- Email: [email protected]
Diffusion tensor imaging of normal-appearing white matter in neuromyelitis optica
Journal of Neuroradiology (2012) 39, 295—300
Available online at
www.sciencedirect.com
ORIGINAL ARTICLE
Diffusion tensor imaging of normal-appearing white matter in neuromyelitis optica Imagerie en tenseur de diffusion de la substance blanche d’allure normale dans la neuromyélite optique J. Jeantroux a,∗, S. Kremer a, X.Z. Lin a, N. Collongues b, J.-B. Chanson b, B. Bourre b, M. Fleury b, F. Blanc b, J.-L. Dietemann a, J. de Seze b a b
Department of radiology 2, hôpital de Hautepierre, university hospitals of Strasbourg, avenue Molière, 67000 Strasbourg, France Department of neurology, university hospitals of Strasbourg, avenue Molière, 67000 Strasbourg, France
KEYWORDS Neuromyelitis optica; MRI; Diffusion tensor imaging
Summary Objectives: Neuromyelitis optica (NMO) is an inflammatory disease of the central nervous system characterized by severe attacks of optic neuritis and myelitis. Brain was classically, unlike in multiple sclerosis (MS), spared. Nevertheless recent studies showed that brain lesions can be seen with MRI. We studied the diffusion characteristics of normal-appearing white matter (NAWM) and abnormal white matter in NMO patients compared with NAWM in healthy subjects. Patients and methods: Diffusion tensor imaging (DTI) scans of the brain and spinal cord were obtained from 25 patients with NMO and 20 age- and gender-matched healthy subjects. Region of interest (ROI) analysis of the apparent diffusivity coefficient (ADC) and fractional anisotropy (FA) was performed in brain NAWM (optic radiations, corpus callosum [CC] and anterior and posterior limbs of the internal capsule [IC]) and in spinal cord NAWM and in lesions. Results: ADC was increased and FA decreased in NMO patients in the posterior limb of the IC in the optic radiations and in spinal cord NAWM. FA was lower in spinal cord lesions. In contrast, there was no difference between the two groups in the anterior limb of the IC nor in the CC. Conclusion: These results suggest that DTI abnormalities are very severe in NMO spinal cord lesions. In our study, DTI abnormalities in NAWM were restricted to optic radiations and corticospinal tracts, suggesting secondary Wallerian degeneration. In contrast, NAWM outside these tracts (CC and anterior IC) remained normal suggesting that, unlike what is observed in MS, there is no infra-lesional abnormality in NMO. © 2011 Elsevier Masson SAS. All rights reserved.
∗ Corresponding author. Tel.: +33 3 88 12 78 90; fax: +33 3 88 12 71 18. E-mail address: [email protected] (J. Jeantroux).
Neuromyelitis optica (NMO) is an inflammatory disease of the CNS characterized by severe attacks of optic neuritis and myelitis, which, unlike multiple sclerosis (MS), commonly spares the brain in the early stages. It has become increasingly apparent, however, that both
0150-9861/$ – see front matter © 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.neurad.2011.10.003
296 symptomatic and asymptomatic MRI brain lesions may occur in NMO [1—6]. Postmortem studies have confirmed that the rare brain lesions observed on MRI in NMO patients have the same immunohistochemical characteristics as spinal cord lesions [4,7,8]. A few studies have reported that patients with NMO appear to have little damage in either normalappearing grey matter (NAGM) or normal-appearing white matter (NAWM) and others showed a lack of abnormalities, for example using MR spectroscopy [9—12]. Diffusion tensor imaging (DTI) helps to measure the random motion of water molecules and provides information on the structural and orientational features of tissues at a microscopic level [13—17]. Some rotationally invariant indices, such as the apparent diffusivity coefficient (ADC) and fractional anisotropy (FA), which are derived from DTI, can provide information on the magnitude and directionality of water diffusion in brain tissues. ADC reflects average diffusivity of water molecular motion, independently of any tissue directionality and is affected by cellular size and integrity. FA measures the degree of directionality of water diffusion and thus reflects the degree of alignment of cellular structures within fibre tracts, as well as their structural integrity [18,19]. Within a coherently arranged white matter tract, water molecules diffuse faster in the direction parallel to the tract than in perpendicular directions. Pathological processes, by changing the microstructural environment, result in altered diffusion [20,21]. DTI seems to be a promising tool for the quantification of tissue damage and for improving our understanding of conditions that affect the integrity and organization of brain tissues. Up to now, there have only been a few published studies on DTI in NMO, all performed by the same team [11,22,23]. The aim of the present study was to evaluate DTI abnormalities in brain NAWM, optic radiations and spinal cord of NMO patients in order to increase our knowledge of the pathophysiological mechanisms in this rare pathology.
Patients and methods Subjects From April 2007 to December 2008, we prospectively included 25 patients (16 women and nine men; mean age 43.7 ± 12.3 years) with definite NMO (mean disease duration 6.1 ± 2.5 years), diagnosed according to the revised criteria proposed by Wingerchuck et al. [24]. All patients but two had a relapsing form of NMO. The two remaining patients had a monophasic course with myelitis and optic neuritis at the same time. All patients had at least one episode of transverse myelitis during the course of the disease. Forty percent of the patients had at least one MRI brain abnormality on T2-FLAIR images, but only 5% fulfilled MRI criteria for multiple sclerosis. Fourteen patients (56%) were positive for NMO IgG. All patients were tested at least 3 months after a clinical relapse and none of them except two received oral corticosteroids. Fourteen patients were treated with immunosuppressive agents (azathioprine [n = 2], mycophenolate mofetil [n = 6], mitoxantrone [n = 2], cyclophosphamide [n = 4]). The mean EDSS score was 4.2 ± 1.8 after a mean follow-up of 6 ± 1.5 years. The control group included 20 healthy volunteers (13 women
J. Jeantroux et al. and 7 men; mean age 42 ± 14.8 years) with normal brain MRI on T2-FLAIR and without any history of neurological disorders. All patients and healthy subjects gave their written informed consent to participate to the study. The study was approved by the local Ethics Committee.
