Clinical characteristics, diagnostic criteria and

Downloaded from bjo.bmj.com on April 18, 2012 - Published by group.bmj.com

Clinical science

Clinical characteristics, diagnostic criteria and therapeutic outcomes in autoimmune optic neuropathy L Frohman,1,2 K DellaTorre,1 R Turbin,1 L Bielory1,3,4 1

Department of Ophthalmology and Visual Sciences, UMDNJ— New Jersey Medical School, Newark, New Jersey, USA; 2 Department of Neurology and Neurosciences, UMDNJ—New Jersey Medical School, Newark, New Jersey, USA; 3 Department of Medicine, UMDNJ—New Jersey Medical School, Newark, New Jersey, USA; 4 Department of Pediatrics, UMDNJ—New Jersey Medical School, Newark, New Jersey, USA Correspondence to: Dr L Frohman, 90 Bergen Street, Newark, NJ 07103, USA; [email protected] Accepted 28 May 2009 Published Online First 18 August 2009

ABSTRACT Background/aims: Autoimmune optic neuropathy (AON) is characterised by chronically progressive or recurrent visual loss associated with serological or cutaneous evidence of autoimmunity, without a defined systemic autoimmune illness. It may improve with large doses of corticosteroids alone, or in combination with immunosuppressive agents. The aim was to determine the relative effectiveness of various therapeutic regimens in AON. Methods: All patients in this study fulfilled these criteria: (1) multiple attacks in one eye or attacks in both eyes (at least three total attacks); (2) a minimum of 12 months of neuro-ophthalmic follow-up; (3) serological abnormalities or skin biopsy changes consistent with AON; (4) no diagnosis of a defined collagen vascular disease or neurological autoimmune disease throughout follow-up, with the exception of one patient, later shown to be shown to be neuromyelitis optic antibody positive. Results: Nine cases were found (female = 7, male = 2, ages 8–74). One case received corticosteroids alone, the others received corticosteroids in combination with methotrexate/gammaglobulin (n = 1), methotrexate (n = 1), gammaglobulin (n = 1), chlorambucil (n = 2), cyclophosphamide (n = 1) and azathioprine/chlorambucil (n = 1), and one received multiple combinations of agents. Criteria for diagnosis are proposed. Conclusion: As AON is quite rare, no formal recommendation can be made regarding its best therapy, although there is a suggestion that chlorambucil, although potentially toxic, may yield long-term remission.

restrict the term to cases having multiple episodes of optic neuritis (at least three with a bilateral simultaneous attack counting as two attacks) with serological tests or skin biopsy indicating an autoimmune process but in the absence of a definable collagen vascular disease. In most cases seen by institutions with experience with this disorder, initial therapy of an acute attack has been with solumedrol in large doses (ie, 1–2 g/day). If patients do not respond to steroids, or relapse on steroid taper, or cannot tolerate the dose required to suppress disease activity, immunosuppressive agents are usually employed, although the best agent for AON remains unclear due to the paucity of cases. We present a retrospective analysis of eight new cases of autoimmune optic neuropathy, and one previously reported4 (see below) where additional laboratory information was obtained, meeting the criteria for diagnosis and where records were recoverable to assess immunosuppressive treatment outcomes. One patient required treatment with only corticosteroids; the remaining eight were treated with one or a combination of the following: intravenous gammaglobulin (IVIG), methotrexate, cyclophosphamide, chlorambucil, and azathioprine. Because AON can cause progressive and irreversible visual loss, it would be advantageous to identify what measures are most effective in halting the disease process.

MATERIALS AND METHODS Autoimmune optic neuropathy (AON) was described by Dutton et al in 1982.1 It is characterised by recurrent bouts of optic neuropathy, frequently requiring high doses of corticosteroids to control and/recurrent on steroid taper. Patients are predominantly female and often have serological abnormalities such as elevated antinuclear antibody (ANA) or anticardiolipin antibodies (ACA), yet with no evidence of a defined systemic disease. Kupersmith et al2 and Bielory et al3 demonstrated that evidence of vasculitis on haematoxylin and eosin stain, or of immunoreactant deposition on immunofluorescent staining of a non-lesional, non-sun-exposed skin biopsy, was highly correlated with a diagnosis of AON and perhaps the most reliable clinical assay. In subsequent years, the term AON has been applied to several conditions besides what Dutton originally described. This includes optic neuropathy in the setting of a known collagen vascular disease—for example, lupus optic neuropathy. We 1660

