- Email: [email protected]
A functional MRI study of movement-associated cortical changes in patients with Devic's neuromyelitis optica
www.elsevier.com/locate/ynimg NeuroImage 21 (2004) 1061 – 1068
A functional MRI study of movement-associated cortical changes in patients with Devic’s neuromyelitis optica M.A. Rocca, a,b F. Agosta, a,b D.M. Mezzapesa, a,b A. Falini, c V. Martinelli, b F. Salvi, d R. Bergamaschi, e G. Scotti, c G. Comi, b and M. Filippi a,b,* a
Neuroimaging Research Unit, Scientific Institute and University Ospedale San Raffaele, Milan, Italy Department of Neurology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy c Department of Neuroradiology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy d Department of Clinical Neurology, Ospedale Bellaria, University of Bologna, Bologna, Italy e Department of Neurology, Istituto Mondino, Pavia, Italy b
Received 22 July 2003; revised 7 October 2003; accepted 7 October 2003
Movement-associated cortical changes have been shown in several neurological conditions and were found to be associated to the extent of brain and cord damage. Devic’s neuromyelitis optica (DNO) is characterized by a severe involvement of the cord and optic nerve, with sparing of the brain. To assess the actual role of cord pathology on the pattern of movement-associated cortical recruitment, we obtained functional magnetic resonance imaging (fMRI) from patients with DNO and investigated whether the extent of brain activation is correlated with the extent of cervical cord damage. We studied 10 right-handed DNO patients and 15 sex- and agematched healthy controls. The MRI assessment consisted of the following: (a) fMRI during repetitive flexion extension of the last four fingers of the right and left hand, (b) brain and cervical cord conventional MRI, and (c) cervical cord magnetization transfer (MT) MRI. Compared to controls and for both tasks, DNO patients had an increased recruitment of several regions of the sensorimotor network (primary sensorimotor cortex, postcentral gyrus, middle frontal gyrus, rolandic operculum, secondary sensorimotor cortex, precuneus, and cerebellum) and of several other regions mainly in the temporal and occipital lobes, such as MT/V5, the fusiform gyrus, the cuneus, and the parahippocampal gyrus. For both tasks, strong correlations (r values ranging from 0.76 to 0.85) were found between relative activations of cortical sensorimotor areas and the severity of cervical cord damage. This study shows an abnormal pattern of movement-associated cortical activations in patients with DNO, which extends beyond the ‘classical’ sensorimotor network and also involves visual areas devoted to motion processing. The correlation found between fMRI changes and the extent of cord damage suggests that such functional cortical changes might have an adaptive role in limiting the clinical outcome of DNO structural pathology. D 2003 Elsevier Inc. All rights reserved. Keywords: Devic’s neuromyelitis optica; Functional magnetic resonance imaging; Cortical changes
Introduction Several neurophysiological (Green et al., 1998, 1999), positron emission tomography (PET) (Bruehlmeier et al., 1998), and functional magnetic resonance imaging (fMRI) (Cramer et al., 2001; Curt et al., 2002; Lotze et al., 1999; Mikulis et al., 2002; Sabbah et al., 2002) studies have shown that the human brain is capable of extensive reorganization after spinal cord damage of different etiology, including traumatic spinal cord injury (Bruehlmeier et al., 1998; Green et al., 1998, 1999; Lotze et al., 1999; Mikulis et al., 2002) and acute (Cramer et al., 2001) and chronic (Filippi et al., 2002) demyelination. Although a possible association between the extent of cervical cord damage and the degree of movementassociated cortical reorganization has been suggested in patients with multiple sclerosis (MS) (Filippi et al., 2002), the actual role of cord damage in eliciting functional changes of the brain has not been fully elucidate yet since tissue damage in MS is not limited to the cord but it affects diffusely the entire central nervous system (CNS). To elucidate whether and to which extent spinal cord damage in the absence of—or with only minimal—brain damage affects the pattern of movement-associated cortical activations, we investigated, using fMRI, patients with Devic’s neuromyelitis optica (DNO) during the performance of simple motor tasks with their clinically unaffected upper limbs. DNO is a condition characterized by a severe cord and optic nerve damage, with no or only minimal involvement of the brain (Wingerchuk et al., 1999). We also attempted to collect evidence for a potential adaptive role of cortical reorganization in these patients by investigating the correlation between the extent of brain functional recruitment and the extent of cervical cord damage, measured with conventional and magnetization transfer (MT) MRI. Patients and methods
* Corresponding author. Neuroimaging Research Unit Department of Neurology, Scientific Institute and University Ospedale San Raffaele, Via Olgettina, 60, 20132 Milan, Italy. Fax: +39-2-2643-3054. E-mail address: [email protected] (M. Filippi). Available online on ScienceDirect (www.sciencedirect.com.) 1053-8119/$ - see front matter D 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2003.10.013
Patients We studied 10 right-handed individuals with a diagnosis of DNO according to the criteria of Wingerchuk et al. (1999). There
1062
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
were six women and four men. Their mean age was 48.1 years (range = 31 – 68 years), median disease duration was 11.3 years (range = 2 – 25 years), and median Expanded Disability Status Scale (Kurtzke, 1983) score was 4.0 (range = 1.0 – 6.5). All patients had a multiphasic involvement of the optic nerve and spinal cord. At the time of the MRI exams, six patients were on immunosuppressive treatment. Fifteen right-handed healthy volunteers with no previous history of neurological dysfunction and a normal neurological exam (nine women and six men, mean age = 48.3 years, range = 34 – 62 years) served as controls. All subjects were assessed clinically by a single neurologist who was unaware of
the MRI results. Local ethical committee approval and written informed consent for each subject were obtained before study initiation. Functional assessment Motor functional assessment was performed for all the subjects on the same day of MRI acquisition using the ninehole peg test (9-HPT) and the maximum finger-tapping frequency (Herndon, 1997). The maximum finger-tapping rate was observed for two 30-s trial periods outside the magnet and the
Fig. 1. Cortical activations on a rendered brain from right-handed healthy subjects (A, C, and E) and patients with Devic’s neuromyelitis optica (B, D, and F) during the performance of a simple motor task with their clinically unimpaired and fully normal functioning right hands (within-group analysis; one-sample t test). Activated foci are shown with a significance threshold set at P < 0.05, corrected for multiple comparisons (color-coded t values). See text for further details.