Magnetic resonance image acquisition Brain and spinal cord MRI scans were obtained at the same time from all the patients and controls using a 1.5-T MRI scanner (SIEMENS Avanto MR, Erlangen, Germany). A birdcage (12 channel) head coil and a tailored spine phased array coil were used for signal reception. For the brain, the following sequences were acquired: • three-dimensional T1 magnetization-prepared rapid acquisition gradient echo (MP-RAGE) (repetition time [TR] = 1900 ms, echo time [TE] = 2.68 ms, inversion time [TI] = 1100 ms, flip angle = 15 degrees, FOV 320*256, matrix 320*256, slice thickness 1.0 mm, number of partitions = 160); • axial turbo spin-echo (TSE) T2-weighted sequence (TR = 4000, TE = 14/109, echo train length = 5, FOV 220*165 mm2 , matrix size 256*192, slice thickness 4.0 mm); • pulsed-gradient spin-echo echoplanar pulse sequence (inter-echo spacing = 0.73, TR = 6800, TE = 99), with diffusion gradients applied in 30 non-colinear directions, chosen to cover three-dimensional space uniformly, with a maximum b factor in each direction of 1000 s/mm2 . To optimize the measurement of diffusion, only two b factors were used (b1 = 0, b2 = 1000 s/mm2 ). For the spinal cord the following sequences were acquired: • T2-weighted turbo spin-echo (TR = 4000 ms, TE = 111 ms, echo train length = 15, FOV = 420 × 420 mm, matrix size = 512 × 389, number of signal averages = 3, 13 contiguous sagittal-oblique, 3-mm slice thickness and an interslice gap of 0.3 mm); • pulsed-gradient spin-echo echoplanar pulse sequence (inter-echo spacing = 0.78, TR = 2700, TE = 71, FOV 230 × 230 mm2 , matrix 104 × 104, 30 contiguous axial slices), 5-mm slice thickness, with diffusion gradients applied in 12 non-colinear directions, chosen to cover three-dimensional space uniformly, with a maximum b factor in each direction of 800 s/mm2 . To optimize the measurement of diffusion, only two b factors were used (b1 = 0, b2 = 800 s/mm2 ).
Image analysis and postprocessing Image analysis and postprocessing were performed on a computer workstation (SIEMENS Syngo). The diffusion tensor (DT) MRI data were reconstructed offline and DT was calculated for each voxel. From the tensor matrix, ADC and FA maps were derived. DTI analysis: regions of interest (ROIs) were successively placed in the corpus callosum (CC) in the midsagittal plane of reconstructed FA images, in the optic radiations in the
Diffusion tensor imaging of normal-appearing white matter in neuromyelitis optica
297
coronal plane of reconstructed FA images at the level of the posterior border of the CC, and in both anterior and posterior limbs of the internal capsule (IC) in the axial plane of FA images (Fig. 1). The lesions and the NAWM areas in the spinal cord were identified on the T2-weighted sagittal images and the ROIs were placed on the axial FA images of the corresponding levels (Fig. 2), following a method detailed previously [25]. For each ROI, the ADC and FA were measured. Using the same method, the ADC and FA in the same regions were measured in the controls. All ROIs were measured by the same observer (X.Z.L.), blinded to the clinical data.
Statistical analysis Statistical analyses were performed using SPSS 11.5 for Windows. A two-tailed Student’s t-test for non-paired data was used to compare ADC and FA values in the two groups. Data are given as mean value ± standard deviation. P < 0.05 were considered statistically significant.
Results Brain scans Table 1 reports DTI-derived measurements of the brain white matter in NMO patients and healthy controls. In the
Table 1 Apparent diffusivity coefficient and fractional anisotropy measurements in brain regions in neuromyelitis optica patients and controls.
Right optic radiation ADC FA Left optic radiation ADC FA Right anterior IC ADC FA Left anterior IC ADC FA Right posterior IC ADC FA Left posterior IC ADC FA Corpus callosum ADC FA
NMO patients (n = 25)
Controls (n = 20)
0.823 ± 0.151** 0.450 ± 0.052*
0.781 ± 0.042 0.480 ± 0.033
0.819 ± 0.139** 0.448 ± 0.049*
0.779 ± 0.039 0.478 ± 0.031
0.710 ± 0.033 0.667 ± 0.051
0.708 ± 0.029 0.663 ± 0.030
0.711 ± 0.039 0.673 ± 0.039
0.706 ± 0.021 0.670 ± 0.034
0.749 ± 0.044* 0.554 ± 0.062**
0.721 ± 0.039 0.609 ± 0.056
0.752 ± 0.056** 0.590 ± 0.073
0.714 ± 0.041 0.629 ± 0.064
0.891 ± 0.108 0.676 ± 0.059
0.880 ± 0.056 0.677 ± 0.047
ADC: apparent diffusivity coefficient; FA: fractional anisotropy; CC: corpus callosum; NMO: neuromyelitis optica; IC: internal capsule; NS: not significant. Values are expressed as mean ± standard deviation; * P < 0.05; ** P < 0.01.
