16ournal ofNeurology, Neurosurgery, and Psychiatry 1997;62:156-162

Ocular myasthenia gravis: response to long termimmunosuppressive treatment

Norbert Sommer, Barbara Sigg, Arthur Melms, Michael Weller, Karsten Schepelmann,Volker Herzau, Johannes Dichgans

Department ofNeurologyN SommerB SiggA MelmsM WellerK SchepelmannJ DichgansDepartment ofOphthalmology,Eberhard-Karls-University Tubingen,72076 Tubingen,GermanyV HerzauCorrespondence to:Dr N Sommer, Departmentof Neurology, Hoppe-Seyler-Str 3, D-72076 Tubingen,Germany.Received 6 March 1996and in revised form17 September 1996Accepted 17 September 1996

AbstractObjective-Ocular myasthenia gravis is asubtype of myasthenia gravis that causesrelatively mild disability, but may convertinto severe generalised muscle weakness.A universal management plan for ocularmyasthenia gravis has not been estab-lished. This study was performed todetermine the outcome of ocular myas-thenia gravis with the currently availabletherapeutic options.Methods-Retrospective analysis of 78patients with ocular myasthenia graviswith a mean disease duration of 8-3(range 0 5-58-3) years.Results-In 54 patients (69%) symptomsand signs remained confined to theextraocular muscles during the observa-tion period. The remaining 24 patients(31%) developed symptoms of generalisedmyasthenia gravis; 50% of them withintwo years, 75% within four years afteronset. A somewhat reduced risk of gener-alisation was found in those with mildsymptoms, normal repetitive nerve stim-ulation test, and low or absent antiacetyl-choline receptor (AChR) antibodies at thetime of diagnosis. Patients receivingimmunosuppressive treatment (corticos-teroids and/or azathioprine) rarely devel-oped generalised myasthenia gravis (sixof 50, 12%). Those without such treat-ment, usually due to uncertain diagnosisand late referral, converted into gener-alised myasthenia gravis significantlymore often (18 of 28, 64%).Conclusions-The prognosis of ocularmyasthenia gravis is good. A conventionalscheme with short term corticosteroidsand long term azathioprine seems ade-quate to achieve remission in mostpatients. The proportion of patientsdeveloping generalised myasthenia graviswas smaller in this population comparedwith previously published groups (usually500/or70%). Early immunosuppressivetreatment is at least partially responsiblefor this finding. Thymectomy (performedhere in 12 patients with an abnormalchest CT) also correlated with a good out-come, but had no apparent advantageover medical treatment alone.

(7 Neurol Neurosurg Psychiatry 1997;62:156-162)

Keywords: myasthenia gravis, extraocular muscleweakness, immunosuppressive treatment

Myasthenia gravis is caused by autoantibodiesagainst the acetylcholine receptor (AChR) atthe neuromuscular junction leading to exer-tional weakness of striated muscle. 1 2 3 Itsprevalence rate has been estimated to be up to10 per 100 000 in recent epidemiological sur-veys, and people of any race at any age may beaffected.45 Due to the currently available treat-ment options the prognosis of myastheniagravis has greatly improved during the past 20years. Anticholinesterase drugs have been usedsince the 1930s and in the 1960s corticos-teroids and other immunosuppressive drugsstarted to be used.3 67 Furthermore, the intro-duction of plasmapheresis, early thymectomy,and improved intensive care facilities hasreduced the mortality of myasthenia gravisessentially close to zero since the mid-1970s."'

Myasthenia gravis is one of the best studiedautoimmune diseases and the remarkableprogress in understanding the underlying aeti-ology and pathogenetic heterogeneity hashelped to improve patient management.According to age of onset, thymic abnormali-ties, and other immune variables patients withgeneralised myasthenia gravis can be dividedinto three main groups: (a) "young onset"patients (< 45 years, mainly female), who reg-ularly have thymic hyperplasia; (b) "old onset"patients (> 45 years, slightly more men), whonormally have thymic atrophy; and (c) thegroup of thymoma patients who have no clearage and sex bias.' 9 As not all patients withocular muscle weakness fit into one of thesecategories, there is often uncertainty about thetreatment options which are best in ocularmyasthenia gravis. Particularly, there is noclearcut guideline as to the need for thymec-tomy, mostly performed in groups (a) and (c),or long term immunosuppression, which maybe beneficial for all three patient groups withgeneralised disease.