After obtaining approval from the New Jersey Medical School Institutional Review Board, a database of all patients seen on the NeuroOphthalmology Service of New Jersey Medical School from 1989 to 2005 was reviewed to find those who potentially had AON. Twenty-eight patients were clinically diagnosed as having AON. To be included, patients had to fulfil all of the following clinical criteria: 1. multiple attacks in one eye or attacks in both eyes with at least three total attacks; 2. either serological evidence of autoimmune diathesis (ANA, rheumatoid factor (RF), lupus anticoagulant (LA), or ACA) or skin biopsy of non-lesional/non-sun exposed skin consistent with vasculitis/collagen vascular disease (CVD), or with immunoreactant deposition on immunofluorescence (IF); 3. absence of defined CVD or other autoimmune systemic illness or other defined neurological autoimmune disease (one case (patient 3), initially diagnosed and treated for several years as AON, Br J Ophthalmol 2009;93:1660–1666. doi:10.1136/bjo.2009.159350


Clinical science Table 1 Patient characteristics on presentation

Patient no

Age at presentation

Sex

Other neurological symptoms

1 2 3 4 5 6 7 8 9

43 62 30 8 45 47 34 32 50

Female Male Male Female Female Female Female Female Female

No Yes No Yes No Yes Yes No No

with a typical skin biopsy, had a positive neuromyelitis optica antibody test after years iof treatment; because of the vasculitis on skin biopsy, and as the delineation between AON and neuromyelitis optica (NMO) is not clear, we have included this case in the series); 4. records available with at least 12 months of neuroophthalmic follow-up.

RESULTS A total of nine cases meeting all inclusion criteria were identified (including the one case who later tested positive for neuromyelitis optica), seven (78%) were female, and two were male, with an age range of 8–74 years on presentation (table 1). On initial assessment, 44% had a history of other neurological symptoms besides visual loss (ie, numbness, tingling, weakness), and 11% had other symptoms consistent with collagen vascular disease (ie, joint pain, dry mouth), yet none met criteria for any known neurological/systemic illness. Only one patient presented with optic disc swelling, while 44% reported a history of eye pain. All patients ultimately had bilateral AON. As this was a retrospective study, the same serological workup was not done in all cases. We reviewed results for abnormalities in erythrocyte sedimentation rate (ESR), syphilitic serologies (Venereal Disease Research Laboratory (VDRL), fluorescent treponemal antibody absorption), complement

Other collagen vascular disease symptoms

Disc oedema right eye

Disc oedema left eye

Ocular pain

Yes No No No No No No No No

No No Yes No No No No No No

No No No No No No No No No

No No Yes No Yes No No Yes Yes

component 3 (C3), complement component four (C4), (ANA), RF, ACA and LA. All nine cases in this series had suggestive laboratory abnormalities. No patients had abnormal VDRL, fluorescent treponemal antibody absorption or ACA IgG. The laboratory abnormalities were as follows (table 2): abnormal C3 in one of three tested (33%), abnormal C4 in one of four tested (25%), abnormal ANA in four of nine tested (44%), abnormal RF in two of eight tested (25%), abnormal ACA IgM in three of nine tested (33%) and abnormal LA in two of five tested (40%). Abnormal anti-DNA antibodies were noted in one patient and an abnormal latex fixation in another patient. Years later, after the neuromyelitis antibody test was available, one case (case 4) was tested and was negative, and one case (case 3) was positive. The false-positive rates for these tests are shown in table 3. All nine patients underwent lumbar puncture, which was normal except for one patient (no 8) who had 16 white cells with a normal protein level. All nine patients underwent MRI of the brain (table 4), with four of nine (44%) showing abnormalities. Eight also underwent MRI of the optic nerves, with five of eight (63%) demonstrating changes. All nine patients underwent skin biopsy and had abnormal results (table 4). Drawing conclusions regarding therapy is difficult, as our series is limited in that it is a retrospective non-controlled study, and due to the rarity of the disorder, the number of patients is small. The choice of initial therapy was largely based upon the prior literature and experience of the authors of the initial papers on

Table 2 Laboratory assessment Erythrocyte Patient sedimentation rate no

VDRL

FTAabs

C3

C4

Antinuclear antibody

Rheumatoid factor

Anticardiolipin antibodies IgG

Anticardiolipin antibodies IgM

Lupus anticoagulant

1 2

14 5

Neg TND

Neg TND

Nl TND

Nl TND

Nl Nl

1:2 TND

Nl Nl

Nl Nl

3 4 5

100 33 78

Neg Neg TND

Neg Neg Neg

TND Nl Nl

TND Nl Nl

Nl Nl 1:80 (speckled)