Figure 1 Localization of regions of interest (ROIs) in normalappearing while matter on T1/fractional anisotropy superposed images. ROIs were located (a) in the coronal plane through the posterior margin of the corpus callosum (CC) for the optic radiation (b) in the midsagittal plane for the CC, (c) in the axial plane for the anterior and posterior limbs of the IC.
298
J. Jeantroux et al.
Figure 2 Example of regions of interest (ROI) placement in spinal cord lesion: a: sagittal T2 weighted shows extensive myelitis in neuromyelitis optica patient; b: ROI is placed on axial fractional anisotropy image.
posterior limb of the IC, NMO patients had a higher mean ADC (P < 0.05) than healthy controls and a lower FA (P < 0.05). NMO patients also had a higher mean ADC (P < 0.01) in the optic radiations than healthy controls and a decreased FA (P < 0.05). In the anterior limb of the IC and in the CC, we did not observe any difference in mean ADC or FA between NMO patients and controls.
Spinal cord scans Table 2 reports DTI-derived measurements of spinal cord in NMO patients and controls. In the spinal cord, the mean ADC of NAWM in NMO patients was higher (P < 0.05) and the mean FA was lower (P < 0.05) than in control subjects. The mean ADC of lesions in NMO patients was higher (P < 0.001) and
Table 2 Apparent diffusivity coefficient and fractional anisotropy measurements in normal appearing white matter and spinal cord lesions in neuromyelitis optica patients and control subjects. NMO patients (n = 25)
Controls (n = 20)
Spinal cord NAWM ADC FA
1.110 ± 0.119* 0.579 ± 0.054*
1.033 ± 0.076 0.611 ± 0.042
Spinal cord lesions ADC FA
1.288 ± 0.1930** 0.479 ± 0.078**
ADC: apparent diffusivity coefficient; FA: fractional anisotropy; NAWM: normal appearing white matter; NMO: neuromyelitis optica. Values are expressed as mean ± standard deviation; * P < 0.05; ** P < 0.001 compared with NAWM of controls.
the mean FA was lower (P < 0.001) than the corresponding values for NAWM in the spinal cord of control subjects.
Discussion This study shows abnormal DTI findings in optic radiations and cortico-spinal tracts (including spinal cord and posterior limb of the IC) in NMO patients when compared with healthy subjects. In contrast, NAWM outside these regions of particular interest in NMO (i.e. the CC and the anterior limb of the IC) appeared to be spared. These findings are clearly different from those in MS, where the CC is frequently abnormal in lesions and in NAWM. Furthermore, DTI abnormalities are very significant in spinal cord lesions, with a high increase of ADC and a substantial decrease of FA in NMO patients. The ADC and FA are rotationally invariant indices widely used in brain disease studies and in monitoring patients’ treatment [26—28]. The ADC measures the average diffusivity of water molecules. It is therefore affected by cell size and integrity. The FA measures the degree of directionality of diffusion and reflects the structural integrity and degree of structural alignment within fibre tracts. These two indexes provide different but complementary information about water molecular diffusion motion at different views and can give information on the size, shape, orientation and geometry of brain tissues [18]. Recent investigations in patients with NMO revealed occult damage in NAGM and NAWM on magnetization transfer imaging but not on MR spectroscopy or DTI [9,10,12]. In our study, we compared the ADC and FA of NAWM, including the CC and anterior limb of the IC, which are not directly connected to the spinal cord. We also investigated the ADC and FA in the optic radiations and in the posterior limb of IC, which are both connected with anatomical regions impaired in NMO. We found that patients with NMO had a higher mean ADC and a lower FA in NAWM of the posterior limb of the IC, of the optic radiations, and of the spinal cord compared
Diffusion tensor imaging of normal-appearing white matter in neuromyelitis optica with control subjects, while none of the ADC and FA values for the CC or the anterior limb of the IC showed any significant difference. Our results are in accordance with those of previous studies [11,22,23]. These MRI modifications were also more marked in spinal cord lesions. Our results suggest that, in NMO, MRI abnormalities outside visible T2-weighted lesions are limited to the lesions and the regions with direct connection to the spinal cord and optic nerves and argue against the presence of lesions beyond the spatial resolution of conventional MR imaging in NAWM. We can speculate on the possible pathological substrates of these findings. There is evidence that a reduction in FA is consistent with either axonal fibre degeneration and myelin breakdown or myelin loss [26—31]. ADC measurements are affected by cellular size, shape, orientation and integrity; FA is a measurement of anisotropy and reflects the degree of alignment of cellular structures within fibre tracts and their structural integrity [14,18]. The different pathological elements in NMO can alter the permeability or the geometry of structural barriers to water diffusion in both the brain and the spinal cord. Consequently, DTI might provide quantitative information on the structural damage occurring in both NMO lesions and NAWM that complements the information provided by other MR techniques [32], thereby increasing our understanding of the pathophysiology of NMO. Indeed, NMO, which spares regions outside the cortico-spinal and optic tracts, appears to be a less diffuse disease than MS. Although our study has produced interesting data on brain and spinal cord tissue damage in patients with NMO and indicated a possible mechanism, radiological and pathological correlation studies will be needed to clarify the exact relationship between changes in the DTI-derived measurements and the histopathological processes. A longitudinal study should also be performed to elucidate the dynamics of lesional processes in NMO.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
Acknowledgements The authors thank Pascale Anstett and Pierre-Emmanuel Zorn for technical assistance with the MRI scans and Baradi Brice for the management of the subjects.