Apart from symptomatic treatment withanticholinesterase drugs, corticosteroids areregularly considered in all forms of myastheniagravis, with good results in most patients. 10'2The indications for immunosuppressive drugs,such as azathioprine, cyclophosphamide, orcyclosporine are less clear. Although thesedrugs have clearly been shown to be beneficialin generalised myasthenia gravis,613'415 someneurologists prefer to limit their application tothe more severely affected patients.'6 Also,scepticism about the drug treatment is fos-tered by the fact that very few prospective dou-ble blind trials have been performed for anycompound. 1'-15 Thymectomy is the generallyaccepted treatment for patients with thymoma

156

on Decem

ber 31, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.62.2.156 on 1 February 1997. D

ownloaded from

Ocular myasthenia gravis: response to long term immunosuppressive treatment

and in young onset patients. This is based onimmunological evidence that thymic changesplay a central part in the pathogenesis of myas-thenia gravis, at least in these two patientgroups.'7-19 Uncertainty, however, exists as tothe value of thymectomy in older patients withthymic atrophy, patients with generalisedmyasthenia gravis that are negative for serumAchR antibodies, and also in purely ocular

2 20 21myasthenia gravis.In this study, we have reviewed patients

with purely ocular muscle weakness.

Patients and methodsAll patients included in this study were exam-ined in our myasthenia clinic between January1989 and April 1993 at least twice a year byone of us (NS or AM). In patients with alonger history previous hospital charts werereviewed. In a few patients additional followup data were collected from referring neurolo-gists.The diagnosis of ocular myasthenia gravis

was based on conventional clinical and labora-tory criteria. 1 2 7 IfAChR antibodies and repet-itive nerve stimulation were both negative,then clinical signs, edrophonium chloride(Tensilon) testing, and treatment response, aswell as other diagnostic measures (normal CTor MRI and CSF investigation) were re-evalu-ated carefully to exclude other causes of eyemuscle weakness (see also comment on differ-ential diagnosis below). Because of its lowspecificity, a positive edrophonium chloridetest was never used as a single diagnostic crite-rion.22-25

Patients were only included if symptomsand signs were restricted to the extraocularmuscles for at least three months after onset.This arbitrary limit seemed reasonable,because early disease stages (often beforereferral) were usually not well documented.Also, we thought that the development of gen-eralised myasthenia gravis after a few days orweeks of exclusive weakness of extraocularmuscles would not qualify for a diagnosis ofocular myasthenia gravis as a separate entity.In this respect we follow the criteria ofOosterhuis.'6 Also, subtle signs of generalisedweakness, particularly facial weakness, weresought for and, if present, those patients werenot included. Grading of severity of diseasewas semiquantive. "Mild", "moderate", or"severe" ocular myasthenia gravis stands formild, moderate, or severe disability in every-day life, with particular respect to impairmentat work and car driving. This seemed moreappropriate as the clinically more relevant sub-jective impairment does not necessarily corre-late with the actual degree of extraocularmuscle weakness. Improvement was docu-mented when ocular myasthenia gravis wasameliorated by at least one such grade.Remission was stated when there was noremaining disability with or without drugtreatment.

Repetitive nerve stimulation was performedaccording to Schumm and St6hr27 with minormodifications.The accessory nerve was stimu-

lated behind the sternocleidomastoid musclewith supramaximal intensity at 3 Hz with abipolar surface electrode and recorded fromthe trapezius muscle. A decrement of greaterthan 10% between the first and fifth potentialwas considered pathological.

Antibodies to AChRs were measured byradioimmunoassay according to Vincent andNewsom-Davis.'8 Human amputated leg mus-cle was used as a source of antigen, incubatedwith 12'I-a-bungarotoxin (Amersham-Buchler,Brunswick, Germany) and subsequently withdifferent dilutions of the patients' serum. Afterthorough washing the bound radioactivity wasmeasured on a 1counter. Known negative,positive, and equivocal serum samples wererun in each assay. Values greater than05 nmol a-bungarotoxin binding sites/l wereconsidered positive, between 0-2 and0 5 nmol/l equivocal, and less than 0-2 nmol/lnegative. To avoid variations between assays,follow up investigations always includedretesting of the previous one or two serumsamples of a patient together with the newserum.

ResultsPATIENTSWe had clinical data from 178 patients withmyasthenia gravis; 78 of them (44%; 40female, 38 male) were diagnosed to have ocularmyasthenia gravis according to the criteriaoutlined above. Mean duration of disease was 8years 4 months (SD 9 years, range 6 months-58years 2 months). Twenty four of the 78patients with ocular myasthenia gravis (31%,13 female, 11 male) developed symptoms ofgeneralised muscle weakness later during thecourse (see below). Age and sex distribution ofocular, primary, or secondary generalisedpatients were not significantly different (age atonset for all 78 patients with ocular myasthe-nia gravis: mean 50-6 (SD 19-5) years, range10-84 years; data not shown). Remarkably,the diagnosis of myasthenia gravis in the 78ocular patients was made only 39-8 (SD 93-5)months after the onset of symptoms.

