Does detection of anti-AQP4 antibodies trump clinical criteria for neuromyelitis optica?

Neuromyelitis optica (NMO) is a common or even predominant form of CNS demyelinating disease in Asian and African countries, but controversy as to how it can be distinguished from multiple sclerosis (MS) in Asian countries persists. In this issue of Neurology ®, Siritho et al.1 report a retrospective analysis of 135 consecutive Thai patients with idiopathic inflammatory CNS demyelinating disease. These patients were categorized sequentially by satisfying the most specific to least specific clinical criteria, as follows: NMO, based on Wingerchuk et al. 2006 criteria2 (excluding anti-aquaporin 4 [AQP4] seropositivity); NMO spectrum disorder (limited forms of myelitis or optic neuritis syndromes, in some cases with brain lesions now regarded as suggestive of NMO); optic-spinal MS (dominant optic neuritis and myelitis but lacking a longitudinally extensive spinal cord lesion [LESCL], precluding a diagnosis of definite NMO); classic MS; and indeterminate clinically isolated syndromes. Serum was tested blind in Japan for AQP4 autoantibodies using a sensitive cell-based immunofluorescence assay; 40% were seropositive. Although the proportion of seropositive cases in each category declined from NMO (78%) to clinically isolated syndrome (6.3%), the proportion of seropositive cases among the total study population was equally distributed among 3 categories: 34% NMO, 36% NMO spectrum disorder, and 30% others, including classic MS. Considering the specificity of AQP4 antibodies as demonstrated worldwide, at first glance …