References [1] Mandler RN, Davis LE, Jeffery DR, et al. Devic’s neuromyelitis optica: a clinicopathological study of 8 patients. Ann Neurol 1993;34:162—8. [2] Wingerchuk DM, Hogancamp WF, O’Brien PC, et al. The clinical course of neuromyelitis optica (Devic’s syndrome). Neurology 1999;53:1107—14. [3] Filippi M, Rocca MA, Moiola L, et al. MRI and magnetization transfer imaging changes in the brain and cervical cord of patients with Devic’s neuromyelitis optica. Neurology 1999;53:1705—10. [4] Pittock SJ, Lennon VA, Krecke K, et al. Brain abnormalities in neuromyelitis optica. Arch Neurol 2006;63:390—6.
299
[5] Pittock SJ, Weinshenker BG, Lucchinetti CF, et al. Neuromyelitis optica brain lesions localized at sites of high aquaporin 4 expression. Arch Neurol 2006;63:964—8. [6] Guimaraes J, Sa MJ. Devic disease with abnormal brain magnetic resonance image findings: the first Portuguese case. Arch Neurol 2007;64:290—1. [7] Nakamura M, Endo M, Murakami K, et al. An autopsied case of neuromyelitis optica with a large cavitary cerebral lesion. Mult Scler 2005;11:735—8. [8] Jacobs D, Roemer S, Weinshenker B, et al. The pathology of brain involvement in neuromyelitis optica spectrum disorder. Mult Scler 2006;12:S155. [9] Rocca MA, Agosta F, Mezzapesa DM, et al. Magnetization transfer and diffusion tensor MRI show gray matter damage in neuromyelitis optica. Neurology 2004;62:476—8. [10] Rocca MA, Agosta F, Mezzapesa DM, et al. A functional MRI study of movement-associated cortical changes in patients with Devic’s neuromyelitis optica. NeuroImage 2004;21: 1061—8. [11] Yu CS, Lin FC, Li KC, et al. Diffusion tensor imaging in the assessment of normal-appearing brain tissue damage in relapsing neuromyelitis optica. Am J Neuroradiol 2006;27:1009— 15. [12] De Seze J, Blanc F, Kremer S, Collongues N, Fleury M, Marcel C, et al. Magnetic resonance spectroscopy evaluation in patients with neuromyelitis optica. J Neurol Neurosurg Psychiatry 2010;81(4):409—11. [13] Basser PJ, Mattiello J, LeBihan D, et al. MR diffusion tensor spectroscopy and imaging. Biophys J 1994;66:259—67. [14] Basser PJ, Mattiello J, Le Bihan D. Estimation of the effective self-diffusion tensor from the NMR spin-echo. J Magn Reson 1994;103:247—54. [15] Eriksson SH, Rugg-Gunn FJ, Symms MR, et al. Diffusion tensor imaging in patients with epilepsy and malformations of cortical development. Brain 2001;124:617—26. [16] Le Bihan D. Looking into the functional architecture of the brain with diffusion MRI. Nat Rev Neurosci 2003;4:469—80. [17] Rovaris M, Gass A, Bammer R, et al. Diffusion MRI in multiple sclerosis. Neurology 2005;65:1526—32. [18] Basser PJ, Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative diffusion tensor MRI. J Magn Reson 1996;111:209—19. [19] Pierpaoli C, Jezzard P, Basser PJ, et al. Diffusion tensor MR imaging of the human brain. Radiology 1996;201:637—48. [20] Sevick RJ, Kanda F, Mintorovitch J, et al. Cytotoxic brain edema: assessment with diffusion- weighted MR imaging. Radiology 1992;185:687—90. [21] Anderson AW, Zhong J, Petroff OA, et al. Effects of osmotically driven cell volume changes on diffusion-weighted imaging of the rat optic nerve. Magn Reson Med 1996;35:162—7. [22] Yu CS, Zhu CZ, Li KC, et al. Relapsing neuromyelitis optica and relapsing-remitting multiple sclerosis: differentiation at diffusion-tensor MR imaging of corpus callosum. Radiology 2007;244:249—56. [23] Yu CS, Lin FC, Li KC, et al. Pathogenesis of normal-appearing white matter damage in neuromyelitis optica: diffusion-tensor MR imaging. Radiology 2008;246:222—8. [24] Wingerchuk DM, Lennon VA, Pittock SJ, et al. Revised diagnostic criteria for neuromyelitis optica. Neurology 2006;66:1485—9. [25] Marcel C, Kremer S, Jeantroux J, Blanc F, Dietemann JL, de Seze J. Diffusion-weighted imaging in noncompressive myelopathies: a 33-patient prospective study. J Neurol 2010;257:1438—45. [26] Filippi M, Cercignani M, Inglese M, et al. Diffusion tensor magnetic resonance imaging in multiple sclerosis. Neurology 2001;56:304—11. [27] Sotak CH. The role of diffusion tensor imaging in the evaluation of ischemic brain injury — a review. NMR Biomed 2002;15:561—9.
300 [28] Sundgren PC, Dong Q, Gomez-Hassan D, et al. Diffusion tensor imaging of the brain: review of clinical applications. Neuroradiology 2004;46:339—50. [29] Beaulieu C, Does MD, Snyder RE, et al. Changes in water diffusion due to Wallerian degeneration in peripheral nerve. Magn Reson Med 1996;36:627—31. [30] Pierpaoli C, Barnett A, Pajevic S, et al. Water diffusion changes in Wallerian degeneration and their dependence on white matter architecture. NeuroImage 2001;13:1174—85.
J. Jeantroux et al. [31] Schmierer K, Wheeler-Kingshott CA, Boulby PA, et al. Diffusion tensor imaging of post-mortem multiple sclerosis brain. NeuroImage 2007;35:467—77. [32] van Waesberghe JH, Kamphorst W, De Groot CJ, et al. Axonal loss in multiple sclerosis lesions: magnetic resonance imaging insights into substrates of disability. Ann Neurol 1999;46:747—54.