Concomitant autoimmune diseases werediagnosed in 39 (22%) of all 178 patients withmyasthenia gravis reviewed in this study.Thyroid disease was by far the most commoncondition (33 patients, 19%; female:male ratio3-1:1). These proportions were not signifi-cantly different among the ocular patients (forexample, total incidence of thyroid disease 19of 78, 24%).

Family history disclosed that six of the 178patients (3%) had a first or second degree rela-tive with myasthenia gravis. Two pairs ofsiblings (including one pair of monozygotictwins) had generalised myasthenia gravis;another two in the ocular group had relativeswith generalised myasthenia gravis who werenot part of our sample.

DIFFERENTIAL DIAGNOSIS OF SERONEGATIVEOCULAR MYASTHENIA GRAVISIn three female patients (ages 45, 57, and 74years) referred to us with suspected seronega-

157

on Decem

ber 31, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.62.2.156 on 1 February 1997. D

ownloaded from

Sommer, Sigg, Melms, Weller, Schepelmann, Herzau, Dichgans

tive ocular myasthenia gravis, this diagnosishad to be revised during the course. All threehad double vision and unilateral ptosis withmild daily fluctuations and the diagnosis hadmainly been based on a positive edrophoniumchloride test. AChR-antibodies and electro-physiology were normal. One of the patientshad been thymectomised and thymic rem-nants without germinal centres were found.When re-evaluating the diagnosis, we foundunilateral contrast enhancing lesions on CTand MRI in the region of the cavernous sinus inall three. These patients are not included inthe present evaluation, but deserve mention,because their history reflects well the diagnosticproblems encountered in ocular myastheniagravis. The MRI of one of the patients hasbeen published previously.7

PROGNOSTIC FACTORSUsing secondary generalisation as an endpoint, no significant prognostic factor could bedetermined. Nevertheless, mild symptoms atonset, normal results at repetitive nerve stimu-lation, and negative serum AchR antibodytesting were found more often in patientswhose myasthenia remained limited toextraocular muscles (fig lA-C). Patients withmild ocular symptoms developed generalisedmyasthenia gravis in 14% (three of 22). Inmoderate and severe ocular myasthenia gravis

Figure 1 Evaluation ofprognostic factors in ocularmyasthenia gravis.Patients with mildsymptoms (A) or normalrepetitive accessory nervestimulation (B) have asomewhat lower risk ofdeveloping generalisedmyasthenia (NS).Detectable AchRantibodies weresignificantly more frequentin the group withsecondary generalisation.The numbers included inthe bars indicate theproportion ofpatients withimmunosuppressive drugtreatment (for example,15119: among 19 patientsin this group, 15 receivedcorticosteroids andlorazathioprine). Note thatin B and C the total n =66, because data were notavailable in 12 patientsbefore generalisation.

m Remaining ocularE Generalised

U)

a)

00z

40

30

20

10

0

50

X) 40

a)*; 300.'o 200

Z l

A: Clinical severity

these proportions were 36% (15 of 42) and43% (six of 14) respectively (X2 = 4.47, NS;fig 1A).

Patients with a pathological decrement inrepetitive accessory nerve stimulation con-verted to generalised myasthenia gravis in 32%(six of 19), but only 15% (seven of 47) did sowith an initially normal electrophysiologicalresult (x2 = 2-38, NS; fig iB).Among the seronegative patients one of 23

(4%) developed generalised myasthenia gravis.This proportion was significantly higher in theseropositive group (11 of 43, 26%; X2 = 4 54,P < 005; fig 1C). Also, in patients with sec-ondary generalisation the mean antibody titrewas significantly higher at the first diagnostictesting (mean 11 9 (SD 14-4), range 0-47-5 nmol/l) than in the permanently ocularpatients (mean 4 1 (SD 6-0), range 0-25 9 nmol/l; t = 3-02, P < 0-01; not shown).As expected, the maximum levels during thewhole observation period were tenfold higherin the patients with generalisation (61-6 (SD116-7) nmol/l) compared with those whoremained ocular (6-3 (SD 13-7) nmol/l).

Tensilon testing was positive in 97% of thepatients with ocular myasthenia gravis, butwas never used as a single diagnostic criterion(see above).

