Діагностика та лікування оптикомієліту — EFNS 2010


  • demyelinating diseases;
  • diagnosis;
  • longitudinally extensive transverse myelitis;
  • neuromyelitis optica;
  • recurrent optic neuritis;
  • treatment


Background and purpose:  Neuromyelitis optica (NMO) or Devic′s disease is a rare inflammatory and demyelinating autoimmune disorder of the central nervous system (CNS) characterized by recurrent attacks of optic neuritis (ON) and longitudinally extensive transverse myelitis (LETM), which is distinct from multiple sclerosis (MS). The guidelines are designed to provide guidance for best clinical practice based on the current state of clinical and scientific knowledge.

Search strategy:  Evidence for this guideline was collected by searches for original articles, case reports and meta-analyses in the MEDLINE and Cochrane databases. In addition, clinical practice guidelines of professional neurological and rheumatological organizations were studied.

Results:  Different diagnostic criteria for NMO diagnosis [Wingerchuk et al. Revised NMO criteria, 2006 and Miller et al. National Multiple Sclerosis Society (NMSS) task force criteria, 2008] and features potentially indicative of NMO facilitate the diagnosis. In addition, guidance for the work-up and diagnosis of spatially limited NMO spectrum disorders is provided by the task force. Due to lack of studies fulfilling requirement for the highest levels of evidence, the task force suggests concepts for treatment of acute exacerbations and attack prevention based on expert opinion.

Conclusions:  Studies on diagnosis and management of NMO fulfilling requirements for the highest levels of evidence (class I–III rating) are limited, and diagnostic and therapeutic concepts based on expert opinion and consensus of the task force members were assembled for this guideline.


To provide guidelines for best practice diagnosis and management of adult neuromyelitis optica based on the current state of clinical and scientific knowledge.

 Neuromyelitis optica (NMO, also known as Devic’s disease) is a severe idiopathic immune-mediated demyelinating and necrotizing disease that predominantly involves optic nerves and spinal cord. Cases of NMO have been reported in all continents and races, but ethnic variations suggest that genetic factors are relevant [1–10]. In Europe, the prevalence of NMO among autoimmune disorders of the central nervous system (CNS) is lower compared with multiple sclerosis (MS). However, NMO makes up a substantial proportion of inflammatory demyelinating disorders of the CNS in non-Caucasian populations such as Afro-Brazilians (15%), East Asians (up to 48%) and Indians (9%), which is most likely due to divergent prevalence of MS [9,11–13].

The association between optic neuritis (ON) and spinal cord impairment was first described by Sir Clifford Albutt in 1870 [14]. In 1894, Eugene Devic and his student Fernand Gault evaluated further cases and proposed the nature of the pathological process, named the syndrome ‘neuro-myélite optique’ or ‘neuroptico-myélite’, and discussed a relationship with MS [15,16]. However, it was not until the 1990s that further clinical and histopathological studies changed the concept and place of NMO within the expanding range of autoimmune disorders of the CNS.

Historically, the diagnosis of NMO was restricted to a monophasic course of bilateral ON and myelitis as well as cases with short intervals between the index events. More recently, NMO has been recognized as a recurrent autoimmune CNS disorder with clinical, neuroimaging and laboratory findings that are distinct from MS. Accordingly, NMO immunoglobulin G (NMO-IgG, referred to as aquaporin 4 antibody when using antigen-specific detection techniques), an autoantibody that binds to the water channel aquaporin 4 (AQP4), in combination with diagnostic criteria support distinction of NMO from other autoimmune disorders of the CNS [17]. NMO-IgG/AQP4 antibodies are also detected in NMO spectrum disorders, which include (i) spatially limited forms such as longitudinally extensive transverse myelitis (LETM) and recurrent isolated optic neuritis (RION)/bilateral optic neuritis (BON) [18,19], (ii) NMO in the context of organ and non-organ-specific autoimmune diseases [20,21], (iii) atypical cases with clinically manifest or subclinical brain lesions and (iv) Asian opticospinal MS (OSMS) [22].

The natural history of untreated NMO is significantly worse than that of MS with acquisition of residual disability from initial relapses in the majority of patients. Hence, NMO requires early recognition and concepts for treatment of acute attacks and long-term disease modification. However, studies on diagnosis and evaluation of treatments are scarce and most diagnostic and therapeutic recommendations reflect consensus opinion of individual experts in the field.


Evidence for this guideline was collected by searches for original articles, case reports and meta-analyses in the MEDLINE and Cochrane databases by employing relevant keywords, combinations and abbreviations. The period for eligible articles ranged from 1965 to September 2009, and following suggestions from the reviewers also individual articles beyond this period. Clinical practice guidelines were searched for within databases of the European Federation of Neurological Societies (EFNS, https://www.efns.org/EFNS-Guideline-Papers.270.0.html, the American Academy of Neurology (AAN, https://www.aan.com/go/practice/guidelines), the American College of Rheumatology (ACR, https://www.rheumatology.org/index.asp) and the British Society for Rheumatology (https://www.rheumatology.org.uk).

Scientific evidence for diagnostic investigations and treatments was evaluated according to pre-specified levels of certainty (class I, II, III, and IV) [23]. The recommendations were graded according to the strength of evidence (grade A, B or C), using definitions given in the EFNS guidance. When sufficient evidence for recommendations A–C was not available, we gave a recommendation as a ‘best practice recommendation’ if agreed by all members of the task force.

The task force was initiated by the EFNS Scientific Panel on Multiple Sclerosis/Demyelinating Disorders and commissioned by the Scientific Committee of the EFNS. JS and BH wrote the draft. Consensus was reached after three rounds of circulating questionnaires and drafts to the task force members.


Diagnostic strategies

Demographics and disease-specific features

Ethnic variations in the condition suggest that genetic factors are important, and rare familial cases have been reported (class IV) [24,25]. NMO primarily affects non-whites and populations with a minor European contribution to their genetic composition (class IV) [13,26–29]. Among a cohort of 850 patients with demyelinating disorders in North East Tuscany, the prevalence of NMO spectrum disorders was 1.5% (N = 13), the MS:NMO ratio was 42.7:1 (class IV) [30]. Similar neuroimaging, autoantibody and immunopathological characteristics of NMO and Asian opticospinal MS support the hypothesis that the latter, or a subset, is identical to Western NMO (class IV) [9,29,31]. Yet Japanese opticospinal MS was associated with the human leucocyte antigen (HLA) HLA class II DPB1*0501 allele, which could not be confirmed in a European NMO cohort (class III) [1,10,32]. Moreover, different diagnostic criteria for Japanese and Western forms impede a detailed comparison (class IV) [22].

In monophasic NMO [no recurrence, simultaneous or closely related ON and LETM (<30 days)], men and women are affected equally but in the more frequent recurrent disease course (80–90%), women (ratio 5–10:1) are overrepresented (class IV) [2,4,7,8,29,33,34]. The age of onset in NMO ranges from childhood to late adulthood with the incidence tapering off after the fifth decade, the median onset age is in the late 30s [35–38] (class IV). In MS, 70% of first manifestations occur between the age of 20 and 40 years (median onset at around 30 years of age) [39].

The majority of patients suffer from a recurrent course (80–90%) whilst monophasic (10–20%) and primary or secondary progressive courses are rare (class IV) [2,8,40,41]. Relapses usually occur early, in clusters and unpredictable intervals (class IV) [8,40]. In the Mayo Clinic series, the second relapse occurred within 1 year in 60%, within 3 years in 90%, and also decades after the index events (class IV) [8]. Likewise, in a Cuban NMO study, the next relapse developed after a mean time of 15 months (range 1–158 months) from the index events (class IV) [33]. In contrast, among Mexican patients with NMO who were followed up for more than 3 years, 82% (14/17) still had a monophasic course (class IV) [7]. Predictors of a recurrent course were a longer inter-attack interval between the first two clinical episodes, older age at onset, female gender and less severe motor impairment at the sentinel myelitis event (class IV) [40].

There are no clinical features that are disease-specific. Clinical manifestations include involvement of optic nerve and/or spinal cord and also many other symptoms and signs. Resulting from a cumulative attack-related injury of the CNS, the natural history of NMO is characterized by a stepwise deterioration of motor, sensory, visual and bowel/bladder function (class IV) [8,33,40]. Severe attacks of myelitis or ON should raise awareness for NMO. Most relapses worsen over several days and then slowly improve in the weeks or months after the maximum clinical deficit is reached (class IV) [8]. An antecedent viral illness was reported in 30% of patients with monophasic and 23% of patients with recurrent NMO (class IV) [8].

Conclusions:  NMO is a clinical entity different from MS, is characterized by a worldwide occurrence with ethnic variations, and is – compared to MS – a relatively rare disorder in Europe. Most frequently, NMO develops as a recurrent disorder, which predominantly affects women.

Best practice recommendations:  In the diagnosis of NMO, it is recommended to take following disease characteristics into consideration: The predominant course is characterized by recurrent severe attacks of myelitis and/or uni- or bilateral ON with incomplete recovery and is up to 10 times more prevalent in women than in men. Age of onset, the late 30s, is approximately 10 years later than age of onset in MS, but NMO may also occur in children and elderly people.

Clinical features

Optic neuritis  Visual loss is generally more severe in NMO than in MS (class IV) [42]. The occurrence of bilateral simultaneous ON or sequential ON in rapid succession is more suggestive of NMO (class IV) [40]. Other clinical features of ON including pain, pattern of visual loss, occurrence of positive visual phenomena such as movement-induced phosphenes and findings on examination do not differ from MS-related ON attacks (class IV) [8]. Blindness in at least one eye developed in 60% of recurrent (mean follow-up time 16.9 years) and 22% of patients with monophasic NMO (mean follow-up time 7.7 years) (class IV) [8]. Ophthalmoscopic examination may be normal or exhibit signs of ON, and optic atrophy with disc pallor is typically more pronounced than in MS. In addition, demyelination and necrosis are predominantly seen at the centre of the nerve and may form cavitation (class IV) [43,44].

Visual field testing typically reveals central scotoma, although other visual field changes such as colour blindness, bitemporal hemianopsia, paracentral scotoma and altitudinal deficits are possible. Optic coherence tomography studies in NMO reported a thinner retinal fibre layer than in MS, indicating a more widespread axonal injury (class IV) [45,46].

Myelitis  Spinal cord involvement usually presents in the form of complete transverse myelitis with para- or tetraparesis, an almost symmetrical sensory level and sphincter dysfunction (class IV) [2,8,34,47]. In contrast, spinal cord symptoms in MS are milder and asymmetric and caused by acute partial transverse myelitis (class IV) [48,49]. Radicular pain, paroxysmal tonic spasms and Lhermitte′s signs develop in 1/3 of the recurrent cases, but are rare or absent in patients with monophasic NMO (class IV) [8]. Nausea and intractable hiccups were found in 8/47 (17%) of patients with recurrent NMO due to expansion of the lesion to the medulla (class IV) [50]. Other brainstem symptoms include vomiting, vertigo, hearing loss, facial weakness, trigeminal neuralgia, diplopia, ptosis and nystagmus (class IV) [8,51,52]. Due to involvement of medullary centres of neuromuscular respiration control, neurogenic respiratory failure and subsequent death can occur (class IV) [8].