Available online at
www.sciencedirect.com
ORIGINAL ARTICLE
Diffusion tensor imaging of normal-appearing white matter in neuromyelitis optica Imagerie en tenseur de diffusion de la substance blanche d’allure normale dans la neuromyélite optique J. Jeantroux a,∗, S. Kremer a, X.Z. Lin a, N. Collongues b, J.-B. Chanson b, B. Bourre b, M. Fleury b, F. Blanc b, J.-L. Dietemann a, J. de Seze b a b
Department of radiology 2, hôpital de Hautepierre, university hospitals of Strasbourg, avenue Molière, 67000 Strasbourg, France Department of neurology, university hospitals of Strasbourg, avenue Molière, 67000 Strasbourg, France
KEYWORDS Neuromyelitis optica; MRI; Diffusion tensor imaging
Summary Objectives: Neuromyelitis optica (NMO) is an inflammatory disease of the central nervous system characterized by severe attacks of optic neuritis and myelitis. Brain was classically, unlike in multiple sclerosis (MS), spared. Nevertheless recent studies showed that brain lesions can be seen with MRI. We studied the diffusion characteristics of normal-appearing white matter (NAWM) and abnormal white matter in NMO patients compared with NAWM in healthy subjects. Patients and methods: Diffusion tensor imaging (DTI) scans of the brain and spinal cord were obtained from 25 patients with NMO and 20 age- and gender-matched healthy subjects. Region of interest (ROI) analysis of the apparent diffusivity coefficient (ADC) and fractional anisotropy (FA) was performed in brain NAWM (optic radiations, corpus callosum [CC] and anterior and posterior limbs of the internal capsule [IC]) and in spinal cord NAWM and in lesions. Results: ADC was increased and FA decreased in NMO patients in the posterior limb of the IC in the optic radiations and in spinal cord NAWM. FA was lower in spinal cord lesions. In contrast, there was no difference between the two groups in the anterior limb of the IC nor in the CC. Conclusion: These results suggest that DTI abnormalities are very severe in NMO spinal cord lesions. In our study, DTI abnormalities in NAWM were restricted to optic radiations and corticospinal tracts, suggesting secondary Wallerian degeneration. In contrast, NAWM outside these tracts (CC and anterior IC) remained normal suggesting that, unlike what is observed in MS, there is no infra-lesional abnormality in NMO. © 2011 Elsevier Masson SAS. All rights reserved.
∗ Corresponding author. Tel.: +33 3 88 12 78 90; fax: +33 3 88 12 71 18. E-mail address: [email protected] (J. Jeantroux).
Neuromyelitis optica (NMO) is an inflammatory disease of the CNS characterized by severe attacks of optic neuritis and myelitis, which, unlike multiple sclerosis (MS), commonly spares the brain in the early stages. It has become increasingly apparent, however, that both
0150-9861/$ – see front matter © 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.neurad.2011.10.003
296 symptomatic and asymptomatic MRI brain lesions may occur in NMO [1—6]. Postmortem studies have confirmed that the rare brain lesions observed on MRI in NMO patients have the same immunohistochemical characteristics as spinal cord lesions [4,7,8]. A few studies have reported that patients with NMO appear to have little damage in either normalappearing grey matter (NAGM) or normal-appearing white matter (NAWM) and others showed a lack of abnormalities, for example using MR spectroscopy [9—12]. Diffusion tensor imaging (DTI) helps to measure the random motion of water molecules and provides information on the structural and orientational features of tissues at a microscopic level [13—17]. Some rotationally invariant indices, such as the apparent diffusivity coefficient (ADC) and fractional anisotropy (FA), which are derived from DTI, can provide information on the magnitude and directionality of water diffusion in brain tissues. ADC reflects average diffusivity of water molecular motion, independently of any tissue directionality and is affected by cellular size and integrity. FA measures the degree of directionality of water diffusion and thus reflects the degree of alignment of cellular structures within fibre tracts, as well as their structural integrity [18,19]. Within a coherently arranged white matter tract, water molecules diffuse faster in the direction parallel to the tract than in perpendicular directions. Pathological processes, by changing the microstructural environment, result in altered diffusion [20,21]. DTI seems to be a promising tool for the quantification of tissue damage and for improving our understanding of conditions that affect the integrity and organization of brain tissues. Up to now, there have only been a few published studies on DTI in NMO, all performed by the same team [11,22,23]. The aim of the present study was to evaluate DTI abnormalities in brain NAWM, optic radiations and spinal cord of NMO patients in order to increase our knowledge of the pathophysiological mechanisms in this rare pathology.
Patients and methods Subjects From April 2007 to December 2008, we prospectively included 25 patients (16 women and nine men; mean age 43.7 ± 12.3 years) with definite NMO (mean disease duration 6.1 ± 2.5 years), diagnosed according to the revised criteria proposed by Wingerchuck et al. [24]. All patients but two had a relapsing form of NMO. The two remaining patients had a monophasic course with myelitis and optic neuritis at the same time. All patients had at least one episode of transverse myelitis during the course of the disease. Forty percent of the patients had at least one MRI brain abnormality on T2-FLAIR images, but only 5% fulfilled MRI criteria for multiple sclerosis. Fourteen patients (56%) were positive for NMO IgG. All patients were tested at least 3 months after a clinical relapse and none of them except two received oral corticosteroids. Fourteen patients were treated with immunosuppressive agents (azathioprine [n = 2], mycophenolate mofetil [n = 6], mitoxantrone [n = 2], cyclophosphamide [n = 4]). The mean EDSS score was 4.2 ± 1.8 after a mean follow-up of 6 ± 1.5 years. The control group included 20 healthy volunteers (13 women
J. Jeantroux et al. and 7 men; mean age 42 ± 14.8 years) with normal brain MRI on T2-FLAIR and without any history of neurological disorders. All patients and healthy subjects gave their written informed consent to participate to the study. The study was approved by the local Ethics Committee.