Data on HIA were available in 21 out ofthe 78 of the ocular patients and there wasdetectable trend towards a specific association;further investigation was not performed.

COURSE AND TREATMENT OF OCULARMYASTHENIA GRAVISOutcomeTwenty four of the 78 patients with ocularmyasthenia gravis (31%) developed secondarilygeneralised muscle weakness. The medianonset of generalisation was 24-5 months(range 3-132) after the first ocular symptoms(fig 2). Among the 54 permanently ocularpatients 29 (54%) had reached remission by

18j

15

12

U) 9

a) 60._4- 300z c

A

I1

6

42-

Seronegative Seropositive

B

D 12 24 36 48 60 72 84 96 108 120 132Time (months)

Figure 2 Cumulative incidence ofgeneralised symptomsin patients with initially purely ocular myasthenia gravis.A shows the patients without, B with immunosuppressivedrug treatment. The median time to generalisation was 16months in A and 33 months in B. The risk of converting togeneralised symptoms is greatest during the first years, but isstill present after many years.

U1)4)CL

0.,00z

158

on Decem

ber 31, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.62.2.156 on 1 February 1997. D

ownloaded from

Ocular myasthenia gravis: response to long term immunosuppressive treatment

the end of the observation period. All othersimproved (n = 18) or remained the same(n= 7); none of the permanently ocularpatients worsened. Although patients aftergeneralisation were not analysed in detail inthis study, it is worthwhile mentioning that 11of the 24 (46%) had also reached remission bythe end of the observation time.

Symptomatic treatmentSixty of the 78 patients received pyridostig-mine at some time during their follow up. Themaximum individual daily dose was 149-8(SD 74 2) mg. The duration of pyridostigminetreatment was on average 45-6 (SD 39 9)months. Fourteen patients were treated withpyridostigmine alone for some time. Of those,seven (50%) reported considerable improve-ment, six (43%) mild improvement, and inone patient no effect was reported.

Immunosuppressive drug treatmentForty five patients received prednisolone and27 azathioprine. Figure 3 gives an overview ofthe various combinations of immunosuppres-sion and thymectomy. Prednisolone was givenfor an average of 32-3 (SD 32.3) months, witha maximum individual daily dose of 51-9 (SD26-1) mg. (In those patients who alwaysreceived alternate day regimens 50% of theirmaximum dose was used for the calculation.)The mean duration of treatment with azathio-prine was 43-7 (SD 35.2) months with a meanmaximum dose of 145-3 (SD 25-0) mg.Among the patients who were treated withprednisolone but without azathioprine, a clearpositive effect was found in 19 of 22 (86%).Among the four patients treated with azathio-prine alone, we found an unequivocally posi-tive effect in three of them. A combination ofboth drugs was used in 23 and was beneficialin 21 (91%).

Figure 3 Frequency ofthe differentimmunosuppressivetherapies. Total number ofpatients = 78. Thenumbers in parentheses(total = 24) give thenumbers of those whodeveloped generalisedmyasthenia gravis. Notethe low proportion ofpatients with secondarygeneralisation with anyimmunosuppressivetreatment.

ThymectomyThymectomy was performed in 12 patientswith purely ocular symptoms, because of anabnormal chest CT (fig 3). Probably due tothe late diagnosis (see above) the procedurewas performed on average 50 (range 2-253)months after onset of symptoms. Histologywas available in 11 and disclosed thymoma infour (three non-invasive, one invasive), lym-phofollicular hyperplasia in two, thymic rem-nants in two, and thymic involution in threepatients. Thymectomised patients were fol-lowed up for an average of 81 (range 18-190)months after the operation, and all but onewere treated with additional immunosuppres-sants (fig 3). Six patients reached remissionduring follow up; in another four ocular weak-ness improved. The remaining two patients(both with thymoma, one non-invasive, theother invasive) developed generalised myas-thenia after six or 125 months respectively.

SIDE EFFECTSNo lasting sequelae of thymectomy werereported. Pyridostigmine was associated withside effects in two patients. In one a systemicallergic reaction (generalised exanthema twodays after starting the drug) was reported, inanother patient bradycardia was ascribed todrug overdose. Corticosteroids led to moder-ate side effects in 12 patients (cushingoid infive, restlessness and sleeplessness in five,severe acne vulgaris in one, and gastric upsetin one), and severe side effects in two patients(one each with decompensation of diabetesmellitus or psychosis). Moderate side effects ofazathioprine occurred in six patients (leucope-nia between 2000 and 4000/Ml in two patients,hair loss in two, nausea in one, and skin andnail changes in one). Severe leucopenia(< 2000/,l) occurred in one patient. Inanother two (not considered in fig 3) azathio-prine had to be stopped within the firstmonths due to gastrointestinal upset. Nosevere infections were found under immuno-suppressive treatment in this patient group.

