, Lambert EH, Howard FM. Immune complexes (IgG and C3) at the motor endplate in myasthenia gravis: Ultrastructural and light microscopic localization and electrophysiological correlation. Mayo Clin Proc 1977; 52: 267-80. 126. Kao I, Drachman DB. Myasthenic immunoglobulin accelerates acetylcholine receptor degradation. Science 1977; 196: 527-29. 127. Drachman DB, De Silva S, Ramsay D, Pestronk A. Humoral pathogenesis of myasthenia gravis. Ann NY Acad Sci 1987; 505: 90-105. 128. Vincent A, Newsom-Davis J. Acetylcholine receptor antibody characteristics in myasthenia gravis I: Patients with generalized myasthenia or disease restricted to ocular muscles. Clin Exp Immunol 1982; 49: 257-65. 129. Oda K. Ocular myasthenia gravis: Antibodies to endplates of human extraocular muscle. Ann NY Acad Sci 1987; 505: 861-63. 130. Mossman S, Vincent A, Newsom-Davis J. Myasthenia gravis without acetylcholinereceptor antibody: A distinct clinical entity. Lancet 1986(i): 116-19. 131. Yamamoto T, Vincent A, Ciulla TA, Lang B, Johnston I, Newsom-Davis J. Seronegative myasthenia gravis: A plasma factor inhibiting agonist-induced acetylcholine receptor function copurifies with IgM. Ann Neurol 1991; 30: 550-57. 132. Newsom-Davis J, Harcourt G, Sommer N, Beeson D, Willcox N, Rothbard JB. T-cell reactivity in myasthenia gravis. J Autoimmunity 1989; 2(suppl): 101-8. 133. Hohlfeld R, Toyka KV, Heininger K, Grosse-Wilde H, Kalies I. Autoimmune human T lymphocytes specific for acetylcholine receptor. Nature 1984; 310: 244-46. 134. Harcourt GC, Sommer N, Rothbard J, Willcox HNA, Newsom-Davis J. A juxtamembrane epitope on the human acetylcholine receptor recognized by T cells in myasthenia gravis. J Clin Invest 1988; 82: 1295-1300. 135. Sommer N, Harcourt GC, Willcox N, Beeson D, Newsom-Davis J. Acetylcholine receptor-reactive T lymphocytes from healthy subjects and myasthenia gravis patients. Neurology 1991; 41: 1270-76.

333 136. Melms A, Malcharek G, Gem U, Wieth61ter H, Mfiller CA, Schoepfer R, Lindstrom J. T-cells from normal and myasthenic individuals recognize the human acetylcholine receptor: Heterogeneity of antigenic sites on the a-subunit. Ann Neurol 1992; 31: 311-18. 137. Willcox N, Schluep M, Sommer N, Campana D, Janossy G, Brown AN, Newsom-Davis J. Variable corticosteroid sensitivity of thymic cortex and medullary peripheral-type lymphoid tissue in myasthenia gravis patients: Structural and functional effects. Q J Med 1989; 271: 1071-87. 138. Melms A, Schalke BCG, Kirchner T, Miiller-Hermelink HK, Albert E, Wekerle H. Thymus in myasthenia gravis: Isolation of T-lymphocyte lines specific for the nicotinic acetylcholine-receptor from thymuses of myasthenic patients. J Clin Invest 1988; 81: 902 8. 139. Sommer N, Willcox N, Harcourt GC, Newsom-Davis J. Myasthenic thymus and thymoma are selectively enriched in acetylcholine receptor-reactive T cells. Ann Neurol 1990; 28: 312-19.

Address .for correspondence: Dr N. Sommer, Department of Neurology, Hoppe-Seyler-Str. 3, D-72076 T/ibingen, Germany Phone: (7071)292049; Fax: (7071)296 507