Other manifestations  Central nervous system involvement beyond optic nerve and spinal cord/brainstem is observed in about 15% of the patients with NMO and include encephalopathy, hypothalamic dysfunction and cognitive impairment (class IV) [8,53,54]. The latter is present in NMO with a similar frequency as reported in MS (class IV) [55]. Magana et al. reported five NMO-IgG seropositive women (N = 3 NMO, N = 2 recurrent LETM), who developed confusion and depressed consciousness consistent with posterior reversible encephalopathy syndrome (PRES) (class IV) [53]. The immune-mediated disruption of AQP4 water channel function may play a central role in the pathogenesis of PRES in these patients. Endocrinopathies associated with NMO include amenorrhea, galactorrhea, diabetes insipidus, hypothyroidism or hyperphagia (class IV) [56].

Outcome  The index events (visual acuity, motor, sensory and sphincter dysfunction) are more severe in monophasic than recurrent NMO (class IV) [40]. Repeated NMO attacks are the main cause of accumulation of neurological impairment, whereas permanent disability in MS is primarily a feature of secondary progression. In a Brazilian cohort, incomplete recovery from the index event predicted future disability (class IV) [4].

A history of other autoimmune diseases, higher attack frequency during the first 2 years of disease and better motor recovery following the index myelitis event were associated with increased risk of fatality in recurrent NMO (class IV) [40]. In this study published in 2003, which may have been biased towards more severe and complicated cases, 32% of patients with recurrent NMO died (median follow-up time 60.2 months) whilst no death occurred in patients with monophasic NMO [40]. There were 24/96 (25%) deaths in the French West Indies cohort, and predictors for mortality were higher attack frequency during the first year of disease, blindness and sphincter signs at onset (class IV) [57]. Mortality in a cohort of recurrent NMO from Brazil (published in 2002) was even higher (50%) (class IV) [12], but recent progress in understanding the disease and its management are likely to have decreased mortality rates.

Conclusions:  A recurrent course with rapid development of further events is frequent. Individual episodes of ON/LETM are severe, and permanent disability is more attack-related in NMO than in MS. CNS involvement beyond optic nerve and spinal cord/brainstem need to be taken into account. NMO is associated with a more detrimental short- and long-term outcome than MS, and neurogenic respiratory failure is the most frequent cause of death. Prognostic factors need to be confirmed in independent prospective studies.

Best practice recommendations:  ON with severe visual loss and in rapid succession may be indicative of NMO. Optic atrophy is more pronounced than in MS and may form cavitation. Complete transverse myelitis is typical for NMO, and partial transverse myelitis syndromes are more indicative of MS. Expansion of the spinal cord lesion may lead to brainstem symptoms and life-threatening complications.


Optic nerve  There are no studies evaluating differences in magnetic resonance imaging (MRI) presentation of ON in the setting of NMO and MS. Short-tau inversion recovery (STIR) sequences provide fat suppression, which is advantageous for evaluation of the optic nerve. Using STIR sequences, increased signal intensity on T2-weighted scans of the optic nerve is reported in acute ON in 84% and during remission in 20% (class IV) [58]. Disruption of the blood-nerve barrier leads to gadolinium-enhancement on T1-weighted spin echo sequences. Gadolinium-enhancement is a sensitive finding in acute ON (94%) (class IV) [59], is of variable extent and can occasionally extend into the optic chiasm. In a Cuban study in patients with long-duration NMO, gadolinium-enhancement of the optic nerve was found in 32.5% (class III) [60].

Spinal cord  Spinal cord lesions extending over three or more vertebral segments are the most reliable finding for the diagnosis of NMO (class IV) [52]. However, normal appearances or shorter lesions can be found very early during relapse or in residual atrophic stage [52]. Lesions are predominantly located in the cervical and thoracic cord with a central grey matter pattern, reflected by hyperintensities on T2-weighted axial scans and corresponding T1-hypointensities (class IV) [47,61]. Cervical lesions may extend to the lower medulla. Cavity-like longitudinally extensive lesions are seen in cases of severe disease. Acute spinal cord lesions tend to occupy most of the cross-sectional area of an affected segment and are associated with swelling and gadolinium-enhancement (detectable days to months following relapse) (class IV) [2,62,63]. In a study on LETM relapses in NMO (N = 11), gadolinium-enhancement vanished in all patients following treatment with high-dose methylprednisolone, and lesions may almost entirely disappear during remission (class IV) [61].

Conversely, spinal cord lesions expanding over two or more vertebral segments are rarely found in MS [48]. Lesions in MS usually appear as short-segment, asymmetric and often posterior cord lesions (class IV) [49,64]. ‘Snake-eye’ or ‘owl-eye’ sign is a common feature of spinal artery ischaemia and may be a transient finding at an early stage of NMO (class IV) [61].

Brain  Normal brain MRI is initially present in 55–84% of patients with NMO although a development of cerebral white matter lesions can be expected over the course of the disease (class IV) [2,8,34]. Indeed, brain MRI lesions can be detected with serial scans in up to 84.8% of patients with NMO (class IV) [4,51,65–67]. Distribution of NMO-typical brain lesions (8/120 patients) corresponded to structures with high AQP4 expression such as ependymal cells, hypothalamus and brainstem (class IV) [51,65]. Presence of brain lesions may be more common in childhood NMO. A cross-sectional study reported that brain lesions were found in 68% and were predominantly involving the periventricular region (class IV) [68].

Most lesions are non-specific and asymptomatic; Pittock et al. reported brain MRI lesions in 60% of patients, of which 10% were rated MS-like and fulfilled the Barkhof criteria for dissemination in space (class III) [51]. Two or more brain MRI lesions were found in 66% of a Cuban NMO cohort, but presence of lesions did not correlate with disease severity (class IV) [60]. In a Chinese study, most supratentorial lesions were rated as non-specific punctate or small round dot, and located in juxtacortical, subcortical and deep white matter regions (class IV) [67]. Callosal lesions were described in 18.2% (4/22) of AQP4 antibody-positive Japanese patients with NMO spectrum disorders (class III) [69]. These callosal lesions were classified as acute, large and oedematous in three (of four patients) and appeared with a heterogeneous lesion intensity (‘marbled pattern’), whereas callosal lesions in MS (36/56, 64.3%) were reported to be small, isolated and not oedematous. Another study described a cloud-like enhancement of brain lesions in terms of multiple patchy enhancement with blurred margin as typical of NMO (class IV) [70].

Symptomatic brain lesions are not an exclusion criterion for NMO [54,56]. Brain MRI of NMO spectrum disorder patients experiencing a PRES episode revealed bilateral T2-hyperintensities primarily in frontal, parieto-occipital and cerebellar regions (class IV) [53].

Conclusions:  The most characteristic MRI finding is a spinal cord lesion expanding over three or more vertebral segments on T2-weighted images, occupying most of the cross-sectional area, frequently hypointense on T1-weighted images and displaying gadolinium-enhancement. Brain MRI can be normal, disclose NMO-typical lesions, show non-specific white matter lesions or rarely exhibit MS-like lesions fulfilling the Barkhof criteria for dissemination in space. NMO-typical brain lesions are present in areas with high AQP4 expression such as peri-ependymal, in the hypothalamus and the brainstem. Gadolinium-enhancement within the optic nerve is a frequent finding on MRI in the setting of ON, but may also be present during remission. Typical MRI characteristics of brain lesions need to be confirmed in larger studies.

Best practice recommendations:  MRI appearance of spinal cord lesions plays a central role in the diagnosis of NMO. Brain MRI is a mainstay in the work-up and may display gadolinium-enhancement of the optic nerve and both symptomatic and asymptomatic brain lesions.

Cerebrospinal fluid analysis

Cerebrospinal fluid (CSF) abnormalities are detected in most patients with NMO and concern cell count, protein level and oligoclonal bands (OCB). Pleocytosis, usually consisting of monocytes and lymphocytes, is present in 14–79% of patients (class IV) [8,34]. CSF pleocytosis can include or be dominated by neutrophils, also eosinophils may be found (class IV) [8,71]. CSF cell count is greater than 50 cells/μl in 13–35% of patients and in a few cases up to 1000 cells/μl (class IV) [8,34,72]. Patients with LETM are more likely to exhibit a pleocytosis than patients with ON (class IV) [73]. Increased protein levels are present in 46–75% of cases (class IV) [2,34]. The frequency of OCB in NMO ranges from 0% to 37% and the presence of OCB can be transient in NMO in contrast to MS (class IV) [2,4,8,34,74]. CSF analysis in MS rarely reveals a pleocytosis exceeding 50 cells/μl, and OCB are present in over 90% in established MS (class III) [74]. One study found a positive MRZ (measles, rubella and zoster) reaction, as defined by a combination of at least two positive antibody indices, in MS in 37/42, but in NMO in only 1/20 (class III) [75].

Neurofilaments are released into the CSF following axonal injury. Neurofilament heavy chain (NfH) levels are significantly higher in the CSF of NMO than in patients with MS (class III) [76]. Also, glial fibrillary acidic protein (GFAP) is significantly higher in the CSF of patients with NMO compared with MS, spinal cord infarction and acute disseminated encephalomyelitis (ADEM) (class III) [77].

Conclusions:  Analysis of CSF provides supportive data for the diagnosis of NMO and CSF should be obtained during or shortly after an acute attack. The findings are useful, but not highly sensitive or specific. The value of NfH and GFAP determination needs to be evaluated in further studies.

Best practice recommendations:  CSF findings with a lymphomononuclear pleocytosis >50 cells/μl, occasional presence of neutrophils/eosinophils, and lack of OCB may be indicative of, but not specific for, NMO and NMO spectrum disorders.

Electrophysiological evaluation

There are only a few studies reporting electrophysiological examinations in NMO. An Australian study revealed abnormal visual evoked potentials (VEP) more frequently in opticospinal demyelinating disease than in conventional MS (85% and 71.4%, respectively) (class IV) [27]. In a multi-ethnic Cuban NMO study, somatosensory evoked potentials (SEP) were abnormal in 86% (42/49), VEP in 83% (44/53) and brainstem acoustic evoked potentials (BAEP) in 37% (19/51) (class IV) [33]. BAEP abnormalities were more frequent in blacks than in other patients (78% vs. 29%P = 0.003). Peripheral motor and sensory nerve conduction was normal in all patients of a Japanese study with nine opticospinal MS patients (class IV) [78].

Conclusions:  VEP, SEP and BAEP examination in patients with NMO frequently reveal abnormalities whilst peripheral nerve conduction studies are expected to be normal.

