Article Text

Download PDFPDF

Magnetic resonance spectroscopy evaluation in patients with neuromyelitis optica
  1. Jérôme de Seze1,
  2. Frédéric Blanc1,
  3. Stéphane Kremer2,
  4. Nicolas Collongues1,
  5. Marie Fleury1,
  6. Christophe Marcel1,
  7. Izzie-Jacques Namer3
  1. 1Department of Neurology, Strasbourg University, Strasbourg, France
  2. 2Neuroradiology, Strasbourg University, Strasbourg, France
  3. 3Nuclear Medicine, Strasbourg University, Strasbourg, France
  1. Correspondence to Professor Jérôme de Seze, Department of Neurology, Hôpital Civil, 1 place de l'hôpital, BP 426, 67091, Strasbourg cedex, France;{at}


Objective Neuromyelitis optica (NMO) is an inflammatory disease associated with optic neuritis and myelitis. Although some studies have reported multiple sclerosis (MS)-like lesions in 10–30% of NMO patients, brain MRI is usually normal. Several studies have observed metabolic abnormalities on MR spectroscopy in MS, even in normal-appearing white matter (NAWM). To the authors' knowledge, MR spectroscopy has never been used to investigate NMO. The aim of this study was to evaluate metabolic abnormalities in the NAWM and normal-appearing grey matter (NAGM) of NMO patients.

Methods The authors evaluated 24 patients (17 women and seven men, with a mean age of 44.6 years). NMO was diagnosed according to revised criteria. All patients had a brain and spinal cord MR imaging including MR spectroscopy sequences in both NAWM and NAGM. Patients were compared with 12 healthy subjects.

Results NAA/creatinine ratios in NAWM (1.89±0.26 in NMO compared with 1.91±0.15 in control subjects) and NAGM (1.62±0.21 compared with 1.59±0.18) were normal, as were choline/creatinine ratios in NAWM (1.03±0.18 compared with 1.08±0.14) and NAGM (0.89±0.2 compared with 0.94±0.2). Myo-inositol values in NAWM were also normal (0.42±0.12 compared with 0.42±0.18).

Conclusion Our results are clearly different from those found in MS, where NAA is frequently decreased and choline increased, even in NAWM. Our findings could have an impact on the differentiation between MS and NMO.

  • Spectroscopy
  • NMO
  • MRI
  • MRS
  • multiple sclerosis
  • neuroradiology

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Neuromyelitis optica (NMO) is an inflammatory disease associated with optic neuritis and myelitis.1–3 In the past, NMO was frequently considered to be a form of multiple sclerosis (MS), but there is now much evidence to suggest that they are different diseases. The brain has generally been considered to be spared in NMO, although recent MRI and immunopathological findings suggest that tissue damage in NMO is more extensive than previously thought, involving the brainstem or brain.4–9 Although some studies have reported MS-like lesions in about 10–30% of NMO patients, brain MRI is usually normal.3 However, recent evidence suggests that brain lesions are more frequent than previously thought in NMO; indeed, lesions unlike those occurring in MS have been detected in the hypothalamus and around the 4th ventricle, corresponding to areas with a high concentration of aquaporine 4, in NMO patients.6

In MS, several studies observed metabolic abnormalities on MR spectroscopy, even in normal-appearing white matter (NAWM) and normal-appearing grey matter (NAGW).10–12 These studies showed a decrease in N-acetyl-aspartate (NAA) in MS patients, suggesting infra-radiological destruction, including axonal loss. NMO is a severe disease that is frequently more disabling than MS.3 However, progression between relapse episodes is very rare.13 This suggests that, in contrast to MS, NMO is a more regionally restricted disease and that axonal loss and necrosis are only located in the region affected by inflammatory lesions, without whole-brain abnormalities. To our knowledge, MR spectroscopy has never been used to investigate NMO.

The aim of our study was to evaluate metabolic abnormalities in the NAWM and NAGM of NMO patients.



The study was cross-sectional. We prospectively and consecutively included 24 NMO patients (17 women and seven men, with a mean age of 44 years). NMO was diagnosed according to the revised criteria recently proposed by Wingerchuck et al.14 Clinical, demographic and laboratory data are summarised in table 1. All patients had definite NMO according to the recently revised criteria.14 All but one patient had a relapsing form of NMO. All patients but four had bilateral optic neuritis (ON). We compared these 24 NMO patients with a control group of 12 healthy volunteers from our database. All patients and controls gave their written informed consent to participate in the study. The study was approved by the ethics committee of Strasbourg hospital.

Table 1

Clinical, laboratory and MR spectroscopy data for neuromyelitis optica (NMO) patients and controls

MRI evaluation

All patients had a brain and spinal cord MRI on a 1.5 T machine (Siemens, Ehrlangen, Germany). In addition to the usual protocol for the follow-up of these patients, we performed MR spectroscopy sequences in both NAWM and NAGM.

