Diffusion tensor imaging characterization of occult brain damage in relapsing neuromyelitis optica using 3.0T magnetic resonance imaging techniques

Abstract

Studies of relapsing neuromyelitis optica (RNMO) using advanced MRI techniques are limited compared with those done on multiple sclerosis (MS). The present study used diffusion tensor imaging (DTI) to investigate whether occult brain damage exists in RNMO patients. DTI scans using a 3.0T MRI scanner were performed in 24 clinically confirmed RNMO patients whose conventional brain MRI results were normal, and also in 24 age- and sex-matched healthy control subjects. DTI data were processed to generate fractional anisotropy (FA) and mean diffusivity (MD) maps, and region of interest (ROI) analyses were performed to obtain these parameters in white matter (including medulla oblongata, cerebral peduncle, optic radiation, genu of corpus callosum, splenium of corpus callosum, and internal capsule) and gray matter (including thalamus and putamen). Regional measures from patients at stable and acute phases were compared with healthy controls. Both acute and stable NMO patients had a higher average FA in ROIs of the thalamus and putamen. Acute NMO patients had significantly higher average MDs than controls in the genu of corpus callosum and optic radiation, and significantly lower average MDs in the medulla oblongata. Stable NMO patients had increased MDs in the genu of corpus callosum and optic radiation, but lower MDs in the medulla oblongata, internal capsule and thalamus. The DTI findings confirm the presence of occult tissue damage in normal-appearance white and gray matter, especially deep gray matter, in RNMO patients. This study adds further to the evidence that DTI is suitable as a tool for characterizing subtle brain tissue damage.

Introduction

Neuromyelitis optica (NMO) is a severe form of demyelinating autoimmune disease characterized by the selective involvement of the optic nerves and spinal cord in a mono- or multi-phasic manner (Devic, 1984, Ghezzi et al., 2004, Wingerchuk et al., 1999). The incidence rate of NMO is about 1 in 100,000 in Western countries (Marignier et al., 2010), however, there are no precise statistics for China. The disease is also known as Devic's disease or Devic's syndrome to commemorate the author of the initial report (Devic, 1984). Age at onset follows a unimodal distribution with a peak between 21-and 41-years-of-age. The distinctive clinical manifestation is the co-occurrence of optic neuritis with myelitis (Yamakawa et al., 2000). Viral infection may have an important relationship with the disease, although whether NMO is evoked directly by viral infection, as suggested by Merle et al. (1998), is unclear. It remains contentious whether NMO is a form of multiple sclerosis or a separate neurological entity, although more recent evidence favors the latter view. The evidence includes epidemiology, course of disease, clinical manifestation, pathology, CSF analyses, and especially, the discovery of serum autoantibody IgG, which may be a sensitive and specific marker for NMO (de Seze et al., 2002, de Seze et al., 2003, Lucchinetti et al., 2002). Despite the apparent differences between multiple sclerosis and NMO, it remains difficult to distinguish the two conditions, both clinically and in lab analyses. MRI is an important tool in diagnosing NMO. As a consequence, a new set of diagnostic criteria was proposed for NMO, with a supportive criterion being a contiguous spinal cord MRI lesion extending three or more segments (Wingerchuk et al., 1999). A more recent revision of the criteria considered a brain MRI not meeting the diagnostic criteria for MS as a supportive criterion for NMO (Wingerchuk et al., 2006).

Recently, along with the development of MRI and increased recognition of NMO, quantitative MRI techniques including DTI, MTI and fMRI have been used to study NMO. These techniques may provide more detail than is possible using MRI, and may be useful in the diagnostic work-up of patients and increasing the understanding of NMO pathobiology. However, there has been no study involving a large number of patients concerning the possible presence of abnormalities in normal-appearance brain tissue. The present study investigated whether and where occult brain damage exists in RNMO patients using DTI.