Nl Nl Nl

Nl Nl Nl

Nl 23 (Nl to 6) 1:200, Nl,1:100

Nl PTT_LA 53.7 (Nl to 52), dRVVT 33.0 (Nl) Nl Neg TND

6

35

Neg

Neg

Nl

Nl

1:20 (,1:20) Nl

Nl

TND

7

16

Neg

TND

TND

TND

Nl

Nl

Nl

ABNL*

8

53

Neg

Neg

TND

TND

Nl

Nl

Nl

TND

9

45

TND

TND

TND

TND

1:40 (speckled) 1:40 (speckled) 1:2560 (centromere) Nl

Nl

Nl

13.0 (Nl to 11)

Nl

Other abnormalities Latex fixation Thrombin time 19.7 (Nl to 18)

Elevated anti-DNA, IgG3 deficiency (21, Nl 41–129) cerebrospinal fluid protein, 59 (Nl to 45)

*Level/dilution not reported by lab. ABNL, abnormal result, all abnormals are in bold; dRVVT, Russell Viper Venom time; FTA-abs, fluorescent treponemal antibody absorption; Neg, negative result (normal); Nl, normal result; TND, test not done. Venereal Disease Research Laboratory (VDRL).

Br J Ophthalmol 2009;93:1660–1666. doi:10.1136/bjo.2009.159350

1661


Clinical science Table 3 False-positive values Test

False-positive rate

Antinuclear antibody Rheumatoid factor Anticardiolipin antibodies Lupus anticoagulant Anti-double-stranded DNA Latex fixation

5–10%5 5%5 2–4%6 ,1%6 0%7 5–8%8

AON. These cases have been treated for their acute phase with corticosteroids. This is because, aside from intravenous cyclophosphamide, the onset of immunosuppressive effect is delayed several weeks in the other agents. Immunosuppressives (or IVIG) are used as steroid sparing and/or maintenance drugs once the acute event has been treated with corticosteroids. Early in the series, our strategy was to use the drug that had been the most successful in prior series, chlorambucil. However, as this agent’s long-term toxicity became more apparent, this was relegated to an agent held in reserve for when agents with better safety profiles failed. Thus, later in the collection of these cases, methotrexate and azathioprine were used as initial agents. If a patient needed rapid induction of immunosuppression because of the rapidity or severity of the visual decline, intravenous cyclophosphamide was used. Mycophenolate mofetil was not used in the patients in this series. It may turn out to be a viable agent for controlling visual loss, but it too has recently been associated with serious toxicities, such as progressive multifocal leukencephalopathy.9 Because of the capability of the antitumour necrosis factor monoclonal antibodies to cause an autoimmune illness directed against the central nervous system, including optic neuritis and a multiple sclerosis-like syndrome, which would impact our ability to separate disease from iatrogenic illness, we, in the latter part of the series when they were available, considered these agents but held them in reserve and did not need to use them.10 One case did not require any agent aside from corticosteroids to control her disease process. She was weaned off steroids after 50 months. The remaining eight cases (89%) required either immunosuppressive therapy or IVIG. Two cases were initially treated with

methotrexate (case nos 2 and 3). One of these patients (no 2) was weaned off the drug after eight uneventful months, relapsed 2 months later and was switched to IVIG. The second (no 3) was successfully treated for a year, weaned and relapsed after 2 months. In this case, methotrexate was restarted and failed to control visual loss. Cyclophosphamide was tried and also failed. IVIG was given as a third agent and led to some recovery of vision (table 5) The patient with most resistance to therapy and with the worst outcome (patient 3) is the patient who subsequently tested positive for NMO. Two cases received IVIG as their initial therapy (nos 4 and 5). One of these (no 5) relapsed on IVIG and was switched to chlorambucil. This patient had multiple recurrences and required multiple regimens, sometimes with three drugs, to control her optic neuropathy over her long course. The other patient, a child (no 4) was successfully treated with IVIG, which was chosen to avoid immunosuppression in childhood. Two patients received chlorambucil as initial therapy (nos 6 and 7). Both cases responded to this therapy. One patient was successfully treated with cyclophosphamide (no 8). Another patient (no 9) was initially treated with azathioprine and, after it was tapered off, relapsed, and was successfully treated with chlorambucil.

