The impact of 2015 neuromyelitis optica spectrum disorders criteria on diagnostic rates

Introduction

Patients with clinical features consistent with what we now call neuromyelitis optica (NMO) were described in the literature approximately two centuries ago, in case reports. ¹ But it was given a disease status ‘de la neuromyélite optique aiguë’ in 1870 by Gault and Devic. ² Since then, at least eight iterations of diagnostic criteria have been proposed. ³ The discovery of antibodies to aquaporin 4 (AQP4 IgG) in 70%–90% cases of NMO has changed the diagnostic and treatment approach to disorders presenting with presumed demyelinating etiology. ⁴ This was reflected in the previous diagnostic criteria (Table 1). ⁵ However, further advances in the field and the segregation of several clinical syndromes, symptoms, signs or imaging features, ⁶ hitherto not associated with the classical phenotype of NMO (i.e. optic neuritis with longitudinally extensive transverse myelitis), have been reported in patients who have AQP4 IgG although no formal criteria were proposed.^7,8 There also remains a substantial group of patients with otherwise typical clinical or imaging features of NMO that do not have AQP4 IgG. Some may have other serum markers, for example, myelin oligodendrocyte glycoprotein (MOG) IgG. ⁹ The recent diagnostic criteria developed by an International Panel attempts to encompass and incorporate these developments ¹⁰ (Table 2). The term ‘neuromyelitis optica spectrum disorders’ (NMOSD) was felt to better represent the disease and has replaced the term NMO. It is as yet uncertain how much these new criteria will increase the number of cases diagnosed. Therefore, we systematically applied both the new and old criteria to a large cohort of cases.

OPEN IN VIEWER

Table 1.

2006 NMO diagnostic criteria from Wingerchuk et al. ⁵

2006 NMO diagnostic criteria
Optic neuritis and transverse myelitis + two of the following
1. Normal MRI brain
2. Transverse myelitis with MRI changes >3 vertebral segment
3. AQP4 IgG positive

NMO: neuromyelitis optica; MRI: magnetic resonance imaging; AQP4: aquaporin 4; IgG: immunoglobulin G.

OPEN IN VIEWER

Table 2.

The 2015 International Panel for NMO Diagnosis (IPND) criteria for neuromyelitis optica spectrum disorders.

Diagnostic criteria for NMOSD with AQP4 IgG

1. At least one core clinical characteristic

2. Positive test for AQP4 IgG using best available detection method (cell-based assay strongly recommended)

3. Exclusion of alternative diagnoses

Diagnostic criteria for NMOSD without AQP4 IgG or NMOSD with unknown AQP4 IgG status

1. At least two core clinical characteristics occurring as a result of one or more clinical attacks and meeting all of the following requirements:

a. At least one core clinical characteristic must be optic neuritis, acute myelitis with LETM, or area postrema syndrome

b. Dissemination in space (two or more different core clinical characteristics)

c. Fulfilment of additional MRI requirements, as applicable

2. Negative tests for AQP4 IgG using best available detection method or testing unavailable

3. Exclusion of alternative diagnoses

Core clinical characteristics

1. Optic neuritis

2. Acute myelitis

3. Area postrema syndrome: episode of otherwise unexplained hiccups or nausea and vomiting

4. Acute brainstem syndrome

5. Symptomatic narcolepsy or acute diencephalic clinical syndrome with NMOSD-typical diencephalic MRI lesions

6. Symptomatic cerebral syndrome with NMOSD-typical brain lesions

Additional MRI requirements for NMOSD without AQP4 IgG and NMOSD with unknown AQP4 IgG status

1. Acute optic neuritis: requires brain MRI showing (a) normal findings or only nonspecific white matter lesions OR (b) optic nerve MRI with T2-hyperintense lesion or T1-weighted gadolinium-enhancing lesion extending over 1/2 optic nerve length or involving optic chiasm

2. Acute myelitis: requires associated intramedullary MRI lesion extending over three contiguous segments (LETM) OR three contiguous segments of focal spinal cord atrophy in patients with history compatible with acute myelitis

3. Area postrema syndrome: requires associated dorsal medulla/area postrema lesions

4. Acute brainstem syndrome: requires associated periependymal brainstem lesions

Source: Adapted from Wingerchuk et al. ⁵

LETM: longitudinally extensive transverse myelitis; NMO: neuromyelitis optica; NMOSD: neuromyelitis optica spectrum disorders; AQP4: aquaporin 4; IgG: immunoglobulin G; MRI: magnetic resonance imaging.

