Neuromyelitis optica (NMO) is a severe inflammatory disorder of the CNS of putative autoimmune aetiology, which predominantly affects optic nerves and spinal cord. Gluten sensitivity refers to an autoimmune condition characterised by the presence of antibodies to tissue transglutaminase usually associated with gastrointestinal symptoms. However, neurological complications associated with gluten sensitivity, ranging from peripheral neuropathy to inflammatory brain lesions mimicking multiple sclerosis (MS), have been described. In a recent case report, published in this journal, some of us reported on the first and, to our knowledge, so far the only cases of serologically and histologically confirmed gluten sensitivity and NMO.1

A major contribution to the diagnosis of NMO was the recent description of a newly detected serum reactivity in patients with NMO (called NMO-IgG by the authors).2 Later, aquaporin-4 (Aqp4), the most abundant water channel in the CNS, was identified as its target antigen3 and several assays have been developed.3 4 Histopathological studies demonstrating Aqp4 loss in association with immunoglobulin and complement deposits, and reports on a beneficial effect of rituximab and plasma exchange in a subset of NMO patients, suggest a potential pathogenic role of the antibody.

We have established a recombinant, fluorescence based immunoprecipitation assay for the detection of Aqp4-Ab, which has been found to have 76% sensitivity and 100% specificity,5 comparable with the original assays. Both patients reported previously1 were found to be positive for Aqp4-Ab (table 1); our results were confirmed by indirect immunofluorescence testing for NMO-IgG (performed at the Mayo Medical Laboratories).

NMO has been reported to be associated with other autoimmune conditions in 28–38% of cases, and non-organ specific antibodies can be found in approximately 50%. A recent study reported NMO-IgG in patients with NMO associated with connective tissue disorders (CTD), such as Sjögren syndrome or systemic lupus erythematosus, with almost the same frequency as in patients with uncomplicated NMO. This favours the coexistence of the two autoimmune disorders rather than a causal link between NMO and CTD or non-organ specific autoimmunity.6 The same may hold true for our patients, taking into account the apparent rarity of gluten sensitivity among NMO patients.

If patients with NMO report gastrointestinal symptoms, it is important to consider the possibility of gluten sensitivity. Such symptoms are often ignored, particularly if they are receiving treatment with azathioprine which can cause similar symptoms.

During the past decades, a broad variety of further disorders have been reported in the literature in association with NMO (including tuberculosis, sarcoidosis, hepatitis C, etc) and a causal link was proposed by many of the authors, resulting in the impression that NMO might represent a common phenotype with diverse aetiologies rather than a distinct medical entity. Now that reliable assays for Aqp4 antibodies are available, it should be possible to reinvestigate these cases and evaluate the relationship between these conditions, NMO and Aqp4 antibodies.

Acknowledgements

We thank the patients and their consultants, Dr H Allroggen and Dr A Kenton for allowing us to publish the updated laboratory results.