DISCUSSION The differential diagnosis of AON is as follows: 1. demyelinating optic neuritis; 2. optic neuropathy in CVD/autoimmune/vasculitic disease; 3. recurrent optic neuropathy associated with autoimmune neurologic/non-systemic disorders (ie, myasthenia gravis); 4. sarcoidosis; 5. idiopathic pachymeningitis with optic nerve involvement; 6. lymphocytic adenohypophysitis; 7. autoimmune-related retinopathy and optic neuropathy syndrome (ARRONS) (tends not to have relapsing/remitting episodes); 8. paraneoplastic optic neuropathy (tends not to have relapsing/remitting episodes); 9. neuromyelitis optica;

Table 4 MRI and skin biopsy results Patient no

Initial MRI brain

Initial MRI optic nerve

Skin biopsy result

1

Normal

Lymphocytic chronic vasculitis

2

Age-appropriate atrophy

3

Non-specific foci in subcortical R frontal white matter

Enhancement of R ON and faint enhancement of L ON Mild enhancement LON in orbital apex, bilateral intracranial enhancement Prominent enhancement of R ON

4

Marked white-matter changes in the cerebrum and cerebellum

Not done, optic nerves/chiasm normal on MRI brain

5

Normal

6 7

Enhancing white-matter hyperintensities proximal to the frontal horn, lesion in right brachium pontis Normal Normal

8

Normal

Enhancement L ON, chiasm, and tracts L.R

9

Multiple scattered increased signal foci in bilateral cerebral hemispherical Enhancement of R ON white matter

Normal Normal

2+ granular C3D and 1+ patchy granular IgM at dermal epidermal zone low-grade lymphocytic vasculitis with basal vacuolar degeneration, +/2 perivascular IgM deposition, 1+ perivascular C1q deposition Thrombin deposition in vessels consistent with anticardiolipin antibody syndrome, subtle IgM granular deposits and moderate C3D deposits at dermal–epidermal junction Vasculitis, immune complexes present Consistent with scleroderma 1+ IgG and C3D at dermal– epidermal junction Mild chronic perivascular inflammatory infiltrate and mild deposition of mucinous material in the dermis 1+ IgM deposition along basement membrane and borderline IgG deposition

L, left; ON, optic nerve; R, right.

1662



Clinical science Table 5 Visual outcome Patient no

Therapy

Initial VA*

Best VA{

Final VA{

Best VA % of final VA1

1

Corticosteroids only

2

Methotrexate, IVIG

3

Methotrexate (two courses), cyclophosphamide, IVIG

3/400 CF 60 20/800 20/40 20/25

5/100 CF 39 20/25 20/20 20/25

5/100 HM 20/25 20/30 NLP

100% OD 10% OS 100% OD 67% OS 0% OD

4

IVIG

5

IVIG, chlorambucil, multiple other agents"

20/20 20/800 20/400 20/20

20/20 20/25 20/25 20/20

NLP 20/25 20/25 20/20

0% OS 100% OD 100% OS 100% OD

6

Chlorambucil

7

Chlorambucil

8

Cyclophosphamide

9

Azathioprine, chlorambucil

20/40 20/70 20/400 20/50 20/80 20/25 LP 20/30

20/20 20/30 20/300 20/20 20/20 20/20 20/200 20/20

HM CF 19 CF 39 20/20 20/20 20/25 20/400 20/20

0.1% OS 1.5% OD 15% OS 100% OD 100% OS 80% OD 50% OS 100% OD

20/20

20/15

20/40

37.5% OS

Mean follow-up

Length of follow-up (months) 50 33 65

65 190

175 102 30 96

90 months (7.5 years)

*Initial visual acuity (VA) defined as VA on presentation to the neuro-ophthalmology service. {Best VA defined as best VA in each eye after therapy (not necessarily concurrent). {Final VA defined as VA at termination of neuro-ophthalmic follow-up. 1Calculations using Halladay’s method with CF = 0.01 and HM 0.001 (http://www.iovs.org/cgi/content/full/47/3/1236). LP and NLP assigned values of 0 for statistical purposes. "Cyclophosphamide, plaquenil, and two additional courses of chlorambucil. CF, count fingers; HM, hand motion; IVIG, intravenous gammaglobulin; LP, light perception; NLP, no light perception; OD, right eye; OS, left eye.