All MR studies were conducted with a 1.5 T MR imager (Siemens 1.5 T Avanto system). MR spectra were obtained using a double spin-echo point-resolved spectroscopy (PRESS) sequence with two-dimensional 16×16 phase-encoding (2D-MRSI). A transverse plane was positioned in the centrum ovale within the volume of interest resulting in maps of 8×8 spectra (figure 1). Measurement parameters were as follows: TR, 1500 ms; TE, 135 ms; slice thickness, 15 mm; FOV, 240 mm; data points, 1024; and acquisition time, 7 min. An additional single-voxel acquisition was performed with an 8 cm3 voxel (2×2×2 cm); TE, 30 ms; and acquisition time, 3 min. The field homogeneity achieved in automated non-localised multiple angle projection shimming resulted in water peak linewidths <8 Hz in the volume of interest. Water suppression was achieved with 2.56 ms sinc-Hanning-shaped RF pulses for chemical shift selective excitation (CHESS). The 2D-MRSI and single-voxel raw data were analysed with standard postprocessing software (Siemens, Germany). For 2D-MRSI, postprocessing included zero-filling to 32×32 and apodisation in the spatial domain, which resulted in an effective voxel size of 0.84 cm3. The time-domain data were zero-filled to 2048 data points and apodized with a Gaussian function, which corresponded to 2 Hz line-broadening in the frequency domain. Spectral phasing and a polynomial baseline correction were also performed. The signals of NAA, choline-containing compounds (Cho), total creatine (Cr), the lactate doublet (Lac) and mobile lipids (Lip) were quantitated by means of curve fitting. The spectral regions for the numerical integration of peaks were: 3.22 ppm for Cho, 3.03 ppm for Cr, 2.02 ppm for NAA, 1.32 ppm for Lac and 0.8–1.5 ppm for Lip peak. Quantification of the metabolites was done from the CSI 135 ms by taking the mean of three spectra for white matter and grey matter for Cho, Cr and Lac, from a single voxel from the SVS 30 ms for myo-inositol.

Figure 1

Voxel placement of MR spectroscopy.

Data analysis

Results were expressed as mean±SD. Statistical evaluation was performed using a χ2 test for qualitative values and non-paired Student t test for quantitative values.


Eleven patients (46%) had T2-weighted abnormalities on brain MRI, including seven patients (29.1%) with non-specific lesions and four (16.7%) with MS lesions.

The main MRI and MR spectroscopy findings are summarised in table 1. In NMO patients, the NAA/creatinine ratios in NAWM (NMO: 1.89±0.26; controls: 1.91±0.15) and NAGM (NMO: 1.62±0.21; controls: 1.59±0.18) and the choline/creatinine ratios in NAWM (NMO: 1.03±0.18; controls: 1.08±0.14) and NAGM (NMO: 0.89±0.2; controls: 0.94±0.2 in healthy controls) were not different from the values in normal subjects. Lastly, myo-inositol values in NAWM were not different in NMO patients when compared with the control group (NMO: 0.42±0.12; controls: 0.42±0.18).

Furthermore, we did not find any differences in spectroscopy findings when we compared patients with and patients without brain MRI T2 lesions.


Our study demonstrates that MR spectroscopy findings are normal in both NAWM and NAGM in NMO patients for the main metabolic parameters (NAA, choline and myo-inositol), corresponding to axonal loss, inflammation and gliosis, respectively. This is clearly different from the findings in MS, where NAA is frequently decreased and choline increased, even in NAWM.10–12 Our present results are somewhat surprising considering our recent findings on cognitive disturbances in NMO.15 In that study we found a similar profile and frequency of cognitive impairment in two MS and NMO cohorts. Those results led us to suspect diffuse brain tissue damage in NMO. This apparent discordance could be explained by the choice of brain region for spectroscopy. In the present study, we focused on NAWM in the centrum ovale within the volume of interest, resulting in maps of typically 8×8 spectra. Whole-brain spectroscopy or the analysis of other regions, such as the temporal lobe (and especially the hippocampus), would therefore be interesting. However, the cognitive abnormalities in NMO might also be explained by mechanisms other than NAWM and NAGM involvement.

Our results could have an impact on the differentiation between MS and NMO in cases presenting differential diagnosis difficulties. It is interesting to note that, even in the few patients with brain T2 lesions, MR spectroscopy findings remained normal, suggesting that NMO is a more regionally restricted disease than MS, where diffuse tissue damage is frequently observed. In this respect, it would be interesting to perform MR spectroscopy in patients with acute demyelinating encephalomyelitis, which is unlikely to be associated with NAWM/NAGM alterations. However, MR spectroscopy is rarely used as part of the routine diagnosis strategy, as the majority of studies showed intergroup differences but also found a considerable overlap between normal controls and MS patients.

Our results are in good correlation with clinical findings in NMO showing a monophasic or relapsing evolution but relatively restricted to optic nerve and spinal cord. Furthermore, inter-relapse progression, which has been reported only very rarely in previous studies on NMO,13 was not observed in any of the patients in our study. In MS, however, disability progression is frequently observed, even in patients without relapses, especially in secondary progressive MS, suggesting subclinical destruction. A recent study with diffusion tensor tractography found that, in NMO, abnormalities remained restricted to both optic and corticospinal tracts.9 All these results argue in favour of a different pathophysiological process in NMO compared with MS, as demonstrated by recent neuropathological studies.4 5

Our study, based on metabolic measurement using MR spectroscopy, provides further evidence that the brain is frequently spared in NMO. MR spectroscopy could usefully be added to the diagnostic work-up for differentiation between MS and NMO in some difficult cases.



  • Competing interests None.

  • Ethics approval Ethics approval was provided by the Strasbourg.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.