DiscussionOcular myasthenia gravis had a good progno-sis in most patients in this study. In 69% thedisease remained confined to the extraocularmuscles. Of these, 54% were in remission,33% improved, and 13% patients wereunchanged at the end of the study. The mostremarkable difference from previous studies isthe relatively low fraction of patients convert-ing to generalised myasthenia gravis, whichwas 31% in our population, but 49-69% inother previous reports. The development ofgeneralised symptoms was used as a major endpoint by us and others, because it indicatesmajor progression of disease and managementhas to be re-evaluated.

In the study published in 1983 by Bever etal29 53 of 108 patients (diagnosed 1957-69 inNew York City)-that is, 49%-later gener-alised. Grob reported in his large collective30that about 40%, and in a later survey3l 34% of202, of his patients with initial ocular myasthe-

159

on Decem

ber 31, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.62.2.156 on 1 February 1997. D

ownloaded from

Sommer, Sigg, Melms, Weller, Schepelmnann, Herzau, Dichgans

nia gravis would remain ocular. The authors ofthose studies do not report details on possibleimmunosuppressive treatments applied topatients with ocular myasthenia gravis. Asboth series started decades ago it might safelybe assumed that only a small proportion ofpatients had corticosteroids and even fewerhad azathioprine.

Oosterhuis reviewed an even earlier series ofpatients who were first documented between1926 and 1965 in Amsterdam and had to bemanaged without steroid or other immuno-suppressive treatment.26 In this group 24 of 35(69%) developed generalised myastheniagravis, probably most realistically reflectingthe natural course of the disease. Althoughpatient groups in the various studies may notbe comparable with each other, it is evidentthat there is a gradual evolution towards alower proportion with secondary generalisa-tion. Common to all studies, including ours, isthe finding that the risk of generalisation isgreatest soon after onset, gradually decliningwith time. Of the patients developing gener-alised myasthenia gravis 50% did so by twoyears after onset in our study, 88% inOosterhuis' group,26 83% in the study of Beveret al.29The proportion of 64% seropositive patients

in our group was comparable with most of thepreviously published results (45%-71%).28 32 30

Whereas diagnosis of ocular myasthenia gravisis straightforward in a patient with a compatiblehistory and detectable serum AchR antibodies,diagnostic problems arise in the seronegativepatients. Suspicion should be aroused espe-cially if extraocular muscle weakness is unilat-eral and follows a neurogenic pattern. Thebriefly summarised history of three patientswith fluctuating extraocular muscle weaknessand a positive edrophonium chloride test (seeabove) adds to other reports of patients with amissed tumour of the skull base. The largestsuch series, consisting of eight patients withextraocular muscle weakness, positiveresponse to anticholinesterase treatment, andintracranial mass lesions was reported byMoorthy et al.2 '3 Thus seronegative ocularmyasthenia gravis must, in our opinion,remain a mainly clinical diagnosis after acareful diagnostic work up including skullimaging. The lack of specificity of the edro-phonium chloride test has been discussedrepeatedly,7 222425 and we cannot but repeat thewarning not to interpret response to anti-cholinesterase drugs as a certain diagnosticsign for ocular myasthenia gravis.The good outcome in patients with ocular

myasthenia gravis in this study undoubtedlyreflects the overall improvements in the treat-ment of autoimmune diseases. However, thebeneficial effects of the different treatmentsare difficult to distinguish. Steroids and aza-thioprine clearly improve symptoms and signsin all forms of autoimmune myastheniagravis12 and it would therefore seem natural togive both drugs, steroids for short or mediumterm treatment and azathioprine for long termimmunosuppression. Nevertheless, detailedreports on therapeutic experiences in ocular