Aquaporin 4

Expression of AQP4 within the CNS  Aquaporin 4 is an osmosis-driven, bidirectional water channel that belongs to the subfamily of mammalian aquaporins. In the CNS, AQP4 is expressed at the astrocytic foot processes in close vicinity to the basement membranes, in the optic nerve, in a subpopulation of ependymal cells, in hypothalamic nuclei and in the subfornical organ [79,80]. In NMO, the third extracellular loop of AQP4 is considered as the major epitope for AQP4 antibodies [81]. Astrocytes were shown to undergo necrosis in a complement-dependent manner when exposed to AQP4 antibody containing sera, and disease was induced by passive transfer of antibodies to rats, suggesting a primary pathogenic role of AQP4 antibodies in NMO [82,83]. A recent study evaluated the plasma cell population taken from an early patient with NMO at the molecular level and reported that AQP4-specific IgG is synthesized intrathecally at disease onset and contributes directly to CNS pathology [84].

NMO-IgG/AQP4 antibody assay  The diagnosis of NMO and its distinction from MS has been facilitated by the discovery of NMO-IgG/AQP4 antibodies. The initial assay was based on an indirect immunofluorescence (IIF) method using mouse cerebellum (class II) [17]. With this assay, testing for NMO-IgG had a reported sensitivity of 58–76% and a specificity of 85–99% for NMO. In addition, there are four other assay techniques allowing detection of AQP4 antibodies including cell-based assays (CBA), radioimmunoprecipitation assays, fluoroimmunoprecipitation assays (FIPA) and enzyme-linked immunosorbent assays (ELISA) [85,86]. Sensitivities and specificities of the assays differ and the gold standard remains to be elucidated [85–91]. Cell-based techniques such as CBA may show the best results (class IV) [86].

NMO-IgG/AQP4 antibody and disease characteristics  Neuromyelitis optica-IgG/AQP4 antibodies can be detected years before the onset of NMO [92]. A French study using IIF did not report differences with regard to age and onset of disease, annualized relapse rate, brain MRI findings and CSF abnormalities when comparing NMO-IgG positive and negative patients (class II) [87]. A Cuban study evaluating 48 recurrent patients with NMO (diagnostic criteria of 1999) reported a relatively low prevalence of NMO-IgG (33.3%, method: IIF); however, presence of NMO-IgG was associated with a worse course defined by more frequent relapses, occurrence of myelitis and greater attack-related disability (class II) [87]. Detection of NMO-IgG was also associated with a higher probability for >3 periventricular lesions and localization within the deep white matter, and a more extensive lesion on spinal cord images during remission. A European study reported that AQP4 antibody serum levels correlated with disease activity and were reduced by treatments with rituximab, azathioprine, cyclophosphamide (method: FIPA) (class IV) [93]. Antibody titres were attenuated by methylprednisolone and remained low during remission whilst high titres were associated with complete blindness and more extensive brain lesions (methods: CBA) (class II) [88]. The study also revealed that AQP4 antibody titres positively correlated with the length of spinal cord lesions at the nadir of exacerbations. In another study, measures of complement-mediated cell injury, but not AQP4 antibody titres, were indicative of disease severity [94]. A recent Japanese study evaluated the frequency of AQP4 antibodies in idiopathic demyelinating diseases [10] and found that the HLA-DPB1*0501 allele was associated with seropositive opticospinal MS, but not with classical MS or seronegative opticospinal MS (class II). A French NMO study reported that the presence of NMO-IgG was linked to HLA-DRB1*01*03 (majority DR3) [1]. In a cohort of 130 patients with relapsing–remitting MS, none was tested positive for NMO-IgG [95].

AQP4 antibodies in spatially limited syndromes  Neuromyelitis optica-IgG is also found in patients with LETM (37.9–50%) or RION (14.3–20%) and importantly appear to predict outcome in terms of conversion to NMO (class II) [18,19]. Detection of NMO-IgG in serum of patients with RION (5/25) was associated with a more severe initial ON episode, poor visual outcome and development of NMO (class II) [18]. Of nine NMO-IgG positive LETM patients, five (65%) experienced either another episode of myelitis (N = 4) or ON (N = 1) within 1 year (class II) [19]. The probability of detecting NMO-IgG in the serum of patients with acute partial transverse myelitis (lesion <3 vertebral segments) is low; NMO-IgG was found in 1 of 22 patients (4.5%) (class III) [96]; this patient subsequently developed recurrent LETM. Recently, four cases of AQP4 antibody negative recurrent LETM, which did not seem to be related to NMO, were reported (class IV) [97]. Of note, some NMO or spectrum disorder cases can be NMO-IgG positive and AQP4 antibody negative, or viceversa (class IV) [85]. In a series of three rapidly recurring LETM patients, AQP4 antibodies could not be detected in serum, but in CSF (class IV), confirming the diagnosis of a spatially limited NMO spectrum disorder and mandating initiation of immunosuppressive treatment [98].

Conclusions:  Presence of NMO-IgG/AQP4 antibodies supports the diagnosis of NMO (level A) and is a prognostic marker for high-risk syndromes (level A). NMO-IgG/AQP4 antibodies are routinely evaluated in serum. Whether determination of NMO-IgG/AQP4 antibodies in CSF of seronegative patients is helpful is currently unclear. NMO-IgG/AQP4 antibodies can be present in patients years before and after clinical disease activity. The ideal method for determination of the antibodies has not yet been established and testing with different methods may be reasonable in cases with high suspicion for AQP4 CNS autoimmunity. It also remains to be determined whether NMO-IgG/AQP4 antibodies are a reliable marker of disease activity and treatment response.

Recommendation:  Testing for NMO-IgG/AQP4 antibodies is an important element in the work-up of NMO and NMO spectrum disorders. Standard specimen is serum. It may be helpful to use different detection methods and determine AQP4 antibodies in CSF in highly suspicious AQP4 seronegative NMO and NMO spectrum disorders. There are different assays, and the ideal detection method remains to be elucidated.

Diagnostic criteria

Criteria for diagnosis of NMO

The following two diagnostic criteria were developed on the basis of the most recent clinical, MRI and laboratory findings and have incorporated the determination of the NMO-IgG/AQP4 antibody. Of note, the Wingerchuk criteria refer to NMO-IgG and were set up when antigen-specific assays were not available.

Revised diagnostic criteria by Wingerchuk et al. (2006) (class IV) [52]  Two absolute criteria:

  • (i) optic neuritis,

  • (ii) myelitis.

At least two of three supportive criteria:

  • (i) presence of a contiguous spinal cord MRI lesion extending over three or more vertebral segments,

  • (ii) MRI criteria not satisfying the revised McDonald diagnostic criteria for MS, and

  • (iii) NMO-IgG in serum.

These revised diagnostic criteria of 2006 and previous criteria of 1999 (excluding patients with brain lesions and without considering NMO-IgG) were applied in a series of Spanish and Italian patients with suspected NMO (n = 28) and NMO spectrum disorders (LETM and RION, n = 18), 115 patients with MS served as controls (class IV) [91]. IIF was used for determination of NMO-IgG. Compared with the criteria of 1999, the revised criteria were found to be associated with a higher specificity (83.3% vs. 25%), but slightly lower sensitivity (87.5% vs. 93.7%).

NMSS task force on differential diagnosis of MS, Miller et al. (2008) (class IV) [99]  Major criteria:

  • (i) ON in one or two eyes,

  • (ii) transverse myelitis, clinically complete or incomplete, but associated with radiological evidence of spinal cord lesion extending over three or more spinal segments on T2-weighted MRI images and hypointensities on T1-weighted images when obtained during acute episode of myelitis, and

  • (iii) no evidence for sarcoidosis, vasculitis, clinically manifest systemic lupus erythematosus (SLE) or Sjögren’s syndrome (SS), or other explanation for the syndrome.

All major criteria are required, but may be separated by an unspecified interval.

Minor criteria, from which at least one must be fulfilled consist of (i. and/or ii):

  • 1. Most recent brain MRI scan of the head must be normal or may show abnormalities not fulfilling the Barkhof criteria used for McDonald diagnostic criteria including:

    • (i) Non-specific brain T2-signal abnormalities not satisfying the Barkhof criteria for dissemination in space used in the revised McDonald criteria,

    • (ii) lesions in the dorsal medulla, either in contiguity or not in contiguity with a spinal cord lesion,

  • (iii) hypothalamic and/or brainstem lesions,

    • (iv) ‘linear’ periventricular/corpus callosum signal abnormality, but not ovoid, not extending into the parenchyma of the cerebral hemispheres in Dawson finger configuration.

  • 2. Positive test in serum or CSF for NMO-IgG/AQP4 antibodies.

NMO spectrum disorders

Neuromyelitis optica spectrum disorders comprise the spatially limited syndromes, presentations with atypical features including brain lesions and comorbidities, and OSMS. Among 13 patients NMO spectrum disorders from North East Tuscany, seven had clinically definite NMO after a follow-up time of at least 2 years, the other six (46%) remained NMO spectrum disorders (class IV) [30].

Spatially limited NMO spectrum disorders: NMO-IgG/AQP4 positive LETM and RION/BON  NMO-IgG/AQP4 positive LETM and RION/BON are limited or inaugural syndromes of NMO. For details on the risk of such patients for converting to NMO refer to the preceding paragraph. The National Multiple Sclerosis Society (NMSS) task force concluded that these limited syndromes should not qualify as NMO, even in the presence of NMO-IgG/AQP4 antibody (class IV) [99]. Currently, there are no diagnostic criteria for spatially limited NMO syndromes and hence the panel decided to propose potential guidelines for work-up and diagnosis (Fig. 1).

Figure 1.  Flow chart of panel recommendations for work-up and diagnosis of suspected spatially limited NMO spectrum disorders LETM and RION/BON.


NMO and spatially limited NMO spectrum disorders: association with comorbidities  Neuromyelitis optica and spatially limited manifestations can occur in association with organ and non-organ-specific autoimmune disorders. Hypothyroidism, pernicious anaemia, ulcerative colitis, primary sclerosing cholangitis and idiopathic thrombocytopenic purpura are organ-specific autoimmune disorders associated with NMO. Patients with NMO can be seropositive for acetylcholine receptor (AChR) antibodies and 1–2% have clinical and electrophysiological findings compatible with myasthenia gravis (MG) (class IV) [21,100,101]. Moreover, NMO and NMO spectrum disorders can occur together with endocrinopathies (class IV) [56,102–105].

Among the non-organ-specific autoimmune diseases are SLE, SS, antiphospholipid syndrome or sarcoidosis (class IV) [20,106–112]. Autoantibodies against nuclear antigens can be detected in NMO without clinical evidence of systemic autoimmune diseases. Some authors consider this as an epiphenomenon in the context of disordered humoral autoimmunity (class IV) [2,20,113]. Pittock et al. evaluated the presence of autoantibodies in US American patients with NMO (N = 78) and LETM (N = 75) and most frequently detected antinuclear antibodies (43.8%) followed by antibodies for SS (15.7%) (class IV) [20]. Both antibodies were found more frequently in NMO-IgG positive than in negative patients (P = 0.001). No patient was positive for NMO-IgG among the 49 control patients with SS/SLE; none of these patients had a presentation involving the optic nerve or the spinal cord. The frequency for detection of autoantibodies was even higher in paediatric NMO, where autoantibodies were found in 57/75 (76%) (class IV) [68]. In these patients, a coexisting autoimmune disease was diagnosed in 16 of 58 patients (28%). Occurrence of ON and LETM in these syndromes can either be considered as vasculitic neurological complication or the currently more favoured coexistence of two autoimmune disorders (class IV) [20]. In addition, in a cohort of patients with NMO (N = 78), 3% met the international criteria for SLE or SS (class IV) [20]. Quoting that it was a conservative decision, the NMSS task force concluded that, pending further studies, clinical evidence of SLE or SS should exclude a diagnosis of NMO [99]. Conversely, that panel commented that in the lack of clinical evidence for SLE or SS, seropositivity for antinuclear antibodies or SSA/SSB would not exclude the diagnosis of NMO.