Magnetic resonance image acquisition Brain and spinal cord MRI scans were obtained at the same time from all the patients and controls using a 1.5-T MRI scanner (SIEMENS Avanto MR, Erlangen, Germany). A birdcage (12 channel) head coil and a tailored spine phased array coil were used for signal reception. For the brain, the following sequences were acquired: • three-dimensional T1 magnetization-prepared rapid acquisition gradient echo (MP-RAGE) (repetition time [TR] = 1900 ms, echo time [TE] = 2.68 ms, inversion time [TI] = 1100 ms, flip angle = 15 degrees, FOV 320*256, matrix 320*256, slice thickness 1.0 mm, number of partitions = 160); • axial turbo spin-echo (TSE) T2-weighted sequence (TR = 4000, TE = 14/109, echo train length = 5, FOV 220*165 mm2 , matrix size 256*192, slice thickness 4.0 mm); • pulsed-gradient spin-echo echoplanar pulse sequence (inter-echo spacing = 0.73, TR = 6800, TE = 99), with diffusion gradients applied in 30 non-colinear directions, chosen to cover three-dimensional space uniformly, with a maximum b factor in each direction of 1000 s/mm2 . To optimize the measurement of diffusion, only two b factors were used (b1 = 0, b2 = 1000 s/mm2 ). For the spinal cord the following sequences were acquired: • T2-weighted turbo spin-echo (TR = 4000 ms, TE = 111 ms, echo train length = 15, FOV = 420 × 420 mm, matrix size = 512 × 389, number of signal averages = 3, 13 contiguous sagittal-oblique, 3-mm slice thickness and an interslice gap of 0.3 mm); • pulsed-gradient spin-echo echoplanar pulse sequence (inter-echo spacing = 0.78, TR = 2700, TE = 71, FOV 230 × 230 mm2 , matrix 104 × 104, 30 contiguous axial slices), 5-mm slice thickness, with diffusion gradients applied in 12 non-colinear directions, chosen to cover three-dimensional space uniformly, with a maximum b factor in each direction of 800 s/mm2 . To optimize the measurement of diffusion, only two b factors were used (b1 = 0, b2 = 800 s/mm2 ).
Image analysis and postprocessing Image analysis and postprocessing were performed on a computer workstation (SIEMENS Syngo). The diffusion tensor (DT) MRI data were reconstructed offline and DT was calculated for each voxel. From the tensor matrix, ADC and FA maps were derived. DTI analysis: regions of interest (ROIs) were successively placed in the corpus callosum (CC) in the midsagittal plane of reconstructed FA images, in the optic radiations in the
Diffusion tensor imaging of normal-appearing white matter in neuromyelitis optica
297
coronal plane of reconstructed FA images at the level of the posterior border of the CC, and in both anterior and posterior limbs of the internal capsule (IC) in the axial plane of FA images (Fig. 1). The lesions and the NAWM areas in the spinal cord were identified on the T2-weighted sagittal images and the ROIs were placed on the axial FA images of the corresponding levels (Fig. 2), following a method detailed previously [25]. For each ROI, the ADC and FA were measured. Using the same method, the ADC and FA in the same regions were measured in the controls. All ROIs were measured by the same observer (X.Z.L.), blinded to the clinical data.
Statistical analysis Statistical analyses were performed using SPSS 11.5 for Windows. A two-tailed Student’s t-test for non-paired data was used to compare ADC and FA values in the two groups. Data are given as mean value ± standard deviation. P < 0.05 were considered statistically significant.
Results Brain scans Table 1 reports DTI-derived measurements of the brain white matter in NMO patients and healthy controls. In the
Table 1 Apparent diffusivity coefficient and fractional anisotropy measurements in brain regions in neuromyelitis optica patients and controls.
Right optic radiation ADC FA Left optic radiation ADC FA Right anterior IC ADC FA Left anterior IC ADC FA Right posterior IC ADC FA Left posterior IC ADC FA Corpus callosum ADC FA
NMO patients (n = 25)
Controls (n = 20)
0.823 ± 0.151** 0.450 ± 0.052*
0.781 ± 0.042 0.480 ± 0.033
0.819 ± 0.139** 0.448 ± 0.049*
0.779 ± 0.039 0.478 ± 0.031
0.710 ± 0.033 0.667 ± 0.051
0.708 ± 0.029 0.663 ± 0.030
0.711 ± 0.039 0.673 ± 0.039
0.706 ± 0.021 0.670 ± 0.034
0.749 ± 0.044* 0.554 ± 0.062**
0.721 ± 0.039 0.609 ± 0.056
0.752 ± 0.056** 0.590 ± 0.073
0.714 ± 0.041 0.629 ± 0.064
0.891 ± 0.108 0.676 ± 0.059
0.880 ± 0.056 0.677 ± 0.047
ADC: apparent diffusivity coefficient; FA: fractional anisotropy; CC: corpus callosum; NMO: neuromyelitis optica; IC: internal capsule; NS: not significant. Values are expressed as mean ± standard deviation; * P < 0.05; ** P < 0.01.
Figure 1 Localization of regions of interest (ROIs) in normalappearing while matter on T1/fractional anisotropy superposed images. ROIs were located (a) in the coronal plane through the posterior margin of the corpus callosum (CC) for the optic radiation (b) in the midsagittal plane for the CC, (c) in the axial plane for the anterior and posterior limbs of the IC.