myasthenia gravis as a separate entity are rare.Whereas steroids are probably widely appliedin ocular myasthenia gravis, the use of otherimmunosuppressants has hardly been dis-cussed. Evoli et al report a series of 48 patientswith ocular myasthenia gravis.37 They statethat corticosteroids are effective in most cases,but may cause problems when being taperedoff. Azathioprine was not considered in thosepatients followed up between 1968 and1986.17 Considering the beneficial effect ofazathioprine alone or in combination withprednisolone in our patients, we would pro-pose that azathioprine is a valuable immuno-suppressant in ocular myasthenia gravis. Thesteroid sparing effect of the drug is generallyaccepted,2 and side effects are usually tolera-ble. Haematological (in 11%) and other sideeffects were rare compared with other studies.Hohlfeld et al38 reported haematological sideeffects in 18%, Kissel et al39 in 22%, and Witteet al 10 in 17% of their patients. However, allthese studies were performed in generalisedmyasthenia gravis, and often side effects wereconsidered to be dose related in patientsrequiring relatively high dosage for stabilisa-tion of symptoms (up to 6 6 mg/kg in thestudy of Hohlfeld et al,38 and up to 3 4 mg/kgin that of Witte et al -10). In our clinic patientswith generalised myasthenia gravis are treatedwith 2-2-5 mg/kg azathioprine with anincrease in dose (usually not above 3 0 mg/kg),if the treatment result is not satisfactory. Inpatients with ocular myasthenia gravis, how-ever, we find that in most patients 2 0 mg/kg issufficient to lead to improvement (results notshown in detail).Thymectomy is now a standard treatment

for patients with myasthenia gravis with sus-pected thymoma and in young onset patientsin whom thymic hyperplasia with germinalcentres can be expected.1 2 However, in mostneurologists' opinions thymectomy is not astandard treatment for ocular myasthenia."l Inour series thymectomy was only performedwhen chest CT indicated thymoma.Thymoma was found in four of 12 patients,and two of those developed generalised myas-thenia gravis, whereas all patients without thy-moma remained ocular. Remarkably, in twoyoung onset patients (onset < 45 years) withshort duration of disease there was histologicalevidence for lymphofollicular hyperplasia,whereas in two other young onset patientswith much longer duration of disease, similarto those with onset over 45 no hyperplasia wasfound. This supports the hypothesis that atleast a proportion of ocular patients withmyasthenia gravis have thymic changes similarto generalised young onset patients, and thus asimilar pathogenesis may be assumed. How-ever, evaluation of the retrospective data is dif-ficult, because all except one patient requiredimmunosuppressive drug treatment. In 1985Schumm et al reported a series of 18 mainlyyoung patients with ocular myasthenia graviswho seemed to benefit from thymectomy and13 of whom showed thymic hyperplasia.2'Although immunosuppressants were used inparallel and clear criteria for an expected ther-

160

on Decem

ber 31, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.62.2.156 on 1 February 1997. D

ownloaded from

Ocular myasthenia gravis: response to long term immunosuppressive treatment

apeutic effect could not be identified, theauthors concluded that thymectomy should beconsidered in ocular myasthenia gravis.What general treatment plan should be fol-

lowed in ocular myasthenia gravis? Ocularmyasthenia gravis is treated as an entity forpractical reasons in this study. However, thereis ample evidence that in most patients it isprobably simply a mild form of myastheniagravis, without fundamental pathogenetic dif-ference to the three forms of generalised myas-thenia gravis outlined above. All clinicalfindings, such as thymic changes, associatedautoimmune disease, and occasional familialoccurrence (see results) can also be made inocular myasthenia gravis with roughly similarfrequency as in generalised myastheniagravis.4243 Therefore, treatment of ocularmyasthenia gravis should follow the generalprinciples of myasthenia gravis therapy.The use of pyridostigmine and corticos-

teroids is generally accepted in ocular myas-thenia gravis. Practically, we start with 20-30 mg pyridostigmine three to four timesdaily, gradually increasing the dose untilsymptoms disappear. Normally, 240 mg perday are not exceeded. Most patients willrequire additional steroids. For mild symp-toms we start with 20-30 mg prednisolone onalternate days. This will often lead to consider-able improvement after four to six weeks. Formore severe ocular symptoms up to 50-60 mgprednisolone daily (or an equivalent steroid)are given for two to three months untilimprovement is detectable. Dose reductionshould not be faster than 5 mg in six to eightweeks. Too fast a reduction of steroids is themost common mistake leading to a relapse. 1 44From our experience in ocular myasthenia

gravis, we suggest that azathioprine should beused as the immunosuppressant of choice, ifocular symptoms do not remit completelyunder steroid treatment or cannot be con-trolled on a very low dose of steroids. Longterm azathioprine treatment seems safe, pro-vided blood count and liver enzymes arechecked regularly. Thymectomy might be anadditional treatment option, but cannot clearlybe evaluated on current evidence. A prospec-tive trial could settle this issue and seems mostrewarding in patients with onset before 45years of age. If thymectomy is considered inocular myasthenia gravis, it should in any casebe limited to patients with detectable AChR-antibodies, because seronegative patients vir-tually never have thymoma in our experienceand that of others4546 and do not normallyhave lymphofollicular hyperplasia like theseropositive young onset patients.20

1 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.