Atypical cases  Patients with clinically manifest or subclinical brain MRI lesions (hypothalamic, periventricular, brainstem) are referred to as atypical cases when other clinical characteristics are typical of NMO and when they are seropositive for NMO-IgG/AQP4 antibodies [114]. Brain lesions in some NMO spectrum disorder patients are accompanied by vasogenic oedema and may manifest as PRES (class IV) [53].

Asian opticospinal MS (OSMS)  Asian opticospinal MS shares some clinical, immunological and MRI features with recurrent NMO, and it has been suggested that OSMS may be the same entity as NMO (class IV) [31]. However, differences in nomenclature in Asia and Western countries hamper comparative analyses. Whilst the diagnosis of NMO requires ON and LETM in Western countries, in Asia – regardless of the length of their spinal cord lesion – associations of ON and myelitis are classified as OSMS [99]. NMO-IgG/AQP4 antibody seropositivity has been reported in OSMS in about 60% of the patients (class IV) [17,31]. NMO-IgG/AQP4 antibody negative OSMS is associated with significantly fewer brain lesions, and some authors suggest that the immunopathogenesis of OSMS is more heterogeneous (class IV) [22].

Differential diagnosis

Despite the availability of diagnostic criteria, there is an overlap between NMO and other disorders, and MS remains the most important differential diagnosis. In a cohort of 320 patients with clinically isolated syndrome suggestive of MS, 23 patients (7.2%) fulfilled the revised absolute NMO criteria of 2006 at some time (class III) [115]. In paediatric NMO particularly, ADEM has to be considered (class IV) [68]. Other differential diagnoses include spinal cord and/or optic nerve manifestations of viral, bacterial and fungal infections. Toxic exposures, nutritional and metabolic disorders, ischaemia, neoplasia and neurodegenerative diseases may mimic inflammation within the spinal cord and optic nerve. Also, hereditary optic neuropathies and retinal disorders have to be considered. Furthermore, spinal cord compression, arteriovenous malformations and non-cord mimics such as Guillain-Barré syndrome and MG have to be taken into account.

Conclusions:  NMO and NMO spectrum disorders need to be distinguished from MS due to divergent course, treatment strategies and outcome. ‘High-risk’ syndromes among NMO spectrum disorders include NMO-IgG/AQP4 antibody positive LETM and RION/BON. Overlap with organ and non-organ-specific autoimmune disorders and a broad range of other differential diagnoses have to be taken into account.

Best practice recommendations:  Diagnostic criteria are the cornerstones of NMO diagnosis, and inception cohort studies in patients with first episodes are needed to refine the diagnostic criteria. A diagnostic guideline for spatially limited NMO spectrum disorders was assembled by the European task force (Figure 1).


Treatment of acute exacerbations

Steroids: Initial or recurrent episodes are usually treated with high-dose intravenous methylprednisolone (1 g daily for three to five consecutive days). This recommendation is taken from studies of MS and idiopathic ON because there are no controlled therapeutic trials which investigated the effectiveness of steroids specifically in NMO. Acute NMO symptoms respond to short courses of high-dose intravenous corticosteroids in up to 80% of patients within 1–5 days, and treatment is generally well tolerated (class IV) [8]. In many countries of the EU, the intravenous therapy with methylprednisolone for MS-related relapses is followed by an oral taper, which lacks controlled trials and needs to be performed slowly.

Plasma exchange: Therapeutic plasmapheresis was effective in patients with severe symptoms that fail to improve or progress despite treatment with corticosteroids. Plasma exchange (1–1.5 plasma volume per exchange) for the treatment of steroid-unresponsive severe mostly MS-related attacks was compared in a double-blinded, sham-controlled protocol (total of seven treatments every other day) in a North American study (class II only for MS, rating not possible for NMO because only two patients with NMO included) [116]. Patients who did not achieve moderate or greater improvement after the first treatment phase crossed over to the opposite treatment. Improvement was found in 8 of 19 (42.1%) patients receiving plasma exchange and only in 1 of 17 (5.9%) receiving sham exchange. In an extension of this study, moderate or marked improvement was detected in 6 of 10 patients after switching to plasma exchange (class III) [117]. However, in three patients, no improvement could be accomplished. Predictive factors for better prognosis were male gender, preserved reflexes and early initiation of plasma exchange. Llufriu et al. performed plasma exchange in four NMO patients with severe episodes of CNS demyelination unresponsive to steroids; one patient had improved at discharge (25%) and the remainder when reassessed after 6 months (75%) (class IV) [118]. Watanabe et al. reported the therapeutic efficacy of plasma exchange (median four exchanges over a period of 1–2 weeks) in 6 NMO-IgG seropositive patients who were unresponsive to high-dose intravenous methylprednisolone treatment (class IV) [119]. Following plasma exchange, three patients experienced a significant functional improvement, whilst one had mild and two no improvement. Clinical response commenced quickly, after one or two exchanges. Benefits were also seen in patients with severe isolated optic neuritis who were unresponsive to high-dose corticosteroids (class IV) [120]. Efficacy of plasma exchange for severe spinal cord attacks in patients with NMO spectrum disorders was shown to be independent of NMO-IgG seropositivity (class IV) [121]. Miyamoto et al. reported four AQP4 antibody positive NMO patients with steroid-refractory relapses in which plasmapheresis was effective (class IV) [122]. Lymphocyte apheresis, a procedure that removes only lymphocytes but not plasma from the blood, was effectively applied in a single NMO-IgG negative NMO patient who did not respond to methylprednisolone or IVIG (class IV) [123].

Intravenous Immunoglobulins (IVIG): IVIG has not been specifically evaluated for ON/LETM relapses of NMO/NMO spectrum disorders and is rarely used for corticosteroid-refractory attacks [124].

Conclusions:  Disability in NMO is attack-related, and relapses require a rapid treatment approach with high-dose i.v. corticosteroids, followed by a slow tailoring; however, they may have limited or no effect in a subgroup of patients. A repeated course of corticosteroids may be considered prior to further treatment escalation. Corticosteroid tapering is standard for treatment of MS relapses in many countries of the European Union. In case of unresponsiveness to steroids, early initiation of a rescue therapy with plasmapheresis is indicated (up to seven treatments every other day). There is currently no evidence for the effectiveness of other pharmacological approaches in the treatment of NMO relapses. Whether NMO spectrum disorders associated with other autoimmune disorders require a different approach for treatment of acute relapses is yet to be determined.

Best practice recommendations:  We suggest the following approach for treatment of relapses: High-dose steroids (methylprednisolone; 1 g for 3–5 consecutive days) as first-line therapy, followed by an oral prednisolone taper. Early initiation of an escalation therapy with plasma exchange is recommended in steroid-unresponsive relapses. Prior to escalation therapy, a repeated course of high-dose corticosteroids may be considered.

Attack prevention (alphabetical order)

Immunomodulatory treatment  Many patients with recurrent NMO receive the initial diagnosis of MS and are treated with approved immunomodulatory therapies such as interferon-β and glatiramer acetate. A randomized-controlled trial for interferon-β 1b in Japan revealed effectiveness in relapsing-remitting MS, but was not powered for evaluation of treatment effects in opticospinal MS (class I for MS) [125]. However, exacerbations of severe ON and myelitis were reported from a Japanese OSMS trial with interferon-β 1b and treatment was discontinued in five of six patients (class IV) [126]. A retrospective Japanese study of 104 consecutive patients revealed that interferon-β 1b significantly reduced the relapse rate in the first year in MS (N = 69; P < 0.00001), whereas the annualized relapse rate was not altered in patients with NMO (N = 35, P = 0.56) (class III) [127]. The development of extensive brain lesions after treatment with interferon-β was observed in two Japanese patients (class IV) [128]. In a French study, 26 patients with recurrent NMO receiving either immunosuppressive (cyclophosphamide, mitoxantrone, or azathioprine) or immunomodulatory treatment (interferon-β) were followed for a mean of 32.0 months (class III) [129]. The retrospective analysis revealed that the probability of further relapses was significantly lower in patients receiving immunosuppressive treatment (P = 0.0007). A 48-year-old patient who was unresponsive to cyclophosphamide, but was effectively treated with glatiramer-acetate was reported (class IV) [130].

Intravenous Immunoglobulins (IVIG): IVIG might be effective in NMO given the potential humoral immunopathogenesis. However, there are only very few supportive data in the literature, including the case of two recurrent patients with NMO who responded to IVIG whilst prior azathioprine and prednisone therapies were not effective (class IV) [131]. An open-label study including five patients with NMO and three patients with recurrent LETM treated with bi-monthly IVIG (0.7 g per kg bodyweight per day for three consecutive days) reported an effect on the annualized relapse rate (from 1.6 in the previous year to 0.0006 in the follow-up, P = 0.01) and the expanded disability status scale (EDSS) (from 3.3 to 2.6, P = 0.04) (class IV) [132]. A total of 83 infusions during a mean follow-up of 19.3 month (range 4–21) were well tolerated, minor adverse events included headache in three patients and a mild cutaneous rash in one patient.

Immunosuppressive treatment

Azathioprine & steroids: Azathioprine is a purine synthesis inhibitor and interferes with the proliferation of cells, especially leucocytes. Azathioprine (75–100 mg daily) in combination with oral prednisolone (1 mg/kg/daily) was evaluated in a prospective open-label case series of seven patients with newly diagnosed NMO (class IV) [133]. The combination was effective over a treatment period of 19 months by means of sustained improvement in the EDSS scores and absence of relapses. Treatment with azathioprine may be carried out in analogy to the recommendations for myasthenia gravis. Haematological evaluations are required every 2–4 weeks, the dosage may be reduced in the course and treatment duration of up to 5 years may be considered. The evaluation of long-term side effects of azathioprine treatment in MS revealed that gastrointestinal complaints and leucopenia were very common adverse events (>10%), whilst infections, allergy and haematological disturbances were common (1–10%) [134]. Corticosteroids are used for rapid immunosuppression until azathioprine exerts its full effect. Some patients experience clinical worsening when prednisone is reduced below 5–15 mg/day (class IV) [135]. Long-term treatment with corticosteroids requires osteoporosis prophylaxis.