10. chronic relapsing inflammatory optic neuropathy (CRION). Demyelinating optic neuritis is always a consideration at onset, but the presence of recurrent and relapsing optic nerve disease, typically in the absence of other neurological foci, and the lack of typical demyelinating lesions on MRI scan usually eliminate this from diagnostic consideration. The absence of systemic disease and known neurological syndrome eliminates many elements of the differential, although sarcoidosis can be occult and requires thorough consideration. Idiopathic pachymeningitis often has other neurological signs, such as headache, cranial nerve palsies and ataxia, and often has spinal fluid and MRI findings that distinguish it from AON.11 The lack of a temporal link to pregnancy eliminates lymphocytic adenohypophysitis. Retinal disease (paraneoplastic retinal degenerations and ARRONS syndrome) as well as paraneoplastic optic nerve disease can all resemble AON but tend to have a more insidious, chronically progressive, rather than episodic/relapsing course. Nonetheless, assessment with electrophysiology and assays for autoimmune disease against the retina and optic nerve may be indicated.12 13 The borderline between cases of NMO with optic neuropathy14 and AON remains to be defined. Patient 3 demonstrated the clinical course seen in AON, had no other neurological deficits and had a skin biopsy of normal skin read as consistent with vasculitis; thus, he was treated as AON. Nonetheless, years after treatment, he was tested for NMO and was positive. Certainly, most of the cases of AON have not featured the spinal cord disease seen in many of the cases of NMO, but the assay for NMO has not been available for long enough to have tested adequate numbers of AON patients at this point. Br J Ophthalmol 2009;93:1660–1666. doi:10.1136/bjo.2009.159350

Chronic relapsing inflammatory optic neuropathy (CRION), an entity with a similar clinical course to AON, has been reported by Kidd and colleagues.15 In their series, they describe 15 patients over 10 years at two institutions who were characterised by: 1. steroid-dependent, relapsing optic neuropathy; 2. pain with onset; 3. all cases ultimately bilateral, often with virtually simultaneous onset; 4. disc abnormalities common, often swollen when acute, atrophic later; 5. MRI abnormalities of the brain not typically seen, whereas 19/30 imaged optic nerves were abnormal (63%). These cases had workups for sarcoidosis, including Kveim testing, which were negative in the six so tested, and a gallium scan, which is a useful test in trying to diagnosis optic neuritis Table 6 Comparison of features of autoimmune optic neuropathy and chronic relapsing inflammatory optic neuropathy

Symptom/sign/test Ocular pain/headache Disc oedema (eyes) MRI: brain abnormalities MRI: optic nerve abnormalities Abnormal antinuclear antibody Elevated erythrocyte sedimentation rate

Autoimmune optic neuropathy

Chronic relapsing inflammatory optic neuropathy

4/9 (44%) 1/18 (6%) 4/9 (44%) 7/18 nerves (39%)

14/15 17/30 1/15 19/30

(93%) (57%) (7%) nerves (63%)

4/9 (44%)

1/15 (7%)

5/9 (56%)

1/15 (7%)

1663


Clinical science due to occult sarcoidosis,16 negative in the five patients examined. Furthermore, when followed longitudinally, none of their cases ultimately developed sarcoidosis. A comparison of the clinical and laboratory features of CRION and the current series of AON cases is given in table 6. There is no mention of anticardiolipin antibodies or skin biopsies being performed in these CRION cases. Although the frequency of pain, and the swollen discs and absence of serological markers likely make this a distinct entity from AON, it would be interesting to see the results of skin biopsies and anticardiolipin antibodies in such cases. Although ACA is a serological marker for AON, the relationship of AON to the antiphospholipid antibody syndrome (APA) is not clear. We are aware of only one case of overlap between AON and APA. This case is the youngest case of AON reported (case 4 in this series). Details of her case are published elsewhere.4 Here, ACA IgM was repetitively elevated, and her skin biopsy showed thrombin deposition in vessels, consistent with anticardiolipin antibody syndrome. Immunofluorescence showed subtle IgM granular deposits and moderate C3D granular deposits at the dermal–epidermal junction, consistent with AON. Normally, such a case would be treated with immunosuppression, but since she was only 8 years old, we wished to avoid a therapy with long-term potential for causing sterility or malignancy. Thus, we treated her with a daily aspirin and monthly IVIG. Her NMO antibody was tested years after

Table 7 Recommended laboratory/radiological evaluation of suspected autoimmune optic neuropathy Blood studies Complete blood count and differential Venereal Disease Research Laboratory Fluorescent treponemal antibody absorption Erythrocyte sedimentation rate C3, C4, CH50 Immunoglobulin levels/subclasses Antinuclear antibody Rheumatoid factor Angiotensin-converting enzyme Antineutrophilic cytoplasmic antibodies Prothrombin time, partial thromboplastin time Lupus anticoagulant Anticardiolipin antibodies Anti-neuromyelitis optica antibodies Radiological studies Chest x ray MRI of the brain and optic nerves, with and without contrast, including fluidattenuated inversion recovery sequences and fat-suppressed views of the optic nerve Cutaneous testing Purified protein derivative/anergy panel Skin biopsy for H&E and immunofluorescence Miscellaneous testing Urinalysis Lumbar puncture

Table 8 Characteristics of autoimmune optic neuropathy cases reported in the literature Patient