2 Drachman DB. Myasthenia gravis. New Engl JfMed 1994;330:1797-810.

3 Oosterhuis HJGH. Myasthenia gravis. Edinburgh:Churchill Livingstone, 1984.

4 Somnier FE, Keiding N, Paulson OB. Epidemiology ofmyasthenia gravis in Denmark: a longitudinal and com-prehensive population survey. Arch Neurol 1991;48:733-9.

5 Tola MR, Granieri E, Paolino E, et al. Epidemiologicalstudy of myasthenia gravis in the province of Ferrara,Italy. J Neurol 1989;236:388-90.

6 Mertens HG, Balzereit F, Leipert M. The treatment ofsevere myasthenia gravis with immunosuppressiveagents. Eur Neurol 1969;2:321-39.

7 Sommer N, Melms A, Weller M, Dichgans J. Ocular myas-thenia gravis. A critical review of clinical and pathophysi-ological concepts. Doc Ophthalmol 1993;84:309-33.

8 Pinching AJ, Peters DK, Newsom-Davis J. Remission ofmyasthenia gravis following plasma-exchange. Lancet1976;ii: 1373-6.

9 Compston DAS, Vincent A, Newsom-Davis J, BatchelorJR. Clinical, pathological, HLA antigen, and immunolog-ical evidence for disease heterogeneity in myastheniagravis. Brain 1980;103:579-601.

10 Rowland LP. Controversies about the treatment of myas-thenia gravis. J Neurol Neurosurg Psychiatry 1980;43:644-59.

11 Johns TR. Long-term corticosteroid treatment of myastheniagravis. Ann NYAcad Sci 1987;505:568-83.

12 Cornelio F, Peluchetti D, Mantegazza R, Sghirlanzoni A,Collarile C. The course of myasthenia gravis in patientstreated with corticosteroids, azathioprine, and plasma-pheresis. Ann NYAcad Sci 1987;505:517-25.

13 Matell G. Immunosuppressive drugs: azathioprine in thetreatment of myasthenia gravis. Ann NYAcad Sci 1987;505:588-94.

14 Schalke B, Kappos L, Dommasch D, Rohrbach E, MertensHG. Cyclosporin-A treatment of myasthenia gravis: ini-tial results of a double-blind trial of cyclosporin-A versusazathioprine. Ann NYAcad Sci 1987;505:872-5.

15 Tindall RSA, Phillips JT, Rollins JA, Wells L, Hall K. Aclinical therapeutic trial of cyclosporine in myastheniagravis. Ann NYAcad Sci 1993;681:539-51.

16 Genkins G, Sivak M, Tartter PI. Treatment strategies inmyasthenia gravis. Ann NYAcad Sci 1993;681:603-8.

17 Sommer N, Willcox N, Harcourt GC, Newsom-Davis J.Myasthenic thymus and thymoma are selectivelyenriched in acetylcholine receptor-reactive T cells. AnnNeurol 1990;28:312-9.

18 Melms A, Schalke BCG, Kirchner T, Muller-HermelinkHK, Albert E, Wekerle H. Thymus in myasthenia gravis:isolation of T-lymphocyte lines specific for the nicotinicacetylcholine-receptor from thymuses of myasthenicpatients. J Clin Invest 1988;81:902-8.

19 Willcox N, Schluep M, Ritter MA, Schuurman HJ,Newsom-Davis J, Christensson B. Myasthenic and non-myasthenic thymoma. An expansion of a minor corticalepithelial subset? Am J Pathol 1987;127:447-600.

20 Willcox N, Schluep M, Ritter MA, Newsom-Davis J. Thethymus in seronegative myasthenia gravis patients. JNeurol 1991;238:256-61.

21 Schumm F, Wietholter H, Fateh-Moghadam A, DichgansJ. Thymectomy in myasthenia with pure ocular symp-toms. J Neurol Neurosurg Psychiatry 1985;48:332-7.

22 Newsom-Davis J. Edrophonium responsiveness not neces-sarily diagnostic of myasthenia gravis. Muscle Nerve 1990;13:1186.

23 Moorthy G, Behrens MM, Drachman DB, et al. Ocularpseudomyasthenia or ocular myasthenia "plus": a warningto clinicians. Neurology 1989;39: 1150-4.

