FINAL DIAGNOSIS: NEUROMYELITIS OPTICA
An extensive work-up was conducted for potential etiologies of an inflammatory pseudotumor. A full serologic panel was positive only for anti-aquaporin IgG antibody, consistent with the neuromyelitis optica spectrum of disorders. On further questioning, the family recalled that the patient had been evaluated at another facility 2-3 years prior to the current admission for a "bad eye." Formal ophthalmological examination confirmed visual impairment in the right eye.
Neuromyelitis optica was once considered an aggressive variant of multiple sclerosis (MS), but with some differences (5). For example, unlike MS, NMO is relatively common in sub-Saharan Africa, Asia, India, and the Caribbean, and interferon-beta is not an effective therapy (7). Furthermore, detection of serum or CSF anti-aquaporin IgG (a.k.a. NMO-IgG) has been shown to be >90% specific and 75% sensitive for NMO (1, 4). These findings led to the modified NMO diagnostic criteria, which included optic neuritis and acute myelitis as major criteria plus either a spinal MRI lesion longer than 3 vertebral segments or the presence of anti-aquaporin-4 (AQP4) antibody (9).
The mean age of NMO is around 40 years, but it is known to occur in children. In such cases the chief differential at presentation is usually acute disseminated encephalomyelitis, transverse myelitis, or severe MS. Relapses with severe sequelae have been reported, especially if the patient is NMO-IgG positive at initial presentation (1). Other risk factors for a relapsing clinical course include female gender and severe motor weakness at presentation. Each recurrent attack during the first 2 years of disease increases risk of early death (10).
In keeping with the presence of circulating anti-AQP4 antibody and the in vitro work discussed below, longstanding NMO lesions typically show reduced AQP4 immunostaining. Ischemic lesions and inactive MS plaques also show reduced AQP4 reactivity (6). In our case the tissue showed increased AQP4 immunoreactivity via fluorescent microscopy, especially in what appear to be macrophages (Fig. 15). This increased staining may be due to the acute phase of the disease.
The general function of aquaporin channels is to increase plasma membrane osmotic permeability by facilitating the bidirectional movement of water and, in some cases, glycerol. Of the dozen or so aquaporins described in mammals thus far, AQP4 is the main subtype found in the brain and is primarily expressed at key parenchymal/fluid interfaces (e.g. astrocyte foot processes, ependyma, subependymal, and glia limitans), suggesting a critical role in maintaining osmotic balance in the brain. It is present in the spinal cord and optic nerves as well as other areas now recognized as sometimes involved by NMO, including the third and fourth ventricles and central aqueduct (2).
Proposed pathologic mechanisms of action by the anti-AQP4 antibody include endocytosis and degradation of membrane-bound AQP4, disrupting the blood brain barrier, destroying AQP4-producing astrocytes via complement fixation, and/or by indirect killing of nearby oligodendroglial cells via glutamate excitotoxicity (2, 3, 8)
After the diagnosis was established, the patient was begun on a course of low-dose oral corticosteroids and azathioprine. Nevertheless, she developed separate relapses 4, 7, and 11 months after the original diagnosis, requiring high-dose IV Solu-Medrol each time. Plasmapheresis was initiated during the second relapse, producing temporary remission and underscoring the humoral nature of this disease. A fourth minor relapse 15 months after diagnosis prompted replacement of azathioprine with mycophenolate. While the patient has not yet demonstrated additional deficits, she has developed side effects to the steroid therapy, including Cushingoid features and diabetes.
Contributed by Craig Horbinski, MD, PhD, Ian F. Pollack, MD, Clayton Wiley, MD, PhD, and Geoff Murdoch, MD, PhD