Neuromyelitis Optica Spectrum Disorders in North Indian Population: Experience from a Tertiary Care Center
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.355118
Source of Support: None, Conflict of Interest: None
Keywords: Neuromyelitis optica, North Indian, NMOSD
Although clinical descriptions of neuromyelitis optica spectrum disorders (NMOSDs), traditionally known as Devic disease date back to times of Allbutt (1870) and Devic and his student Gault (1894), it is only in 2004 following discovery of highly specific neuromyelitis optica immunoglobulin G (NMO IgG) that these disorders gained worldwide acceptance.,, NMOSDs constitute a heterogeneous group of inflammatory demyelinating disorders (IDMs) which primarily affect optic nerves and spinal cord, but can affect other areas of central nervous system as well. The underlying event in pathogenesis of NMOSDs is development of autoimmunity against aquaporin-4 (AQP4) which is a cell membrane water channel protein expressed on the foot processes of astrocytes and involved in maintaining blood–brain barrier integrity. The AQP4 antibodies, also called NMO IgG antibodies, are considered pathogenic marker of NMO., Approximately 25% of the individuals with AQP4-negative NMOSDs are seropositive for antibodies against myelin oligodendrocyte glycoprotein (MOG-IgG)., NMOSDs are believed to represent a higher proportion of IDMs in the non-Caucasian population including India. Despite this, there is paucity of data on epidemiological, clinical, and investigational profile of NMOSDs in developing nations, and none from north Indian population. Thus, there is need to determine epidemiological, clinical, and investigational profile of NMOSDs as well as their response to more commonly used immunosuppressant drugs (for instance, azathioprine, cyclophosphamide, etc.) in developing countries compared to costly treatments such as pulsed intravenous (IV) immunoglobulin therapy in the West. Hence, we planned this study to characterize clinical, serological, and radiological profile of NMOSD in North Indian population.
Aims and objectives
This prospective observational study was conducted at a tertiary care referral center and teaching hospital in Northern India from July 2015 to June 2017. During this period, all patients (n = 81) who fulfilled diagnostic criteria for NMOSDs were recruited in the study. The study was approved by institutional ethics committee and written informed consent was obtained from all the patients before enrolment in the study. The diagnosis of NMOSDs was made as per International Consensus Diagnostic Criteria. The inclusion and exclusion criteria for the study are given below.
Once enrolled, all patients underwent clinical history and examination as per a predesigned proforma. These were followed up at 3 monthly intervals and on an as and when required basis. All the patients underwent gadolinium-enhanced magnetic resonance imaging (MRI) of brain and spine with thin section optic nerve cuts. Decision to carry out follow-up neuroimaging was based on patient's clinical course. All the patients also underwent detailed neuro-ophthalmologic evaluation. All patients underwent routine investigations and special investigations (computed tomography of chest and abdomen, vasculitic work-up, nerve conduction studies, and tissue biopsies wherever indicated) to rule out underlying neoplasm or associated connective tissue disorders. All patients underwent treatment as per standard guidelines for NMOSDs and according to their affordability. All the data were recorded meticulously.
Data were entered in SPSS version 23. The quantitative variables were expressed as mean, median, and standard deviation. Qualitative or categorical variables were described as frequencies and proportions. Proportions were compared using Chi-square or Fisher's exact test whichever was applicable. For normally distributed data, means of two groups were compared using Student's t-test. For skewed data, Mann–Whitney test was applied. P < 0.05 was considered statistically significant.
This study included 81 patients of NMOSDs. The mean age of study population was 33.7 ± 13.4 years (range: 14–78 years), while the mean age of onset of disease was 31.2 ± 13.5 (range: 8–75) years. The female (n = 53) to male (n = 28) ratio was 1.9:1. In all, 26 (32.1%) patients presented with optic neuritis (ON) (unilateral, 19; bilateral, 7), 46 (56.8%) with transverse myelitis (TM), while 9 (11.1%) presented with both ON and TM. 12/26 (46.2%) patients with ON were positive for NMO antibodies. The remaining 14 (53.8%) patients with ON were diagnosed by MRI findings (optic chiasma involvement, 5; involvement of >½ of length of optic nerve, 9]. The mean time from disease onset to diagnosis was 16.17 ± 23.09 months (range: 1–160 months). The mean number of episodes before diagnosis was 2.02 ± 1.643 (range: 1–14). Among patients with TM (n = 55), muscle atrophy, Lhermitte symptom, tonic spasms, flexor spasms, and flaccid bladder were seen in 63.7%, 30.9%, 43.6%, 78.2%, and 11.6%, respectively. Among patients with ON (n = 35), a Foster–Kennedy syndrome-like presentation was seen in 3 (8.6%) patients. These and other data are summarized in [Table 1].