Cyclophosphamide: Cyclophosphamide, an alkylating chemotherapeutic drug related to nitrogen mustards, is a non-specific immunosuppressant that affects both T-cell and B-cell functions. Immunosuppression is transient when cyclophosphamide is given in standard pulse doses, and the immune system returns to baseline within a few months to a year after cessation [136]. Published treatment regimens for i.v. cyclophosphamide vary and range from 7 to 25 mg/kg every month over a period of 6 months. Uromitexan administration with every dose needs to be included for prevention of haemorrhagic cystitis.

A few case reports suggest that cyclophosphamide is partially effective in NMO syndromes associated with other autoimmune disorders including SLE and SS (class IV) [137–140]. In neuropsychiatric SLE, low-dose i.v. cyclophosphamide (200–400 mg per month, N = 37) and oral prednisone were superior to oral prednisone alone (N = 23, average daily dose of prednisone 20.5 mg in both groups) [141]. In MS, cyclophosphamide is considered a treatment option in aggressive courses refractory to immunomodulatory treatment. However, a recent Cochrane systematic review found only four randomized-controlled MS trials with sufficient data and concluded that cyclophosphamide does not prevent worsening of the EDSS [142]. Occurrence of amenorrhea and sepsis has to be taken into account.

Methotrexate: Methotrexate exerts its immunosuppressive activity via inhibition of dihydrofolate reductase and has anti-inflammatory and immunomodulatory effects. Among eight patients with NMO, four treated with a combination of methotrexate (i.v. 50 mg weekly) and prednisolone (1 mg/kg/daily) were stabilized (class IV) [143]. Among the four patients who were treated with cyclophosphamide only, one stabilized and the remaining three responded positively after switching to the combination of methotrexate and prednisolone.

Mitoxantrone: Mitoxantrone is an anthracenedione antineoplastic agent that intercalates with DNA and inhibits both DNA and RNA synthesis, suppressing T-cell and B-cell immunity. A prospective 2-year case series (12 mg/m2 monthly for 6 months, followed by three more treatments, each 3 months apart) reported that four of five patients with recurrent NMO experienced disease stabilization and improvement of MRI measures (class IV) [144]. Mitoxantrone is a toxic agent that must be administered with care to reduce the likelihood of bone marrow suppression, opportunistic infection and cardiomyopathy. Amenorrhoea is a major concern in the treatment of young women. Therapy-related acute leukaemia (TRAL) was studied in 5472 patients with MS treated with a mean dose of 74.2 mg/m2 (range: 12–120 mg/m2) [145]. TRAL occurred in 0.3%, median onset was at 18.5 months (range 40–60 months) after start of mitoxantrone treatment and among the 25 patients for which outcome was reported six died (24%). A relationship with total dose is suggested since over 80% of the patients in which acute leukaemia occurred had received >60 mg/m2.

Mycophenolate mofetil: Mycophenolate (p.o. 1–3 g/day) is suggested in several other conditions requiring immunosuppression and is mostly used when a rapid onset of treatment effects is not required and azathioprine is not tolerated. The occurrence of treatment effect is more rapid for mycophenolate than for azathioprine. Jacob et al. reported a retrospective study of 24 patients suffering from NMO (N = 15) and NMO spectrum disorders (relapsing LETM N = 7, RION and LETM N = 1 each) with a median treatment duration of 27 months (class IV) [146]. At a median follow-up of 28 months (range 18–89 months), 19 patients (79%) were continuing treatment and one had died. The median annualized post-treatment relapse rate was lower than the pre-treatment rate (P < 0.001). A 9-year-old girl treated with mycophenolate achieved sustained improvement over a 2-year follow-up (class IV) [147]. Some panel members have very good experience with mycophenolate as long-term treatment of NMO and would also consider this therapy as first-line treatment (expert opinion).

Prednisolone: Low-dose oral prednisolone (5–20 mg daily) was shown to reduce relapse frequency in a retrospective study of nine Japanese patients with NMO (class IV) [148]. However, relapses occurred significantly more frequently when steroids were tapered to 10 mg/day or less (odds ratio 8.75).

Rituximab: Rituximab, a chimeric anti-CD20 monoclonal antibody capable of depleting mature and precursor B cells, was shown to reduce disease activity and prevent disability in two studies. The first study, a retrospective analysis of 25 patients with NMO treated with rituximab as escalation therapy was performed in patients from the US and UK (class IV) [149]. Two rituximab regimens were used: (i) 375 mg/m2 infused once per week for 4 weeks (N = 18) and (ii) 1000 mg infused twice, with a 2-week interval between the infusions (N = 4). The median annualized pre-treatment relapse rate was 1.7 (range 0–3.2) and dropped to 0 relapses (range 0–3.2) at a median follow-up of 19 months. The EDSS improved in 11 patients, did not change in nine, and worsened in five patients, of whom two died. The second study is an open-label study, which evaluated eight patients with NMO who had failed other therapeutic regimens and were treated with rituximab (four infusions 375 mg/m2, once per week) (class IV) [150]. Mean follow-up time was 12 months, and 6 of 8 patients remained relapse-free. B-cell counts were evaluated bi-monthly, and patients were given the option to be retreated with rituximab when B-cell counts became detectable (re-treatment two infusions of 1000 mg, 2 weeks apart).

A 2005 report on the safety of rituximab in patients with cancer and rheumatoid arthritis concluded that overall usage is safe [151]. Infusion-related reactions were reported in 84% and included nausea, headache, fatigue, rash, flu-like symptoms. The incidence of these symptoms is highest after the first infusion and decreases in the course. Thus, the use of acetaminophen, antihistamines and corticosteroids is recommended as pre-treatment. Infections are reported in 30% of rituximab-treated patients but only 1–2% acquire severe infections. Of note, concomitant immunosuppressive therapies enhance the susceptibility to infection.

Other treatment options: There are no studies of fingolimod (FTY-720) or natalizumab in NMO.

Plasmapheresis  The potential therapeutic efficacy of plasmapheresis for relapse prevention is derived from a study by Miyamoto et al. in which two patients with NMO received intermittent plasma exchange in combination with immunosuppressants (class IV) [122]. One patient received prednisolone and cyclophosphamide and additional double filtration plasmapheresis. The other patient received additional azathioprine and cyclophosphamide.

Conclusions:  Long-term treatment options should be initiated as soon as the diagnosis of NMO is made because prevention of attacks is the key issue for reducing permanent disability. Seronegative NMO is treated in the same way as seropositive NMO. However, there are no randomized-controlled trials and currently only class IV evidence for effect of any medication for relapse prevention. Hence, data favouring specific therapies are weak. Immunosuppression is the preferred treatment, but optimal drug regime and treatment duration are yet to be determined; the decision on treatment options has to consider the time until treatment effect is reached and potential long-term side effects. Intermittent plasmapheresis might be considered in case of insufficient relapse prevention with immunosuppressants. Whether NMO-IgG/AQP4 antibody titres may serve as a marker of treatment response remains to be elucidated.

Spatially limited NMO spectrum disorders such as LETM and RION/BON should be considered as inaugural manifestations of NMO and treatment should be started in accordance with the clinical course (Table 2). Currently, it is not clear, whether NMO syndromes associated with other autoimmune disorders have a divergent course and require different or additional treatment strategies. However, in analogy to NMO, there is a lack of large randomized trials for the treatment CNS manifestations of systemic autoimmune disorders [152].

Table 2.   Panel recommendations for immunosuppressive treatment of Neuromyelitis optica
  Drug name Regimen
First-line therapy
  Azathioprine Oral 2.5–3 mg/kg/day
Plus Prednisolone Oral 1 mg/kg/day, tapered when azathioprine becomes effective (after 2–3 months)
  Rituximab Option 1: i.v. 375 mg/m2 weekly for 4 weeks (lymphoma protocol)
Option 2: 1000 mg infused twice, with a 2-week interval between the infusions (rheumatoid arthritis protocol)
Options 1 and 2: re-infusion after 6–12 months; however, optimal treatment duration unknown
Second-line therapy Alphabetical order  
  Cyclophosphamide i.v. 7–25 mg/kg every month over a period of 6 months, especially considered in case of association with SLE/SS
  Mitoxantrone i.v. 12 mg/m2 monthly for 6 months, followed by 12 mg/m2 every 3 months for 9 months
  Mycophenolate mofetil p.o. 1–3 g per day
Other therapies IVIG, Methotrexate  
Escalation therapy
AND Intermittent plasma exchange

Best practice recommendations:  First- and second-line therapy schemes for NMO and spatially limited NMO spectrum disorders are suggested by the task force (Table 2). The categorization in first- and second-line therapies is based on expert preference.

First-line therapy: As first-line therapy, oral treatment with azathioprine (2.5–3 mg/kg daily), in combination with oral prednisolone (1 mg/kg daily or equivalent given every other day) until azathioprine has taken full effect, is recommended. Slow tapering of prednisolone should be considered after 2–3 months. Optimal treatment duration is yet to be determined; considering similarities to MG, azathioprine treatment up to 5 years should be considered. A further first-line therapy is rituximab; however, the optimal surrogate measures, treatment intervals and duration are unclear.

Second-line therapy: If first-line treatment is ineffective or the patient develops steroid-dependence for clinical remission, alternative immunosuppressive therapies need to be considered. We suggest either cyclophosphamide (7–25 mg/kg every month over a period of 6 months), mitoxantrone (12 mg/m2 every 3 months for 9 months) or mycophenolate mofetil (1–3 g/day) as second-line therapy (alphabetical order). Other potentially effective drugs include IVIG and methotrexate. Additional intermittent plasma exchange may be an option for treatment escalation.

Spatially limited NMO spectrum disorders: Whether spatially limited NMO spectrum disorders should be handled as if they are NMO or whether long-term treatment should be deferred until the relationship with NMO is better established by the occurrence of corresponding symptoms (LETM or ON, respectively) remains unclear. The panel has set up a treatment scheme adapted from classical NMO, which considers initiation of immunosuppressive treatment depending on the clinical course (Table 1). Of note, the likelihood for NMO-IgG/AQP4 negative ON to be associated with MS or other disorders is higher than for NMO-IgG/AQP4 negative LETM. Hence, the scheme of Table 1 is expected to be of greater value for AQP4 negative LETM. Patients with recurrent AQP4 negative ON, particularly with mild relapses and good recovery, may not be necessarily part of the NMO spectrum. However, seronegative patients with recurrent severe ON attacks and incomplete recovery might require similar treatment strategies as spatially limited NMO spectrum disorders.

Table 1.   Panel recommendation for starting immunosuppressive treatment in spatially limited NMO spectrum disorders RION/BON and LETM
  NMO-IgG/AQP4 antibody Severity/recovery of acute attacks Immunosuppressive treatment

    *Depending on follow-up and number of episodes.

RION/BON + Severe/poor +
+ Acceptable/good +/−*
Severe/poor +/−*
LETM + Severe/poor +
+ Acceptable/good +/−*
Severe/poor +/−*

Neuromyelitis optica spectrum disorders associated with SLE/SS: Keeping in mind that further studies are required to support the NMSS task force decision about SLE/SS and NMO, the European panel suggests that NMO cases with clinical evidence of SLE or SS should be treated according to ACR and European League against Rheumatism (EULAR) treatment protocols of neurological manifestations of systemic autoimmune disorders and SS [153,154].