Age/sex

ABNL serology and/or spinal fluid

Kupersmith*12 Kupersmith 2 Kupersmith 3 Kupersmith 4 Kupersmith 5 Kupersmith 6 Kupersmith 7

44, 42, 26, 26, 56, 47, 45,

ANA C3, C4; CSF: oligoclonal bands ANA, anti-double-stranded-DNA ANA, C3 ANA, anti-RNP, C4, ENA; CSF: oligoclonal bands ANA; CSF: elevated protein ANA ACA IgM, ANA, anti-RNA

Kupersmith 8

27, female

ANA, C3, C4; CSF: lymphocytosis

Kupersmith 9 Kupersmith 10

43, female 44, male

ANA ACA IgM, anti-RNA, C3

Kupersmith 11

30, female

ACA IgM

Kupersmith 12 Kupersmith 13

40, female 45, female

ANA, RF ACA IgM, ACA IgA

Kupersmith 14

47, female

ANA, RF

Riedel 17

17, male

None

Riedel 2

25, female

ACA IgM

female female female female female female male

MRI

Swollen chiasm

Central white matter and periventricular signal increase Retrobulbar R ON enhancement

Scar-like lesion in thalamus, mild enhancement of R intraorbital ON and mild enlargement of lacrimal glands

Skin biopsy

Nl H&E; IF: granular IgM deposits at the epidermal–dermal junction Nl H&E; IF: granular IgM deposits in the collagen of the dermis

Immunotherapy CLB CLB CPM, AZA CLB, AZA AZA AZA CLB, AZA, CLB CLB, AZA

Steroids only Nl H&E; IF: IgG deposits around epidermal Steroids only blood vessels and at the dermal–epidermal junction, IgG and fibrin within papillary blood vessels Nl H&E; IF: IgG deposits around epidermal Steroids only blood vessels and at the dermal–epidermal junction Steroids only Nl H&E; IF: linear IgG, C3, and IgA at Steroids only the epidermal–dermal junction H&E: leucoclastic vasculitis; IF: nl Steroids only

Steroids only H&E: mononuclear perivascular infiltrate in dermis; IF: IgM and C3 in granular pattern in walls of small blood vessels and within superficial papillary dermis and at dermal– epidermal border Steroids only IF: IgM and C3 granular staining of dermal–epidermal junction and focal granular deposits in small vessel walls

If abnormality was reported at any point in the follow-up, it was noted in the chart. *Kupersmith cases 1–3 are the three cases from the Dutton series. ACA, anticardiolipin antibodies; ANA, antinuclear antibody; AZA, azathioprine; CLB, chlorambucil; CPM, cyclophosphamide; CSF, cerebrospinal fluid; IF, immunofluorescence; Nl, normal result; ON, optic nerve; RF, rheumatoid factor.

1664



Clinical science Table 9 Immunosupressants/immunomodulating agents Agent

Major action(s)

Major toxicities

Azathioprine

Purine synthesis inhibitor

Chlorambucil

Alkylating agent

Cyclophosphamide

Alkylating agent

Methotrexate

Antimetabolite/antifolate

Mycophenolate mofetil

Purine synthesis inhibitor

Intravenous immunoglobulin

Numerous, including: Blockage of macrophage Fc receptors Modulation of complement activation Suppression of inflammatory mediators

Susceptibility to infection Bone-marrow suppression Gastrointestinal dysfunction Carcinogen (especially acute non-lymphocytic leukaemia) Susceptibility to infection Amenorrhoea Seizures Bone-marrow suppression Nausea and vomiting Haemorrhagic cystitis Susceptibility to infection Bone-marrow suppression Susceptibility to infection Bone-marrow suppression Liver toxicity Lung toxicity Renal toxicity Bone-marrow suppression Susceptibility to infection Thrombosis Diarrhoea Vomiting Progressive multifocal leucoencephalopathy Allergic reactions Anaphylaxis Acute renal failure Thrombosis Aseptic meningitis Shortness of breath Chest pain