24 Seybold ME. The office Tensilon test for ocular myasthe-ma gravis. Arch Neurol 1986;43:842-3.

25 Daroff RB. The office Tensilon test for ocular myastheniagravis. Arch Neurol 1986;43:843-4.

26 Oosterhuis HJGH. The natural course of myastheniagravis: a long term follow up study. Y Neurol NeurosurgPsychiatry 1989;52:1121-7.

27 Schumnm F, Stohr M. Accessory nerve stimulation in theassessment of myasthenia gravis. Muscle Nerve 1984;7:147-51.

28 Vincent A, Newsom-Davis J. Acetylcholine receptor anti-body as a diagnostic test for myasthenia gravis: results in153 validated cases and 2967 diagnostic assays. Y Neurol1985;48:1246-52.

29 Bever CT, Aquino AV, Penn AS, Lovelace RE, RowlandLP. Prognosis of ocular myasthenia gravis. Ann Neurol1983;14:516-9.

30 Grob D. Clinical manifestation of myasthenia gravis. In:Albuquerque EX, Eldefravi AT, eds. Myasthenia gravis.London: Chapman and Hall, 1983:319-45.

31 Grob D, Arsura EL, Brunner NG, Namba T. The courseof myasthenia gravis and therapies affecting outcome.Ann NYAcad Sci 1987;505:472-99.

32 Lefvert AK, Bergstrom K, Matell G, Osterman PO,Pirskanen R. Determination of acetylcholine receptorantibody in myasthenia gravis: clinical usefuilness andpathogenic implications. Y Neurol Neurosurg Psychiatry1978;41:394-403.

33 Oda K, Goto I, Kuriowa Y, Onoue K, Ito Y. Myastheniagravis: antibodies to acetylcholine receptor with humanand rat antigen. Neurology 1980;30:543-6.

34 Tindall RSA. Humoral immunity in myasthenia gravis: bio-chemical characterization of acquired antireceptor anti-bodies and clinical correlations. Ann Neurol 1981 ;10:437-47.

35 Limburg PC, The TH, Hummel-Tappel E, OosterhuisHJGJ. Anti-acetylcholine receptor antibodies in myasthe-nia gravis. Part 1: relation to clinical parameters in 250patients. J Neurol Sci 1983;58:357-70.

36 Toyka KV, Heininger K. Acetylcholine-receptor antibodiesin the diagnosis of myasthenia gravis. Dtsch MedWochenschr 1986;111:1435-9.

37 Evoli A, Tonali P, Bartoccioni E, LoMonaco M. Ocularmyasthenia: diagnostic and therapeutic problems. ActaNeurolScand 1988;77:31- 5.

161

on Decem

ber 31, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.62.2.156 on 1 February 1997. D

ownloaded from

Sommer, Sigg, Melms, Weller, Schepelmann, Herzau, Dichgans

38 Hohlfeld R, Michels M, Heininger K, Besinger U, ToykaKV. Azathioprine toxicity during long-term immunosup-pression of generalized myasthenia gravis. Neurology1988;38:258-61.

39 Kissel JT, Levy RJ, Mendell JR, Griggs RC. Azathioprinetoxicity in neuromuscular disease. Neurology 1986;36:35-9.

40 Witte AS, Cornblath DR, Parry GJ, Lisak RP, Schatz NJ.Azathioprine in the treatment of myasthenia gravis. AnnNeurol 1984;15:602-5.

41 Lanska DJ. Indications for thymectomy in myastheniagravis. Neurology 1990;40:1828-9.

42 Bergoffen J, Zmijewski CM, Fischbeck KH. Familialautoimmune myasthenia gravis. Neurology 1994;44:

551-4.43 Christensen PB, Jensen TS, Tsiropoulos I, Sorensen T,

Kjaer M. Associated autoimmune diseases in myastheniagravis. A population-based study. Acta Neurol Scand1995;91: 192-5.

44 Miano MA, Bosley TM, Heiman-Patterson TD, et al.Factor influencing outcome of prednisone dose reductionin myasthenia gravis. Neurology 1991 ;41 :919-21.

45 Soliven BC, Lange DJ, Penn AS, et al. Seronegative myas-

thenia gravis. Neurology 1988;38: 514-7.46 Sanders DB, Howard JF. Disorders of the neuromuscular

transmission. In: Bradley WG, Daroff RB, Fenichel GM,Marsden CD, eds. Neurology in clinical practice. Boston:Butterworth-Heinemann, 1991:1819-42.

162

on Decem

ber 31, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.62.2.156 on 1 February 1997. D

ownloaded from