298
J. Jeantroux et al.
Figure 2 Example of regions of interest (ROI) placement in spinal cord lesion: a: sagittal T2 weighted shows extensive myelitis in neuromyelitis optica patient; b: ROI is placed on axial fractional anisotropy image.
posterior limb of the IC, NMO patients had a higher mean ADC (P < 0.05) than healthy controls and a lower FA (P < 0.05). NMO patients also had a higher mean ADC (P < 0.01) in the optic radiations than healthy controls and a decreased FA (P < 0.05). In the anterior limb of the IC and in the CC, we did not observe any difference in mean ADC or FA between NMO patients and controls.
Spinal cord scans Table 2 reports DTI-derived measurements of spinal cord in NMO patients and controls. In the spinal cord, the mean ADC of NAWM in NMO patients was higher (P < 0.05) and the mean FA was lower (P < 0.05) than in control subjects. The mean ADC of lesions in NMO patients was higher (P < 0.001) and
Table 2 Apparent diffusivity coefficient and fractional anisotropy measurements in normal appearing white matter and spinal cord lesions in neuromyelitis optica patients and control subjects. NMO patients (n = 25)
Controls (n = 20)
Spinal cord NAWM ADC FA
1.110 ± 0.119* 0.579 ± 0.054*
1.033 ± 0.076 0.611 ± 0.042
Spinal cord lesions ADC FA
1.288 ± 0.1930** 0.479 ± 0.078**
ADC: apparent diffusivity coefficient; FA: fractional anisotropy; NAWM: normal appearing white matter; NMO: neuromyelitis optica. Values are expressed as mean ± standard deviation; * P < 0.05; ** P < 0.001 compared with NAWM of controls.
the mean FA was lower (P < 0.001) than the corresponding values for NAWM in the spinal cord of control subjects.
Discussion This study shows abnormal DTI findings in optic radiations and cortico-spinal tracts (including spinal cord and posterior limb of the IC) in NMO patients when compared with healthy subjects. In contrast, NAWM outside these regions of particular interest in NMO (i.e. the CC and the anterior limb of the IC) appeared to be spared. These findings are clearly different from those in MS, where the CC is frequently abnormal in lesions and in NAWM. Furthermore, DTI abnormalities are very significant in spinal cord lesions, with a high increase of ADC and a substantial decrease of FA in NMO patients. The ADC and FA are rotationally invariant indices widely used in brain disease studies and in monitoring patients’ treatment [26—28]. The ADC measures the average diffusivity of water molecules. It is therefore affected by cell size and integrity. The FA measures the degree of directionality of diffusion and reflects the structural integrity and degree of structural alignment within fibre tracts. These two indexes provide different but complementary information about water molecular diffusion motion at different views and can give information on the size, shape, orientation and geometry of brain tissues [18]. Recent investigations in patients with NMO revealed occult damage in NAGM and NAWM on magnetization transfer imaging but not on MR spectroscopy or DTI [9,10,12]. In our study, we compared the ADC and FA of NAWM, including the CC and anterior limb of the IC, which are not directly connected to the spinal cord. We also investigated the ADC and FA in the optic radiations and in the posterior limb of IC, which are both connected with anatomical regions impaired in NMO. We found that patients with NMO had a higher mean ADC and a lower FA in NAWM of the posterior limb of the IC, of the optic radiations, and of the spinal cord compared
Diffusion tensor imaging of normal-appearing white matter in neuromyelitis optica with control subjects, while none of the ADC and FA values for the CC or the anterior limb of the IC showed any significant difference. Our results are in accordance with those of previous studies [11,22,23]. These MRI modifications were also more marked in spinal cord lesions. Our results suggest that, in NMO, MRI abnormalities outside visible T2-weighted lesions are limited to the lesions and the regions with direct connection to the spinal cord and optic nerves and argue against the presence of lesions beyond the spatial resolution of conventional MR imaging in NAWM. We can speculate on the possible pathological substrates of these findings. There is evidence that a reduction in FA is consistent with either axonal fibre degeneration and myelin breakdown or myelin loss [26—31]. ADC measurements are affected by cellular size, shape, orientation and integrity; FA is a measurement of anisotropy and reflects the degree of alignment of cellular structures within fibre tracts and their structural integrity [14,18]. The different pathological elements in NMO can alter the permeability or the geometry of structural barriers to water diffusion in both the brain and the spinal cord. Consequently, DTI might provide quantitative information on the structural damage occurring in both NMO lesions and NAWM that complements the information provided by other MR techniques [32], thereby increasing our understanding of the pathophysiology of NMO. Indeed, NMO, which spares regions outside the cortico-spinal and optic tracts, appears to be a less diffuse disease than MS. Although our study has produced interesting data on brain and spinal cord tissue damage in patients with NMO and indicated a possible mechanism, radiological and pathological correlation studies will be needed to clarify the exact relationship between changes in the DTI-derived measurements and the histopathological processes. A longitudinal study should also be performed to elucidate the dynamics of lesional processes in NMO.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
Acknowledgements The authors thank Pascale Anstett and Pierre-Emmanuel Zorn for technical assistance with the MRI scans and Baradi Brice for the management of the subjects.