IVIG treatment was discontinued and was negative. Whether she will go on to have the full clinical manifestations of antiphospholipid antibody syndrome or other CVD remains to be seen; she has not shown any systemic illness during 10 years of follow-up. The laboratory evaluation of suspected AON cases focuses on detecting occult systemic inflammatory disease and is summarised in table 7. Table 8 summarises the laboratory, radiological and pathology findings for previously published cases of AON with specified treatment regimens. Two of Dutton’s cases were treated with corticosteroids followed by chlorambucil; the third received both azathioprine and chlorambucil.1 Kupersmith2 and colleagues reported a series of 14 cases of AON from two institutions, including the three original cases from the Dutton series. All patients presented with visual loss and had serological evidence of a collagen vascular process without systemic symptoms. Eighty-six per cent of these cases were female. The mean age was 40 (range 26 to 56). Seven (50%) had prior neurological or systemic symptoms, the latter not being sufficient to diagnose a defined autoimmune illness. The optic neuropathy was bilateral in 9/14 (64%), and was typically progressive and painless. The ANA was abnormal in 11/14 (79%), rheumatoid factor was abnormal in 2/14 (14%), and the Raji cell assay abnormal in 1/3 (33%). Anticardiolipin antibodies were abnormal in 4/6 (67%) individuals. Skin biopsies were performed in seven cases and were abnormal in six (86%) Treatment with steroids was administered to 11 cases, 10 receiving 1–2 g of methylprednisolone daily and one receiving 400 mg prednisone daily. Two others initially received Br J Ophthalmol 2009;93:1660–1666. doi:10.1136/bjo.2009.159350

80–100 mg of prednisone. Ten patients failed corticosteroids with either a flare of optic neuropathy or with intolerance of the drug. Five of these received chlorambucil as the first alternate agent, with the dose adjusted to achieve a white cell count below 4000. Four of these cases did well, while the fifth required azathioprine to be added due to fluctuating vision. One patient received cyclophosphamide as initial therapy, and as prednisone was being tapered, the optic neuropathy flared, necessitating the addition of azathioprine. Two cases were started on azathioprine as the first alternate therapy, one patient was controlled, and one refused further immunosuppression after 1 week. Riedel and colleagues at the University of Iowa described two cases of AON.17 Both were treated with corticosteroids, and the second case was notable for undergoing an optic nerve biopsy due to suspicion of sarcoidosis. This showed perivascular nongranulomatous inflammation and mild gliosis. She underwent a biopsy of non-lesional non-sun exposed skin, which suggested autoimmune connective tissue disease (IgM and C3 granular staining of the dermal–epidermal junction with similar focal deposits in small vessel walls). This finding is not typically seen in normal skin, with the incidence in normals being 0–1.7%.18 Despite our current data and the experience of others, the decision as to what is the best therapy for AON remains scientifically unproven due to the limited number of cases. Originally, when corticosteroids alone would not control the process, or if the patient could not tolerate the dose of corticosteroids requisite to preserve vision, we followed Dutton’s initial therapy, which was chlorambucil after corticosteroid pulse (many of these agents require weeks to reach the 1665


Clinical science peak immunosuppressive effect). If time is crucial, such as when blindness had occurred in the second eye, intravenous cyclophosphamide is a reasonable choice due to the rapidity of onset of its therapeutic effect. In later cases, with potential late effects of chlorambucil, methotrexate, which was safer and helpful in many inflammatory optic neuropathies, was used as the firstline agent. However, the two cases in this series who received methotrexate as their initial immunomodulatory agent both relapsed. Chlorambucil seems to have a better chance at eradicating as opposed to suppressing the disease activity, but its use carries the risk of greater potential side effects. IVIG might also be an appropriate initial therapy, especially if there is a need to avoid immunosuppression.The mechanism of action and major toxicities of several of the immunosuppressive/ immunomodulatory agents are listed in table 9 We generally treat the patient with both corticosteroids and immunosuppression or IVIG for 6–12 months, with an initial bolus of intravenous solumedrol if there has been recent disease activity. If the patient does well on dual therapy, we wean them off the corticosteroids over 6–12 months. If the patient does well on immunosuppressants or IVIG alone, we typically leave them on this regimen for a year and then attempt to wean them off over about 6 months. If the patient develops visual loss during the steroid taper, we generally rebolus with corticosteroids and switch to a different immunosuppressive agent, as it is clear that the agent initially selected was not completely controlling the disease process. This series does not address the vexing cases of recurrent steroid-dependent optic neuropathy that have neither serological markers nor abnormal skin biopsies. Although, in these cases, we may suspect they are autoimmune in aetiology, we do not have good evidence for this and are typically loathe to biopsy an optic nerve unless it has very poor vision for a long period of time. Nonetheless, we typically treat these similarly to the AON cases. Hopefully, in time, diagnostic markers for these cases will be discovered. Patients who have unexplained optic neuropathy, especially if bilateral or recurrent, may benefit from a non-lesional non-sun exposed skin biopsy, which may be diagnostic of AON, even in the absence of abnormal serological studies. When immunoreactant or other changes supporting AON are present, one has evidence that the disease process affects more than just the optic nerve, thereby supporting the use of systemic therapy. Maintaining a high index of suspicion for these cases will lead to earlier diagnosis, unlike when the syndrome was first

1666

described; many of the cases are now diagnosed and treated with much vision preserved. As no single institution has adequate experience in following these cases, a disease registry for AON should be created to better define the natural history and to determine the best therapy for this disorder. Funding: Supported by an unrestricted grant from Research to Prevent Blindness, Inc, and the Lions Eye Research Foundation of New Jersey. Competing interests: None. Ethics approval: Ethics approval was provided by New Jersey Medical School IRB. Provenance and peer review: Not commissioned; externally peer reviewed.