Supportive and symptomatic treatment

Supportive and symptomatic treatment is an essential component of the overall management of NMO and aimed at control or reduction in symptoms impairing functional abilities and quality of life. Such symptoms include spasticity, tonic spasms, NMO-related pain syndromes, bladder symptoms, neurogenic bowel dysfunction, sexual dysfunction and cognitive impairment. Some patients with high cervical cord lesions will require long-term mechanical ventilation.

No trials specifically for symptomatic treatment of NMO have been performed, most evidence is derived from studies in MS, and the readers are referred to recommendations for symptomatic treatment of MS, for instance, from the Multiple Sclerosis Therapy Consensus Group (MSTCG) of the German Multiple Sclerosis Society [155].

Conflicts of interest

  1. Top of page
  2. Abstract
  3. Objectives
  4. Introduction
  5. Search strategy and grading of recommendations
  6. Results
  7. Conflicts of interest
  8. References

JS, MB, ZI and CM declare no conflicts of interest. XM Editorial/advisory board and speakers′s fees from Almirall, Bayer Schering, Biogen Idec, Merck Serono, Sanofi-Aventis, Teva. CP Research grants: Bayer Schering, Biogen Idex, GSK, Merck Serono, Novartis, Teva, UCB. Consultant fees: Actelion, Bayer Schering, Biogen Idec, GSK, Merck Serono, Novartis, UCB, Roche, Teva. RDP Speaker′s fees: Biogen Dompé, Biogen Idec, Merck Serono, Sanofi Aventis, GSK. Research Grants: Bayer Schering, Biogen Dompé. Consultant fees: Biogen Dompé, Biogen Idec, Merck Serono. Travel Grants: Bayer Schering, Biogen Dompé, Merck Serono. RH Research grants: Biogen Idex, Merck Serono, Novartis. Consultant fees: PSS Editorial/advisory board fees: Biogen Idec, Merck Serono, Teva. Speaker′s fees: Bayer Schering, Biogen Idec, Merck Serono, Novartis, Teva. Unrestricted educational grant: Biogen Idec, Bayer Schering, Merck Serono, Novartis. Research grant: Biogen Idec, Merck Serono. Biogen Idec, Merck Serono. BH Editorial/advisory board and speaker′s fees from Bayer Schering, Biogen Idec, Merck Serono, Novartis, Teva. Travel grants from Bayer, Biogen Idec, Merck Serono. Research Grants: Bayer, BiogenIdec, MerckSerono, Novartis.


  1. Top of page
  2. Abstract
  3. Objectives
  4. Introduction
  5. Search strategy and grading of recommendations
  6. Results
  7. Conflicts of interest
  8. References
  • 2
    O’Riordan JI, Gallagher HL, Thompson AJ, et al. Clinical, CSF, and MRI findings in Devic’s neuromyelitis optica. J Neurol Neurosurg Psychiatry 1996; 60: 382–387.