Magnetic resonance imaging (MRI): Gadolinium-enhanced MRI of brain, optic nerves, and spine was done in all the patients. It revealed dorsal brainstem lesions adjacent to fourth ventricle in 7 (8.6%) patients and diencephalic, periependymal, and hemispheric white matter lesions in 1 (1.2%) patient each. MRI of optic nerves revealed optic nerve enhancement in 14 (17.3%) and optic nerve thinning in 3 (3.7%) patients. None of the patients with clinically pure ON had MRI evidence of lesions in cord. Among patients with TM (n = 55), 9 (16.4%) had involvement of <4 vertebral segments. MRI data are summarized in [Table 2]. MRI findings in NMO are shown in [Figure 1].
Cerebrospinal fluid (CSF) and autoantibody profile: CSF revealed pleocytosis only in 13.6% of patients while autoantibodies [NMO antibodies (n = 38)/MOG antibodies (n = 3)] were detected in 50.6% of patients [Table 2].
VEP (visual evoked potentials): VEPs (n = 80) were abnormal in 48 patients. In pure ON group, VEPs were normal in 1 and abnormal in 25 patients. In patients with pure TM, VEPs were abnormal in 17 patients (unilateral in all). In patients with both ON and TM (n = 9), VEPs were abnormal in six patients.
Management of acute attack: All patients received IV methylprednisolone (MP) (1 g daily for 5 days) during acute attack. In addition, five patients needed plasmapheresis and three needed IV immunoglobulin (IVIG). All patients who received plasmapheresis improved. One patient who received both IV MP and IVIG succumbed to his illness because of ventilator-associated pneumonia with sepsis.
Prevention of relapses (n = 80): In this study, monthly IV MP (1 g daily for 3–5 days) was used to prevent relapses in all patients as alternative immunosuppressant drugs often take 3–6 months for their action to start. The alternate drugs included azathioprine (3 mg/kg) (n = 18), IV cyclophosphamide (n = 29), mycophenolate sodium (720 mg twice daily) (n = 8), methotrexate (12.5–20 mg daily) (n = 23), IVIG (2 g/kg over 5 days) + rituximab (375 mg/m2 BSA weekly for 4 weeks) (n = 2). Various drug-related adverse effects included vomiting (n = 5; all during administration of cyclophosphamide), pancytopenia (n = 1; on azathioprine), and varicella zoster (n = 3; one on cyclophosphamide and two on rituximab). Treatment changes were required in 15 patients because of relapses or toxicity. Six of 18 (33%) patients on azathioprine (Poor efficacy-5; pancytopenia-1) required change to alternate drugs (cyclophosphamide, 3; mycophenolate, 1; rituximab, 2). Nine of 23 (39.1%) patients on methotrexate (poor efficacy, 9) needed change to cyclophosphamide. None of the patients in cyclophosphamide group needed drug change. Mycophenolate needed to be replaced by cyclophosphamide in two and rituximab in two patients.
Follow-up: Good response to treatment was noted in most patients. The mean follow-up duration was 15.3 ± 6 months (range: 6–32 months). The mean follow-up was 16.1 months in ON group and 14.8 months in TM group. Schwab and England activities of daily living (ADLs) scale and expanded disability status scale (EDSS) were used to follow patients with TM, whereas patients with ON were followed by ADL scale only. The EDSS and ADLs data are shown in [Table 3].
Various predictors of poor outcome in NMOSDs
We determined various demographic, clinical, and radiological features which could predict poor outcome in NMOSDs. For this, patients with pure ON, pure TM, and both ON and TM were divided into two groups: (a) those with poor outcome (ADL score ≤60%) and (b) those with satisfactory or good outcome (ADL score >60%). Patients with TM were further divided into three groups based on EDSS score: (a) poor outcome (EDSS ≥6), (b) moderate outcome (EDSS = 5–5.5), and (c) good outcome (EDSS ≤4.5). On analysis, we could not find any factor which determines poor outcome at disease onset. The presence of NMO antibodies did not have any bearing on clinical/investigational parameters, response to treatment, and outcome.
NMOSD versus MS
During the study period we enrolled 32 patients in our MS registry. Thus, NMOSDs patients outnumbered patients with MS by a factor of 2.53.