REFERENCES 1. 2. 3. 4.

5. 6. 7.

8. 9.

10. 11. 12. 13. 14.

15. 16. 17. 18.

Dutton, JJ, Burde RM, Klingele TG. Autoimmune retrobulbar optic neuritis. Am J Ophthalmol 1982;94:11–17. Kupersmith MJ, Burde R, Warren F, et al. Autoimmune optic neuropathy: evaluation and treatment. J Neurol Neurosurg Psych 1988;51:1381–6. Bielory L, Kupersmith MJ, Warren F, et al. Skin biopsies in the evaluation of atypical optic neuropathies. Ocul Immunol Inflamm 1993;1:231–41. Frohman L, Turbin R, Bielory L, et al. Autoimmune optic neuropathy with anticardiolipin antibody mimicking multiple sclerosis in a child. Am J Ophthalmol 2003;136:358–60. Colglazier CL, Sutej PG. Laboratory testing in the rheumatic diseases: a practical review. South Med J 2005;98:185–91. Sammaritano LR. Antiphospholipid syndrome: review. South Med J 2005;98:617–25. Witte T, Hartung K, Matthias T, et al. Association of IgA anti-dsDNA antibodies with vasculitis and disease activity in systemic lupus erythematosus. Rheumatol Int 1998;18:63–9. Ravel R. Clinical laboratory medicine: clinical application of laboratory data. 6th edn. Philadelphia: Elsevier Health Sciences, 1994:370. Neff RT, Hurst FP, Falta EM, et al. Progressive multifocal leukoencephalopathy and use of mycophenolate mofetil after kidney transplantation. Transplantation 2008;86:1474–8. Chung JH, Van Stavern GP, Frohman LP, et al. Adalimumab-associated optic neuritis. J Neurol Sci 2006;244:133–6. Kupersmith MJ, Martin V, Heller G, et al. Idiopathic hypertrophic pachymeningitis. Neurology 2004;62:686–94. Thirkill CE. Cancer-induced, immune-mediated ocular degenerations. Ocular Immunol Inflamm 2005;13:119–31. Mizener JB, Kimura AE, Adamus G, et al. Autoimmune retinopathy in the absence of cancer. Am J Ophthalmol 1997;123:607–18. Weinshenker BG, Wingerchuk DM, Vukusic S, et al. Neuromyelitis optica IgG predicts relapse after longitudinally extensive transverse myelitis. Ann Neurol 2006;59:566–9. Kidd D, Burton B, Plant GT, et al. Chronic relapsing inflammatory optic neuropathy (CRION). Brain 2003;126(Pt 2):276–84. Frohman LP, Guirgis M, Turbin RE, et al. Sarcoidosis of the anterior visual pathway: 24 new cases. J Neuroophthalmol 2003;23:190–7. Riedel P, Wall M, Grey A, et al. Autoimmune optic neuropathy. Arch Ophthalmol 1998;116:1121–4. Blenkinsopp WK, Clayton RJ, Haffenden GP. Immunoglobulin and complement in normal skin. J Clin Pathol 1978;31:1143–6.



Clinical characteristics, diagnostic criteria and therapeutic outcomes in autoimmune optic neuropathy L Frohman, K DellaTorre, R Turbin, et al. Br J Ophthalmol 2009 93: 1660-1666 originally published online August 18, 2009

doi: 10.1136/bjo.2009.159350

Updated information and services can be found at: http://bjo.bmj.com/content/93/12/1660.full.html

These include:

References

This article cites 17 articles, 5 of which can be accessed free at: http://bjo.bmj.com/content/93/12/1660.full.html#ref-list-1

Email alerting service

Topic Collections

Receive free email alerts when new articles cite this article. Sign up in the box at the top right corner of the online article.

Articles on similar topics can be found in the following collections Neurology (1053 articles) Optic nerve (561 articles)

Notes

To request permissions go to: http://group.bmj.com/group/rights-licensing/permissions

To order reprints go to: http://journals.bmj.com/cgi/reprintform

To subscribe to BMJ go to: http://group.bmj.com/subscribe/