  • 3
    El Otmani H, Rafai MA, Moutaouakil F, et al. La neuromyelite optique au Maroc. Etude de neuf cas. Rev Neurol (Paris) 2005; 161: 1191–1196.
  • 4
    Bichuetti DB, Oliveira EM, Souza NA, et al. Neuromyelitis optica in Brazil: a study on clinical and prognostic factors. Mult Scler 2009; 15: 613–619.
  • 5
    Wu JS, Zhang MN, Carroll WM, et al. Characterisation of the spectrum of demyelinating disease in Western Australia. J Neurol Neurosurg Psychiatry 2008; 79: 1022–1026.
  • 6
    Rivera VM, Cabrera JA. Aboriginals with multiple sclerosis: HLA types and predominance of neuromyelitis optica. Neurology 2001; 57: 937–938.
  • 7
    Rivera JF, Kurtzke JF, Booth VJ, et al. Characteristics of Devic’s disease (neuromyelitis optica) in Mexico. J Neurol 2008; 255: 710–715.
  • 8
    Wingerchuk DM, Hogancamp WF, O’Brien PC, et al. The clinical course of neuromyelitis optica (Devic’s syndrome). Neurology 1999; 53: 1107–1114.
  • 9
    Kira J. Multiple sclerosis in the Japanese population. Lancet Neurol 2003; 2: 117–127.
  • 10
    Matsushita T, Matsuoka T, Isobe N, et al. Association of the HLA-DPB1*0501 allele with anti-aquaporin-4 antibody positivity in Japanese patients with idiopathic central nervous system demyelinating disorders. Tissue Antigens 2009; 73: 171–176.
  • 11
    Das A, Puvanendran K. A retrospective review of patients with clinically definite multiple sclerosis. Ann Acad Med Singapore 1998; 27: 204–209.
  • 12
    Papais-Alvarenga RM, Miranda-Santos CM, Puccioni-Sohler M, et al. Optic neuromyelitis syndrome in Brazilian patients. J Neurol Neurosurg Psychiatry 2002; 73: 429–435.
  • 13
    Chopra JS, Radhakrishnan K, Sawhney BB, et al. Multiple sclerosis in North-West India. Acta Neurol Scand 1980; 62: 312–321.
  • 14
    Albutt T. On the opthalmoscopic signs of spinal disease. Lancet 1870; 1: 76–78.
  • 15
    Devic E. Myélite subaiguë compliquée de névrite optique. Bull Med (Paris) 1894; 8: 1033–1034.
  • 16
    Gault F. De la neuromyélite optique aiguë. France: Thése Lyon, 1894.
  • 17
    Lennon VA, Wingerchuk DM, Kryzer TJ, et al. A serum autoantibody marker of neuromyelitis optica: distinction from multiple sclerosis. Lancet 2004; 364: 2106–2112.
  • 18
    Matiello M, Lennon VA, Jacob A, et al. NMO-IgG predicts the outcome of recurrent optic neuritis. Neurology 2008; 70: 2197–2200.
  • 19
    Weinshenker BG, Wingerchuk DM, Vukusic S, et al. Neuromyelitis optica IgG predicts relapse after longitudinally extensive transverse myelitis. Ann Neurol 2006; 59: 566–569.
  • 20
    Pittock SJ, Lennon VA, De Seze J, et al. Neuromyelitis optica and non organ-specific autoimmunity. Arch Neurol 2008; 65: 78–83.
  • 21
    McKeon A, Lennon VA, Jacob A, et al. Coexistence of myasthenia gravis and serological markers of neurological autoimmunity in neuromyelitis optica. Muscle Nerve 2009; 39: 87–90.
  • 22
    Kira J. Neuromyelitis optica and asian phenotype of multiple sclerosis. Ann N Y Acad Sci 2008; 1142: 58–71.
  • 23
    Brainin M, Barnes M, Baron JC, et al. Guidance for the preparation of neurological management guidelines by EFNS scientific task forces – revised recommendations 2004. Eur J Neurol 2004; 11: 577–581.
  • 24
    Yamakawa K, Kuroda H, Fujihara K, et al. Familial neuromyelitis optica (Devic’s syndrome) with late onset in Japan. Neurology 2000; 55: 318–320.
  • 25
    Braley T, Mikol DD. Neuromyelitis optica in a mother and daughter. Arch Neurol 2007; 64: 1189–1192.
  • 26
    Wingerchuk DM, Lennon VA, Lucchinetti CF, et al. The spectrum of neuromyelitis optica. Lancet Neurol 2007; 6: 805–815.
  • 27
    Wu JS, Zhang JM, Carroll WM, et al. Characterisation of the Spectrum of Demyelinating disease in Western Australia. J Neurol Neurosurg Psychiatry 2008; 79: 1022–1026.
  • 28
    Cabre P, Heinzlef O, Merle H, et al. MS and neuromyelitis optica in Martinique (French West Indies). Neurology 2001; 56: 507–514.
  • 29
    Misu T, Fujihara K, Nakashima I, et al. Pure optic-spinal form of multiple sclerosis in Japan. Brain 2002; 125: 2460–2468.
  • 30
    Bizzoco E, Lolli F, Repice AM, et al. Prevalence of neuromyelitis optica spectrum disorder and phenotype distribution. J Neurol 2009; 256: 1891–1898.
  • 31
    Weinshenker BG, Wingerchuk DM, Nakashima I, et al. OSMS is NMO, but not MS: proven clinically and pathologically. Lancet Neurol 2006; 5: 110–111.
  • 32
    Yamasaki K, Horiuchi I, Minohara M, et al. HLA-DPB1*0501-associated opticospinal multiple sclerosis: clinical, neuroimaging and immunogenetic studies. Brain 1999; 122 (Pt 9): 1689–1696.
  • 33
    Cabrera-Gomez JA, Kurtzke JF, Gonzalez-Quevedo A, et al. An epidemiological study of neuromyelitis optica in Cuba. J Neurol 2009; 256: 35–44.
  • 34
    De Seze J, Stojkovic T, Ferriby D, et al. Devic’s neuromyelitis optica: clinical, laboratory, MRI and outcome profile. J Neurol Sci 2002; 197: 57–61.
  • 35
    Banwell B, Tenembaum S, Lennon VA, et al. Neuromyelitis optica-IgG in childhood inflammatory demyelinating CNS disorders. Neurology 2008; 70: 344–352.
  • 36
    Filley CM, Sternberg PE, Norenberg MD. Neuromyelitis optica in the elderly. Arch Neurol 1984; 41: 670–672.
  • 37
    Barbieri F, Buscaino GA. Neuromyelitis optica in the elderly. Acta Neurol (Napoli) 1989; 11: 247–251.
  • 38
    Lotze TE, Northrop JL, Hutton GJ, et al. Spectrum of pediatric neuromyelitis optica. Pediatrics 2008; 122: e1039–e1047.
  • 39
    Confavreux C, Vukusic S. The clinical epidemiology of multiple sclerosis. Neuroimaging Clin N Am 2008; 18: 589–622.
  • 40
    Wingerchuk DM, Weinshenker BG. Neuromyelitis optica: clinical predictors of a relapsing course and survival. Neurology 2003; 60: 848–853.
  • 41
    Wingerchuk DM, Pittock SJ, Lucchinetti CF, et al. A secondary progressive clinical course is uncommon in neuromyelitis optica. Neurology 2007; 68: 603–605.
  • 42
    Merle H, Olindo S, Bonnan M, et al. Natural history of the visual impairment of relapsing neuromyelitis optica. Ophthalmology 2007; 114: 810–815.
  • 43
    Mandler RN, Davis LE, Jeffery DR, et al. Devic’s neuromyelitis optica: a clinicopathological study of 8 patients. Ann Neurol 1993; 34: 162–168.
  • 44
    Fardet L, Genereau T, Mikaeloff Y, et al. Devic’s neuromyelitis optica: study of nine cases. Acta Neurol Scand 2003; 108: 193–200.
  • 45
    De Seze J, Blanc F, Jeanjean L, et al. Optical coherence tomography in neuromyelitis optica. Arch Neurol 2008; 65: 920–923.
  • 46
    Naismith RT, Tutlam NT, Xu J, et al. Optical coherence tomography differs in neuromyelitis optica compared with multiple sclerosis. Neurology 2009; 72: 1077–1082.
  • 47
    Nakamura M, Miyazawa I, Fujihara K, et al. Preferential spinal central gray matter involvement in neuromyelitis optica. An MRI study. J Neurol 2008; 255: 163–170.
  • 48
    Sellner J, Luthi N, Buhler R, et al. Acute partial transverse myelitis: risk factors for conversion to multiple sclerosis. Eur J Neurol 2008; 15: 398–405.
  • 49
    Cordonnier C, De Seze J, Breteau G, et al. Prospective study of patients presenting with acute partial transverse myelopathy. J Neurol 2003; 250: 1447–1452.
  • 50
    Misu T, Fujihara K, Nakashima I, et al. Intractable hiccup and nausea with periaqueductal lesions in neuromyelitis optica. Neurology 2005; 65: 1479–1482.
  • 51
    Pittock SJ, Lennon VA, Krecke K, et al. Brain abnormalities in neuromyelitis optica. Arch Neurol 2006; 63: 390–396.
  • 52
    Wingerchuk DM, Lennon VA, Pittock SJ, et al. Revised diagnostic criteria for neuromyelitis optica. Neurology 2006; 66: 1485–1489.
  • 53
    Magana SM, Matiello M, Pittock SJ, et al. Posterior reversible encephalopathy syndrome in neuromyelitis optica spectrum disorders. Neurology 2009; 72: 712–717.
  • 54
    Poppe AY, Lapierre Y, Melancon D, et al. Neuromyelitis optica with hypothalamic involvement. Mult Scler 2005; 11: 617–621.
  • 55
    Blanc F, Zephir H, Lebrun C, et al. Cognitive functions in neuromyelitis optica. Arch Neurol 2008; 65: 84–88.
  • 56
    Vernant JC, Cabre P, Smadja D, et al. Recurrent optic neuromyelitis with endocrinopathies: a new syndrome. Neurology 1997; 48: 58–64.
  • 57
    Cabre P, González-Quevedo A, Bonnan M, et al. Relapsing neuromyelitis optica: long term history and predictors of death. J Neurol Neurosurg Psychiatry 2009; 80: 1162–1164.
  • 58
    Johnson G, Miller DH, MacManus D, et al. STIR sequences in NMR imaging of the optic nerve. Neuroradiology 1987; 29: 238–245.
  • 59
    Kupersmith MJ, Alban T, Zeiffer B, et al. Contrast-enhanced MRI in acute optic neuritis: relationship to visual performance. Brain 2002; 125: 812–822.
  • 60
    Cabrera-Gomez JA, Quevedo-Sotolongo L, Gonzalez-Quevedo A, et al. Brain magnetic resonance imaging findings in relapsing neuromyelitis optica. Mult Scler 2007; 13: 186–192.
  • 61
    Krampla W, Aboul-Enein F, Jecel J, et al. Spinal cord lesions in patients with neuromyelitis optica: a retrospective long-term MRI follow-up study. Eur Radiol 2009; 19: 2535–2543.
  • 62
    Filippi M, Rocca MA, Moiola L, et al. MRI and magnetization transfer imaging changes in the brain and cervical cord of patients with Devic’s neuromyelitis optica. Neurology 1999; 53: 1705–1710.
  • 63
    Nakashima I, Fujihara K, Miyazawa I, et al. Clinical and MRI features of Japanese patients with multiple sclerosis positive for NMO-IgG. J Neurol Neurosurg Psychiatry 2006; 77: 1073–1075.
  • 64
    Sellner J, Luthi N, Schupbach WM, et al. Diagnostic workup of patients with acute transverse myelitis: spectrum of clinical presentation, neuroimaging and laboratory findings. Spinal Cord 2009; 47: 312–317.
  • 65
    Pittock SJ, Weinshenker BG, Lucchinetti CF, et al. Neuromyelitis optica brain lesions localized at sites of high aquaporin 4 expression. Arch Neurol 2006; 63: 964–968.
  • 66
    Cabrera-Gomez J, Saiz-Hinarejos A, Graus F, et al. Brain magnetic resonance imaging findings in acute relapses of neuromyelitis optica spectrum disorders. Mult Scler 2008; 14: 248–251.
  • 67
    Li Y, Xie P, Lv F, et al. Brain magnetic resonance imaging abnormalities in neuromyelitis optica. Acta Neurol Scand 2008; 118: 218–225.
  • 68
    McKeon A, Lennon VA, Lotze T, et al. CNS aquaporin-4 autoimmunity in children. Neurology 2008; 71: 93–100.
  • 69
    Nakamura M, Misu T, Fujihara K, et al. Occurrence of acute large and edematous callosal lesions in neuromyelitis optica. Mult Scler 2009; 15: 695–700.
  • 70
    Ito S, Mori M, Makino T, et al. “Cloud-like enhancement” is a magnetic resonance imaging abnormality specific to neuromyelitis optica. Ann Neurol 2009; 66: 425–428.
  • 71
    Correale J, Fiol M. Activation of humoral immunity and eosinophils in neuromyelitis optica. Neurology 2004; 63: 2363–2370.
  • 72
    Bichuetti DB, Rivero RL, Oliveira DM, et al. Neuromyelitis optica: brain abnormalities in a Brazilian cohort. Arq Neuropsiquiatr 2008; 66: 1–4.
  • 73
    Milano E, Di Sapio A, Malucchi S, et al. Neuromyelitis optica: importance of cerebrospinal fluid examination during relapse. Neurol Sci 2003; 24: 130–133.
  • 74
    Bergamaschi R, Tonietti S, Franciotta D, et al. Oligoclonal bands in Devic’s neuromyelitis optica and multiple sclerosis: differences in repeated cerebrospinal fluid examinations. Mult Scler 2004; 10: 2–4.
  • 75
    Jarius S, Franciotta D, Bergamaschi R, et al. Polyspecific, antiviral immune response distinguishes multiple sclerosis and neuromyelitis optica. J Neurol Neurosurg Psychiatry 2008; 79: 1134–1136.
  • 76
    Miyazawa I, Nakashima I, Petzold A, et al. High CSF neurofilament heavy chain levels in neuromyelitis optica. Neurology 2007; 68: 865–867.
  • 77
    Misu T, Takano R, Fujihara K, et al. Marked increase in cerebrospinal fluid glial fibrillar acidic protein in neuromyelitis optica: an astrocytic damage marker. J Neurol Neurosurg Psychiatry 2009; 80: 575–577.
  • 78
    Kanzaki M, Mochizuki H, Ogawa G, et al. Clinical features of opticospinal multiple sclerosis with anti-aquaporin 4 antibody. Eur Neurol 2008; 60: 37–42.
  • 79
    Graber DJ, Levy M, Kerr D, et al. Neuromyelitis optica pathogenesis and aquaporin 4. J Neuroinflammation 2008; 5: 22.
  • 80
    Tait MJ, Saadoun S, Bell BA, et al. Water movements in the brain: role of aquaporins. Trends Neurosci 2008; 31: 37–43.
  • 81
