What Distinguishes Neuromyelitis Optica from Multiple Sclerosis?

Abstract & Commentary

By Susan A. Gauthier, DO, MPH, Assistant Professor of Neurology, Weill-Cornell Medical College. Dr. Gauthier reports no financial relationships relevant to this field of study.

Synopsis: Neuromyelitis optica (NMO) may be a unique central nervous system demyelinating disease with distinct clinical and MRI manifestations, as well as different immune and pathological characteristics as compared to multiple sclerosis (MS).

Sources: Argyriou AA, Makris N. Neuromyelitis optica: a distinct demyelinating disease of the central nervous system. Acta Neurol Scand 2008;118:209-217; Li Y, Xie P, Lv F, et al. Brain magnetic resonance imaging abnormalities in neuromyelitis optica. Acta Neurol Scand 2008;118:218-225.

Neuromyelitis optica (NMO) is an autoimmune inflammatory disease of the central nervous system, usually included within the spectrum of multiple sclerosis, however, more recently, NMO has been proposed to be a distinct clinical entity. NMO characteristically results in demyelination of the optic nerves and spinal cord, resulting in relapsing episodes of unilateral or bilateral visual loss and/or severe paraparesis; this differs from the milder episodes in MS. Similar to MS, the majority of patients have a relapsing course (80-90%), with a female predominance; however, patients with NMO rarely enter a secondary progressive stage and disability accumulates as a result of severe and frequent relapses.

Pathological damage in NMO ranges from demyelination to necrosis within the optic nerves and spinal cord, with the humoral immune system having a predominant role as compared to MS. Similar to the type II pathological subset of MS, NMO lesions show prominent immunoglobulin and complement deposits. Further implicating the role of B cells in NMO, as opposed to the predominant T-cell mediated damage found in MS, was the discovery of an antibody to aquaporin 4, also referred to as NMO-IgG. Aquaporin 4 is the predominant water channel in the brain and is located in the foot processes of the astrocytes surrounding the blood-brain barrier. A predominant perivascular distribution of immune complex deposition is found in NMO corresponding to the normal expression of aquaporin 4 in the foot processes of astrocytes. The sensitivity of NMO-IgG seropositivity alone for the diagnosis of NMO is only 73%, although the specificity is 91%. However, the recently revised criteria for NMO provides a 99% sensitive and 90% specific diagnostic combination that includes two absolute criteria (optic neuritis and acute myelitis) and at least two of three supportive criteria (a longitudinally extensive cord lesion, brain MRI not meeting criteria for MS at disease onset, or NMO-IgG seropositivity).

Intravenous methylprednisolone is used for acute relapses in NMO in doses similar to MS. In NMO relapses refractory to corticosteroid treatment, rescue therapy with plasma exchange has been found to be beneficial. In general, MS immunomodulatory agents are ineffective in preventing relapses in NMO and the use of an oral immunosuppressant such as azathioprine (2-3 mg/kg/daily) in combination with oral prednisone is the most common prophylactic treatment approach. In a recent study of rituximab, which targets CD20-positive B-cells, stabilization was found in 7 of 8 patients; the small sample size of the study limits any clear conclusions, but is a promising observation.

Classically, NMO was felt to only affect the optic nerves and spinal cord and to spare the brain, yet many recent studies have shown that brain abnormalities are not uncommonly found in patients with NMO. In a series published by the Mayo Clinic, 60% of NMO patients had evidence of brain lesions; the majority were non-specific white matter lesions and occasionally were lesions resembling MS.1 A subsequent publication focused on lesions corresponding to sites of high aquaporin 4 expression.2 Although these were the minority of lesions found in NMO patients, the locations occurred most commonly around the third and forth ventricles, as well as the aqueduct of Sylvius; this corresponded to the periventricular and hypothalamic localization of aquaporin 4. These lesions were felt to be characteristic MRI lesions of NMO. Recently, Li et al described brain lesions in a series of Chinese patients with NMO for which 28 of 33 (84.6%) had brain parenchymal abnormalities. Well defined brain lesions were found in 22 patients (66.7%), with the majority having more than one lesion. Most lesions were scattered and described as small or punctuated (<3 mm), although unique patterns were described; these included symmetrical hyperintense T2 signal changes in the bihemispheric deep white matter regions and belt-like hyperintensities along the margin of the lateral ventricles. Brainstem involvement was higher, specifically in the medulla, in this series of patients (42.4%) as compared to other published clinicopathological studies. Only one patient had involvement of the hypothalamus. The Li et al study concluded that the prevalence of brain lesions in Chinese NMO patients was higher than previously reported in other populations with NMO and that the pattern of involvement differs in the Chinese population.


Distinct clinical differences between NMO and MS have been well recognized; although without evidence of pathophysiological differences, patients were generally treated with MS therapies. With the discovery of NMO-IgG, the treatment approach to this disease has changed; specifically, there is a focus on general immunosuppression and, recently, on anti-B-cell therapies. As the diagnostic use of the NMO-IgG test expands, a better understanding of the clinical spectrum of NMO will follow, as well as the opportunity to determine the actual rate of brain involvement. The study by Li et al found, as others have, that the majority of brain lesions in NMO are non-specific small white matter abnormalities; however, they identified unique features in their patient population. The Chinese population lacked the characteristic NMO lesions described in the Mayo series; however, the small sample sizes in both studies limit any definite conclusions.


1. Pittock SJ, Lennon VA, Krecke K, et al. Arch Neurol 2006;63:390-396

2. Pittock SJ, Weinshenker BG, Lucchinetti CF, et al. Arch Neurol 2006;63:964-968.