Patients and method

The aim of this study was to estimate the change in number of patients diagnosed with the new criteria. The study was done in the NMO and non-multiple sclerosis (MS) central nervous system (CNS) demyelination clinic, part of the UK NMO service at the Walton Centre for Neurology and Neurosurgery NHS foundation Trust. This specially funded clinic in a tertiary hospital in the north of England is a referral centre for cases suspected to be NMO or its variants. After an initial detailed review, prospective follow up is done with imaging and additional investigations including new antibody tests as indicated based on new information from the rapidly evolving field. All patients were tested for AQP4 IgG performed at Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital and Oxford using the best available cell-based assay. ¹¹ This study formed part of the UK NMO study (MREC 02/8/082, Northwest medical research ethics committee, United Kingdom).

We identified all patients seen in the clinic between January 2013 and May 2015 (n = 198). We excluded patients where an alternative diagnosis like MS (n = 15) or other well-described CNS inflammatory disorder (n = 7), for example, Behcet’s disease or sarcoidosis, was made in the clinic on reinvestigation. We systematically applied 2006 and 2015 criteria to the remaining 176 patients.

Results

The 2006 criteria classified 63 of the 176 patients as NMO (42 AQP4 IgG positive and 21 AQP4 IgG negative, 46 females and 17 males, ratio 2.7:1). All these patients had both optic neuritis and longitudinally extensive myelitis, either simultaneously (n = 16) or in subsequent attacks (n = 47). Five had single events (8%) and 58 had a relapsing course (92%). The remaining patients (n = 113) did not satisfy 2006 criteria for NMO (Figure 1).

OPEN IN VIEWER

Figure 1.

Flowchart classifying 176 patients as per 2006 criteria. OSD: optic neuritis with short segment demyelination and normal MRI brain; NMO: neuromyelitis optica; AQP4: aquaporin 4; IgG: immunoglobulin G, ON: optic neuritis; LETM: longitudinally extensive transverse myelitis; TM: transverse myelitis; BS: brainstem demyelination. Combination of clinical events: combination of events other than long myelitis and optic neuritis, for example, optic neuritis and cerebral/brainstem syndrome, LETM and cerebral/brainstem syndrome, DS: diencephalic syndrome.

We then applied the 2015 criteria to the same cohort of 176 patients. A total of 111 patients fulfilled the new criteria (82 females, 29 males, ratio 2.8:1, ratio in relapsing cases 3.3:1 and in AQP4 IgG +ve patients is 5:1). In all, 81 were AQP4 IgG positive and 30 AQP4 IgG negative, an increase of 48 patients (76% rise). All the 2006 cases of NMO (n = 63) remained as NMOSD. The AQP4 IgG +ve NMOSD group included the 42 patients with both ON and TM (as in 2006) and a further 39 newly included cases (Figure 2). The AQP4 IgG negative group included all the seronegative NMO in 2006 (n = 21) and nine more newly included cases. The remaining 65 patients with a variety of clinical presentations (all are AQP4 IgG −ve) did not satisfy the new criteria (Figure 2). All patients who were AQP4 negative were tested for MOG antibodies, 15 of AQP4 IgG –ve NMOSD were positive for MOG IgG (50%).

OPEN IN VIEWER

Figure 2.

Flowchart classifying 176 patients as per 2015 criteria. AQP4 IgG: aquaporin 4 immunoglobulin G; NMOSD: neuromyelitis optica spectrum disorder; ON: optic neuritis; LETM: longitudinally extensive transverse myelitis; OSD: optic neuritis with short segment demyelination and normal MRI brain; TM: transverse myelitis; BS: brainstem demyelination; DS: diencephalic syndrome. Combination of clinical events: combination of events other than long myelitis and optic neuritis, for example, optic neuritis and cerebral/brainstem syndrome, LETM and cerebral/brainstem syndrome.

Discussion

The application of the 2015 International Panel for NMO Diagnosis (IPND) criteria in a large cohort of non-MS demyelination demonstrates a rise in diagnosis of NMOSD by 76%. The AQP4 IgG +ve group contributed 62%, and AQP4 IgG −ve group contributed 14%. All patients diagnosed as NMO by the previous criteria in 2006 are still diagnosable as NMOSD. All cases who met the new criteria and did not meet 2006 criteria were previously classified and followed up in the clinic as ‘atypical non-MS demyelination-probable NMO’.