    Tani T, Sakimura K, Tsujita M, et al. Identification of binding sites for anti-aquaporin 4 antibodies in patients with neuromyelitis optica. J Neuroimmunol 2009; 211: 110–113.
  • 82
    Kinoshita M, Nakatsuji Y, Moriya M, et al. Astrocytic necrosis is induced by anti-aquaporin-4 antibody-positive serum. Neuroreport 2009; 20: 508–512.
  • 83
    Kinoshita M, Nakatsuji Y, Kimura T, et al. Neuromyelitis optica: passive transfer to rats by human immunoglobulin. Biochem Biophys Res Commun 2009; 386: 623–627.
  • 84
    Bennett JL, Lam C, Reddy Kalluri S, et al. Intrathecal pathogenic anti-aquaporin-4 antibodies in early neuromyelitis optica. Ann Neurol 2009; 66(5): 617–629.
  • 85
    Fazio R, Malosio ML, Lampasona V, et al. Antiacquaporin 4 antibodies detection by different techniques in neuromyelitis optica patients. Mult Scler 2009; 15: 1153–1163.
  • 86
    Waters P, Vincent A. Detection of anti-aquaporin-4 antibodies in neuromyelitis optica: current status of the assays. Int MS J 2008; 15: 99–105.
  • 87
    Marignier R, De Seze J, Vukusic S, et al. NMO-IgG and Devic’s neuromyelitis optica: a French experience. Mult Scler 2008; 14: 440–445.
  • 88
    Takahashi T, Fujihara K, Nakashima I, et al. Anti-aquaporin-4 antibody is involved in the pathogenesis of NMO: a study on antibody titre. Brain 2007; 130: 1235–1243.
  • 89
    Paul F, Jarius S, Aktas O, et al. Antibody to aquaporin 4 in the diagnosis of neuromyelitis optica. PLoS Med 2007; 4: e133.
  • 90
    McKeon A, Fryer JP, Apiwattanakul M, et al. Diagnosis of neuromyelitis spectrum disorders: comparative sensitivities and specificities of immunohistochemical and immunoprecipitation assays. Arch Neurol 2009; 66: 1134–1138.
  • 91
    Saiz A, Zuliani L, Blanco Y, et al. Revised diagnostic criteria for neuromyelitis optica (NMO). Application in a series of suspected patients. J Neurol 2007; 254: 1233–1237.
  • 92
    Nishiyama S, Ito T, Misu T, et al. A case of NMO seropositive for aquaporin-4 antibody more than 10 years before onset. Neurology 2009; 72: 1960–1961.
  • 93
    Jarius S, Aboul-Enein F, Waters P, et al. Antibody to aquaporin-4 in the long-term course of neuromyelitis optica. Brain 2008; 131: 3072–3080.
  • 94
    Hinson SR, McKeon A, Fryer JP, et al. Prediction of neuromyelitis optica attack severity by quantitation of complement-mediated injury to aquaporin-4-expressing cells. Arch Neurol 2009; 66: 1164–1167.
  • 95
    Smith CH, Waubant E, Langer-Gould A. Absence of neuromyelitis optica IgG antibody in an active relapsing-remitting multiple sclerosis population. J Neuroophthalmol 2009; 29: 104–106.
  • 96
    Scott TF, Kassab SL, Pittock SJ. Neuromyelitis optica IgG status in acute partial transverse myelitis. Arch Neurol 2006; 63: 1398–1400.
  • 97
    Ravaglia S, Bastianello S, Franciotta D, et al. NMO-IgG-negative relapsing myelitis. Spinal Cord 2009; 47: 531–537.
  • 98
    Klawiter EC, Alvarez E III, Xu J, et al. NMO-IgG detected in CSF in seronegative neuromyelitis optica. Neurology 2009; 72: 1101–1103.
  • 99
    Miller DH, Weinshenker BG, Filippi M, et al. Differential diagnosis of suspected multiple sclerosis: a consensus approach. Mult Scler 2008; 14: 1157–1174.
  • 100
    Pittock SJ. Neuromyelitis optica: a new perspective. Semin Neurol 2008; 28: 95–104.
  • 101
    Furukawa Y, Yoshikawa H, Yachie A, et al. Neuromyelitis optica associated with myasthenia gravis: characteristic phenotype in Japanese population. Eur J Neurol 2006; 13: 655–658.
  • 102
    Gold R, Linington C. Devic’s disease: bridging the gap between laboratory and clinic. Brain 2002; 125: 1425–1427.
  • 103
    Hui AC, Wong RS, Ma R, et al. Recurrent optic neuromyelitis with multiple endocrinopathies and autoimmune disorders. J Neurol 2002; 249: 784–785.
  • 104
    Petravic D, Habek M, Supe S, et al. Recurrent optic neuromyelitis with endocrinopathies: a new syndrome or just a coincidence? Mult Scler 2006; 12: 670–673.
  • 105
    Kira J, Kawano Y. Recurrent optic neuromyelitis with endocrinopathies. Neurology 1997; 49: 1475–1476.
  • 106
    Komolafe MA, Komolafe EO, Sunmonu TA, et al. New onset neuromyelitis optica in a young Nigerian woman with possible antiphospholipid syndrome: a case report. J Med Case Reports 2008; 2: 348.
  • 107
    Birnbaum J, Kerr D. Devic’s syndrome in a woman with systemic lupus erythematosus: diagnostic and therapeutic implications of testing for the neuromyelitis optica IgG autoantibody. Arthritis Rheum 2007; 57: 347–351.
  • 108
    Ferreira S, Marques P, Carneiro E, et al. Devic’s syndrome in systemic lupus erythematosus and probable antiphospholipid syndrome. Rheumatology (Oxford) 2005; 44: 693–695.
  • 109
    Chan AY, Liu DT. Devic’s syndrome in systemic lupus erythematosus and probable antiphospholipid syndrome. Rheumatology (Oxford) 2006; 45: 120–121; author reply 121.
  • 110
    Mehta LR, Samuelsson MK, Kleiner AK, et al. Neuromyelitis optica spectrum disorder in a patient with systemic lupus erythematosus and anti-phospholipid antibody syndrome. Mult Scler 2008; 14: 425–427.
  • 111
    Mochizuki A, Hayashi A, Hisahara S, et al. Steroid-responsive Devic’s variant in Sjogren’s syndrome. Neurology 2000; 54: 1391–1392.
  • 112
    Lehnhardt FG, Impekoven P, Rubbert A, et al. Recurrent longitudinal myelitis as primary manifestation of SLE. Neurology 2004; 63: 1976.
  • 113
    Jacob A, Boggild M. Neuromyelitis optica. Pract Neurol 2006; 6: 180–184.
  • 114
    Jacob A, Matiello M, Wingerchuk DM, et al. Neuromyelitis optica: changing concepts. J Neuroimmunol 2007; 187: 126–138.
  • 115
    Rubiera M, Rio J, Tintore M, et al. Neuromyelitis optica diagnosis in clinically isolated syndromes suggestive of multiple sclerosis. Neurology 2006; 66: 1568–1570.
  • 116
    Weinshenker BG, O’Brien PC, Petterson TM, et al. A randomized trial of plasma exchange in acute central nervous system inflammatory demyelinating disease. Ann Neurol 1999; 46: 878–886.
  • 117
    Keegan M, Pineda AA, McClelland RL, et al. Plasma exchange for severe attacks of CNS demyelination: predictors of response. Neurology 2002; 58: 143–146.
  • 118
    Llufriu S, Castillo J, Blanco Y, et al. Plasma exchange for acute attacks of CNS demyelination: predictors of improvement at 6 months. Neurology 2009; 73: 949–953.
  • 119
    Watanabe S, Nakashima I, Misu T, et al. Therapeutic efficacy of plasma exchange in NMO-IgG-positive patients with neuromyelitis optica. Mult Scler 2007; 13: 128–132.
  • 120
    Ruprecht K, Klinker E, Dintelmann T, et al. Plasma exchange for severe optic neuritis: treatment of 10 patients. Neurology 2004; 63: 1081–1083.
  • 121
    Bonnan M, Valentino R, Olindo S, et al. Plasma exchange in severe spinal attacks associated with neuromyelitis optica spectrum disorder. Mult Scler 2009; 15: 487–492.
  • 122
    Miyamoto K, Kusunoki S. Intermittent plasmapheresis prevents recurrence in neuromyelitis optica. Ther Apher Dial 2009; 13: 505–508.
  • 123
    Nozaki I, Hamaguchi T, Komai K, et al. Fulminant Devic disease successfully treated by lymphocytapheresis. J Neurol Neurosurg Psychiatry 2006; 77: 1094–1095.
  • 124
    Wingerchuk DM, Weinshenker BG. Neuromyelitis optica. Curr Treat Options Neurol 2008; 10: 55–66.
  • 125
    Saida T, Tashiro K, Itoyama Y, et al. Interferon beta-1b is effective in Japanese RRMS patients: a randomized, multicenter study. Neurology 2005; 64: 621–630.
  • 126
    Warabi Y, Matsumoto Y, Hayashi H. Interferon beta-1b exacerbates multiple sclerosis with severe optic nerve and spinal cord demyelination. J Neurol Sci 2007; 252: 57–61.
  • 127
    Tanaka M, Tanaka K, Komori M. Interferon-beta(1b) treatment in neuromyelitis optica. Eur Neurol 2009; 62: 167–170.
  • 128
    Shimizu Y, Yokoyama K, Misu T, et al. Development of extensive brain lesions following interferon beta therapy in relapsing neuromyelitis optica and longitudinally extensive myelitis. J Neurol 2008; 255: 305–307.
  • 129
    Papeix C, Vidal JS, De Seze J, et al. Immunosuppressive therapy is more effective than interferon in neuromyelitis optica. Mult Scler 2007; 13: 256–259.
  • 130
    Gartzen K, Limmroth V, Putzki N. Relapsing neuromyelitis optica responsive to glatiramer acetate treatment. Eur J Neurol 2007; 14: e12–e13.
  • 131
    Bakker J, Metz L. Devic’s neuromyelitis optica treated with intravenous gamma globulin (IVIG). Can J Neurol Sci 2004; 31: 265–267.
  • 132
    Magraner MJ, Bosca I, Simó-Castelló M, et al. An open label study of the effects of intravenous immunoglobulin in neuromyelitis optica spectrum disorders. J Neurol 2010; in press.
  • 133
    Mandler RN, Ahmed W, Dencoff JE. Devic’s neuromyelitis optica: a prospective study of seven patients treated with prednisone and azathioprine. Neurology 1998; 51: 1219–1220.
  • 134
    La Mantia L, Mascoli N, Milanese C. Azathioprine. Safety profile in multiple sclerosis patients. Neurol Sci 2007; 28: 299–303.
  • 135
    Wingerchuk DM, Weinshenker BG. Neuromyelitis optica. Curr Treat Options Neurol 2005; 7: 173–182.
  • 136
    Killian JM, Bressler RB, Armstrong RM, et al. Controlled pilot trial of monthly intravenous cyclophosphamide in multiple sclerosis. Arch Neurol 1988; 45: 27–30.
  • 137
    Bonnet F, Mercie P, Morlat P, et al. Devic’s neuromyelitis optica during pregnancy in a patient with systemic lupus erythematosus. Lupus 1999; 8: 244–247.
  • 138
    Arabshahi B, Pollock AN, Sherry DD, et al. Devic disease in a child with primary Sjogren syndrome. J Child Neurol 2006; 21: 285–286.
  • 139
    Birnbaum J, Kerr D. Optic neuritis and recurrent myelitis in a woman with systemic lupus erythematosus. Nat Clin Pract Rheumatol 2008; 4: 381–386.
  • 140
    Mok CC, To CH, Mak A, et al. Immunoablative cyclophosphamide for refractory lupus-related neuromyelitis optica. J Rheumatol 2008; 35: 172–174.
  • 141
    Stojanovich L, Stojanovich R, Kostich V, et al. Neuropsychiatric lupus favourable response to low dose i.v. cyclophosphamide and prednisolone (pilot study). Lupus 2003; 12: 3–7.
  • 142
    La Mantia L, Milanese C, Mascoli N, et al. Cyclophosphamide for multiple sclerosis. Cochrane Database Syst Rev 2007; 1: CD002819.
  • 143
    Minagar A, Sheremata WA. Treatment of Devic′s disease with methotrexate and prednisone. Int J MS Care 2000; 2: 39–43.
  • 144
    Weinstock-Guttman B, Ramanathan M, Lincoff N, et al. Study of mitoxantrone for the treatment of recurrent neuromyelitis optica (Devic disease). Arch Neurol 2006; 63: 957–963.
  • 145
    Ellis R, Boggild M. Therapy-related acute leukaemia with Mitoxantrone: what is the risk and can we minimise it? Mult Scler 2009; 15: 505–508.
  • 146
    Jacob A, Matiello M, Weinshenker BG, et al. Treatment of neuromyelitis optica with mycophenolate mofetil: retrospective analysis of 24 patients. Arch Neurol 2009; 66: 1128–1133.
  • 147
    Falcini F, Trapani S, Ricci L, et al. Sustained improvement of a girl affected with Devic’s disease over 2 years of mycophenolate mofetil treatment. Rheumatology (Oxford) 2006; 45: 913–915.
  • 148
    Watanabe S, Misu T, Miyazawa I, et al. Low-dose corticosteroids reduce relapses in neuromyelitis optica: a retrospective analysis. Mult Scler 2007; 13: 968–974.
  • 149
    Jacob A, Weinshenker BG, Violich I, et al. Treatment of neuromyelitis optica with rituximab: retrospective analysis of 25 patients. Arch Neurol 2008; 65: 1443–1448.
  • 150
    Cree BA, Lamb S, Morgan K, et al. An open label study of the effects of rituximab in neuromyelitis optica. Neurology 2005; 64: 1270–1272.
  • 151
    Kimby E. Tolerability and safety of rituximab (MabThera). Cancer Treat Rev 2005; 31: 456–473.
  • 152
    Joseph FG, Scolding NJ. Neurolupus. Pract Neurol 2010; 10: 4–15.
  • 153
    Bertsias G, Ioannidis JP, Boletis J, et al. EULAR recommendations for the management of systemic lupus erythematosus. Report of a Task Force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics. Ann Rheum Dis 2008; 67: 195–205.
  • 154
    Guidelines for referral and management of systemic lupus erythematosus in adults. American College of Rheumatology Ad Hoc Committee on Systemic Lupus Erythematosus Guidelines. Arthritis Rheum 1999; 42: 1785–1796.
  • 155
    Henze T, Rieckmann P, Toyka KV. Symptomatic treatment of multiple sclerosis. Multiple Sclerosis Therapy Consensus Group (MSTCG) of the German Multiple Sclerosis Society. Eur Neurol 2006; 56: 78–105.