References [1] Mandler RN, Davis LE, Jeffery DR, et al. Devic’s neuromyelitis optica: a clinicopathological study of 8 patients. Ann Neurol 1993;34:162—8. [2] Wingerchuk DM, Hogancamp WF, O’Brien PC, et al. The clinical course of neuromyelitis optica (Devic’s syndrome). Neurology 1999;53:1107—14. [3] Filippi M, Rocca MA, Moiola L, et al. MRI and magnetization transfer imaging changes in the brain and cervical cord of patients with Devic’s neuromyelitis optica. Neurology 1999;53:1705—10. [4] Pittock SJ, Lennon VA, Krecke K, et al. Brain abnormalities in neuromyelitis optica. Arch Neurol 2006;63:390—6.
299
[5] Pittock SJ, Weinshenker BG, Lucchinetti CF, et al. Neuromyelitis optica brain lesions localized at sites of high aquaporin 4 expression. Arch Neurol 2006;63:964—8. [6] Guimaraes J, Sa MJ. Devic disease with abnormal brain magnetic resonance image findings: the first Portuguese case. Arch Neurol 2007;64:290—1. [7] Nakamura M, Endo M, Murakami K, et al. An autopsied case of neuromyelitis optica with a large cavitary cerebral lesion. Mult Scler 2005;11:735—8. [8] Jacobs D, Roemer S, Weinshenker B, et al. The pathology of brain involvement in neuromyelitis optica spectrum disorder. Mult Scler 2006;12:S155. [9] Rocca MA, Agosta F, Mezzapesa DM, et al. Magnetization transfer and diffusion tensor MRI show gray matter damage in neuromyelitis optica. Neurology 2004;62:476—8. [10] Rocca MA, Agosta F, Mezzapesa DM, et al. A functional MRI study of movement-associated cortical changes in patients with Devic’s neuromyelitis optica. NeuroImage 2004;21: 1061—8. [11] Yu CS, Lin FC, Li KC, et al. Diffusion tensor imaging in the assessment of normal-appearing brain tissue damage in relapsing neuromyelitis optica. Am J Neuroradiol 2006;27:1009— 15. [12] De Seze J, Blanc F, Kremer S, Collongues N, Fleury M, Marcel C, et al. Magnetic resonance spectroscopy evaluation in patients with neuromyelitis optica. J Neurol Neurosurg Psychiatry 2010;81(4):409—11. [13] Basser PJ, Mattiello J, LeBihan D, et al. MR diffusion tensor spectroscopy and imaging. Biophys J 1994;66:259—67. [14] Basser PJ, Mattiello J, Le Bihan D. Estimation of the effective self-diffusion tensor from the NMR spin-echo. J Magn Reson 1994;103:247—54. [15] Eriksson SH, Rugg-Gunn FJ, Symms MR, et al. Diffusion tensor imaging in patients with epilepsy and malformations of cortical development. Brain 2001;124:617—26. [16] Le Bihan D. Looking into the functional architecture of the brain with diffusion MRI. Nat Rev Neurosci 2003;4:469—80. [17] Rovaris M, Gass A, Bammer R, et al. Diffusion MRI in multiple sclerosis. Neurology 2005;65:1526—32. [18] Basser PJ, Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative diffusion tensor MRI. J Magn Reson 1996;111:209—19. [19] Pierpaoli C, Jezzard P, Basser PJ, et al. Diffusion tensor MR imaging of the human brain. Radiology 1996;201:637—48. [20] Sevick RJ, Kanda F, Mintorovitch J, et al. Cytotoxic brain edema: assessment with diffusion- weighted MR imaging. Radiology 1992;185:687—90. [21] Anderson AW, Zhong J, Petroff OA, et al. Effects of osmotically driven cell volume changes on diffusion-weighted imaging of the rat optic nerve. Magn Reson Med 1996;35:162—7. [22] Yu CS, Zhu CZ, Li KC, et al. Relapsing neuromyelitis optica and relapsing-remitting multiple sclerosis: differentiation at diffusion-tensor MR imaging of corpus callosum. Radiology 2007;244:249—56. [23] Yu CS, Lin FC, Li KC, et al. Pathogenesis of normal-appearing white matter damage in neuromyelitis optica: diffusion-tensor MR imaging. Radiology 2008;246:222—8. [24] Wingerchuk DM, Lennon VA, Pittock SJ, et al. Revised diagnostic criteria for neuromyelitis optica. Neurology 2006;66:1485—9. [25] Marcel C, Kremer S, Jeantroux J, Blanc F, Dietemann JL, de Seze J. Diffusion-weighted imaging in noncompressive myelopathies: a 33-patient prospective study. J Neurol 2010;257:1438—45. [26] Filippi M, Cercignani M, Inglese M, et al. Diffusion tensor magnetic resonance imaging in multiple sclerosis. Neurology 2001;56:304—11. [27] Sotak CH. The role of diffusion tensor imaging in the evaluation of ischemic brain injury — a review. NMR Biomed 2002;15:561—9.
300 [28] Sundgren PC, Dong Q, Gomez-Hassan D, et al. Diffusion tensor imaging of the brain: review of clinical applications. Neuroradiology 2004;46:339—50. [29] Beaulieu C, Does MD, Snyder RE, et al. Changes in water diffusion due to Wallerian degeneration in peripheral nerve. Magn Reson Med 1996;36:627—31. [30] Pierpaoli C, Barnett A, Pajevic S, et al. Water diffusion changes in Wallerian degeneration and their dependence on white matter architecture. NeuroImage 2001;13:1174—85.
J. Jeantroux et al. [31] Schmierer K, Wheeler-Kingshott CA, Boulby PA, et al. Diffusion tensor imaging of post-mortem multiple sclerosis brain. NeuroImage 2007;35:467—77. [32] van Waesberghe JH, Kamphorst W, De Groot CJ, et al. Axonal loss in multiple sclerosis lesions: magnetic resonance imaging insights into substrates of disability. Ann Neurol 1999;46:747—54.