This apparent rise may not be evident in specialist clinical settings, where neurologists have been treating all patients with AQP4 IgG as NMO for the past few years based on emerging evidences despite the absence of formal criteria till now. However, in settings where NMO is still diagnosed with the 2006 criteria, there could be a substantial rise in diagnostic rates.

There are many patients who still remain unclassifiable (Figure 2) as they do not satisfy even the present criteria, but are in the authors’ practice being treated as NMOSD. These are all AQP4 IgG −ve patients, with more than one clinical event with dissemination in space but don’t appear typical for MS. Some patients also do not meet the criteria despite typical clinical events for one of the core criteria (optic neuritis and brainstem symptoms) as they did not have magnetic resonance imaging (MRI) abnormalities specified in the new criteria. Some have clinical syndromes other than that prescribed in the criteria but with typical MRI changes, for example, patients who had typical periependymal changes in area postrema without history of nausea, vomiting or hiccups. A further category is patients who may have recurrent long myelitis without history/signs of ON, but abnormal visual evoked responses. Relapsing forms of site-restricted demyelination like ON, TM and brainstem syndromes too remain without a specific nosological category. These and similar syndromes need to be studied prospectively to see if they will evolve into typical NMOSD and may merit inclusion in future revisions of the criteria. Recent collaborative work showed how the diagnosis of this group can be controversial even among specialists. ¹²

The emergence of MOG antibodies associated with AQP4 IgG −ve cases has added another layer of complexity to the NMOSD field. As in this series, MOG IgG can be positive in a proportion of AQP4 IgG −ve NMOSD patients. But not all MOG IgG +ve cases fulfil the NMOSD criteria.^13,14 MOG Ig patients may not have the same severity or clinical prognosis of AQP4 IgG cases. Indeed, many are monophasic. The test is not yet available widely. The non-inclusion of MOG in the present IPND criteria seems appropriate until we learn more about MOG-associated disease.

As most cases of NMOSD are relapsing, an accurate early diagnosis and initiation of immunosuppressive treatment should lead to reduction in relapse rates, disability and better long-term outcomes. An earlier diagnosis also alerts the physician to treat a relapse aggressively with longer duration of steroids and or plasma exchange. ¹⁵ So if the underlying premise of the 2015 IPND criteria is correct, then this increase in the number of patients diagnosed as NMOSD will lead to a larger number of patients being treated earlier ¹⁶ and better and would avoid wrong treatments that could be deleterious, for example, some MS drugs can worsen NMO. ¹⁷ The rise in numbers may also create interest from the pharmaceutical industry and facilitate recruitment into clinical trials.

We are aware of only one other study assessing the impact of the new criteria. ¹⁶ Hyun et al. from South Korea applied the 2006 and 2015 criteria to 594 patients with CNS inflammation (including MS) and compared diagnostic rates. NMOSD was diagnosed in 136 patients with the 2006 criteria (23%) and 252 (42%) with the 2015 criteria, an increase of 85%, quite similar to our study. They estimated that the time to diagnosis reduces to 11 months by 2015 criteria from 53 months by 2006 criteria. There are limitations to our study. The application of new criteria to data (history, MRI) collected in the past poses some problems. Older MRIs may not be available for review. The nature, duration and severity of symptoms that were thought of as nonspecific and poorly documented or remembered (e.g. nausea, vomiting or excessive sleepiness/narcolepsy) may now acquire significance causing recall bias. This is perhaps reflected in the paucity of cases with confirmed area postrema syndrome in our cohort. We also acknowledge that our specialised clinical setting may not reflect the typical settings (MS or general neurology clinics) where such patients are followed up globally. We also have ready access to a sensitive and specific AQP4 IgG assay, which may not be the case elsewhere. Notwithstanding these limitations, we believe that our findings should stimulate clinicians to reassess their patients with non-MS or atypical MS to see if they fit the 2015 IPND NMOSD criteria.

The next revision of the criteria should address the above issues. It is anticipated that both validated cell-based AQP4 and MOG IgG tests will become available globally by that time facilitating our understanding of the differences between these groups. Importantly, the IPND should suggest unifying terminology to the as yet unclassifiable cohort of patients to further research.

Conclusion

The 2015 IPND criteria is a significant change in the approach to the diagnosis and classification of demyelinating syndromes that are not typical MS. It increases the diagnosis of NMOSD by 76%. It also makes NMOSD a differential diagnosis for many previously unclassifiable CNS disorders. It may allow earlier diagnosis and management, which has significant implications to clinical practice, and research including clinical trials and health care costs. Prospective validation of the criteria in a large multi-ethnic, multinational cohort through international collaboration is essential.