It is imperative to study NMOSDs in different populations to promote our understanding of global profile of NMOSDs and implement strategies to address this debilitating disorder. The mean age in this study (33.67 ± 13.41 years) was in agreement with prior studies from French West Indies, Denmark, and Southern India. The lower age of ON (28.42 ± 9.71 years) compared to patients with TM (35.25 ± 14.41 years) was in agreement with Kitley et al. The female-to-male ratio of 1.9:1 in our cohort was lower than previous studies., The differentiation of NMOSDs into ON, TM, and TM + ON groups was also in accordance with previous studies from Spain and Brazil. The man time from disease onset to diagnosis was 16.17 ± 23.09 (range: 1–160) months and the majority (n = 59; 72.8%) of these patients were treated as alternative disorders. These values are alarmingly high for a disorder which is known to present acutely and leave considerable residual neurological sequalae in absence of treatment. Thus, need to educate primary care physicians about NMOSDs cannot be overemphasized. Regarding clinical features of TM, our findings were similar to Jacob et al. and Pandit and Kundapur who reported Lhermitte symptoms, bladder involvement, and tonic spasms to be common in TM. Sato et al. reported cervicomedullary symptoms in 21% of patients with NMOSD compared to 31% in our study. Relapses are reported to affect 80%–90% of patients with NMOSD at 3 years of follow-up. However, we found relapsing course only in 18% of patients. The cause for this discrepancy may be the fact that all our patients (even with one episode) received long-term immunomodulation and relatively short duration of follow-up. Coexisting illnesses (malignancies and other autoimmune diseases) were present in 9 of 81 (11.1%) patients. Thus, NMOSDs should be considered in all patients who present with suggestive clinical features in patients with malignancies or other autoimmune disorders. In all, 38 (46.9%) patients were AQP4 antibody-positive, whereas 3 (3.7%) were anti-MOG antibody-positive. In a study from Southern India, NMO IgG was positive in 39% of patients. The lower positivity rate of NMO IgG antibodies in our cohort compared to western studies may be related to difference in population characteristics and better awareness of existence of NMOSDs whereby a large number of patients with AQP4 antibody-negative NMOSDs were diagnosed and included in our cohort. About 16.3% of patients with TM had <4 vertebral involvement on MRI spine. Similarly, 16.3% of patients had involvement of <50% of cross-sectional area of the cord. Similarly, gadolinium-enhanced MRI of brain with thin section ON cuts was normal in 18 of 35 (51.4%) patients. This stresses on need to keep high index of suspicion for diagnosis of NMOSDs even in absence of MRI findings which are considered classic of NMOSDs. MRI brain revealed abnormalities in 54.3% of patients similar to a Korean study. There is a lack of controlled trials evaluating treatment of NMOSDs and treatment recommendations are based on observational studies and expert opinion. In our protocol, we managed all acute attacks (n = 81) with IV MP (1 g daily for 5 days). Patients who did not improve (n = 8) received plasma exchange (PE) (n = 5) or IVIG. Crout et al. have recommended IV MP in acute attacks of NMO followed by PE, while Elsone et al. have recommended IVIG for acute relapse of NMO. There are no well-established guidelines on prevention of relapses in NMO. Crout et al. have recommended treatment with immunosuppressant drugs such as oral prednisolone, azathioprine, mycophenolate, methotrexate, IV cyclophosphamide, and IV rituximab for at least 5 years after an acute episode. Oral prednisolone is recommended in dosage of 1 mg/kg/day followed by tapering to lowest effective dose. In our protocol, we used pulse IV MP (1 g daily for 3–5 days) every month for 6 months instead of oral prednisolone as maintenance therapy. It is worthwhile to note that none of our patients experienced common steroid related side effects such as cushingoid features, infections, or cataracts which are routinely noted in patients on long-term oral steroids. Thus, IV MP is a suitable alternative to oral steroids. In this study, we chose other immunosuppressant drugs based on the severity of the episode and patient's financial status. For severe episodes with marked disability or patients with relapses on oral therapy, IV cyclophosphamide or IV rituximab (depending on financial status) was used. For patients with financial constraints, we used methotrexate and IV cyclophosphamide. For unmarried women and those planning pregnancy in the future, azathioprine was the drug of choice.
ADL and EDSS
Most of our patients had good outcomes at follow-up [Table 3]. Various predictors of poor outcome described in NMO/NMOSD include older age of onset, severe first episode, seropositive disease, multiple relapses within the initial 2 years of diagnosis, and coexistent autoimmune disorders., In this study, we did not find any factor which could adversely affect outcome of NMOSDs. This is likely related to the fact that we treated all our patients with aggressive immunomodulation form beginning and most of patients in our cohort had good outcomes.
To conclude, NMOSDs are a common cause of demyelinating illnesses in Northern India. The response to treatment is excellent and most patients recover without residual disability. A high index of suspicion, appropriate investigations, and aggressive immunomodulation are the keys to successful outcome.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2], [Table 3]