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
mean frequency to the nearest 0.5 Hz entered the analysis. To avoid differences in task execution between DNO patients and healthy volunteers, patients were asked to perform a task during fMRI acquisition only if they had a functional score within two standard deviations (SD) from the mean score obtained from controls. Experimental design Using a block design (ABAB), where five periods of activation were alternated with six periods of rest (each period of activation and rest consisting of five measurements), the subjects were
1063
scanned while performing two simple motor tasks, each of them consisting of 60 measurements. Task 1 consisted of repetitive flexion extension of the last four fingers of the right hand moving together. Task 2 was identical to Task 1, but performed with the left hand. The tasks were paced by a metronome at a 1-Hz frequency. Patients and controls were trained before performing the experiments. Subjects were instructed to keep their eyes closed during fMRI acquisition and were monitored visually during scanning to ensure accurate task performance and to assess for additional (e.g., mirror) movements. After functional assessment, seven patients were asked to perform both tasks, one only for Task 1, and two only for Task 2.
Fig. 2. Cortical activations on a rendered brain from right-handed healthy subjects (A, C, and E) and patients with Devic’s neuromyelitis optica (B, D, and F) during the performance of a simple motor task with their clinically unimpaired and fully normal functioning upper left hands (within-group analysis; onesample t test). Activated foci are shown with a significance threshold set at P < 0.05, corrected for multiple comparisons (color-coded t values). See text for further details.
1064
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
fMRI acquisition Brain MRI scans were obtained using a magnet operating at 1.5 T (Vision, Siemens, Erlangen, Germany). Sagittal T1-weighted images were acquired to define the anterior – posterior commissural (AC – PC) plane. Functional MR images were acquired using a T2*-weighted single-shot echo-planar imaging (EPI) sequence (TE = 66 ms, flip angle = 90j, matrix size = 128 128, field of view = 256 256 mm, TR = 3 s). Twenty-four axial slices, parallel to the AC – PC plane, with a thickness of 5 mm covering the whole brain were acquired during each measurement. Shimming was performed for the entire brain using an autoshim routine, which yielded satisfactory magnetic field homogeneity. Structural MRI acquisition On the same occasion and using the same magnet, the following additional sequences were obtained: (1) brain dual-echo turbo spin echo (TSE) (TR/TE = 3300/16, 98; echo train length = 5, 24 axial slices, 5-mm thickness) with the same orientation as the fMRI data set; (2) brain sagittal 3D T1-weighted magnetization prepared rapid acquisition gradient echo (MP-RAGE) (TR/TE = 11.4/4.4, FA = 15j, FOV = 256 256, matrix size = 256 256, slab thickness 160 mm, voxel size = 1 1 1 mm3); (3) cervical cord fast short-tau inversion recovery (STIR) (TR = 2288, TE = 60, TI = 110, ETL =
11, FOV = 280 280 mm, matrix size = 264 512, number of signal averages = 4, 8 contiguous sagittal slices, 3-mm thickness); and (4) cervical cord 2D GE (TR/TE = 640/12, flip angle = 20j, 20 contiguous axial slices, 5-mm thickness), with and without an offresonance radio frequency (RF) saturation pulse (offset frequency = 1.5 kHz, Gaussian envelope duration = 16.4 ms, flip angle = 500j). FMRI analysis All image postprocessing was performed on an independent computer workstation (Sun Sparcstation, Sun Microsystems, Mountain View, CA). FMRI data were analyzed using the statistical parametric mapping (SPM99) software developed by Friston et al. (1995). Before statistical analysis, all images were realigned to the first one to correct for subject motion, spatially normalized into the standard space of SPM, and smoothed with a 10-mm 3DGaussian filter. Structural MRI postprocessing All the structural MRI analysis was performed by a single experienced observer, unaware to whom the scans belonged and blinded to the fMRI results. Whenever present, brain lesions were identified on the proton-density (PD)-weighted scans. T2-weighted images were always used to increase confidence in lesion detection.
Fig. 3. Relative cortical activations in patients with Devic’s neuromyelitis optica during a simple motor task with the right hand in comparison to healthy volunteers (color-coded t values). (A) Contralateral fusiform gyrus, posterior lobe of the cerebellum, and ipsilateral parahippocampal gyrus. (B) Ipsilateral superior temporal gyrus. (C) Ipsilateral rolandic operculum, secondary sensorimotor cortex, and MT/V5 complex. (D) Ipsilateral precuneus. (E) Ipsilateral superior frontal sulcus, precentral gyrus, and superior parietal lobule. (F) Bilateral primary sensorimotor cortex.
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068 Table 1 Cervical cord MTR histogram-derived metrics from healthy volunteers and patients with DNO
Average MTR [%] MTR histogram peak height MTR histogram peak location
Healthy volunteers (SD)
DNO patients (SD)
Pa
40.4 (1.4) 50.0 (10.5)
33.2 (2.4) 42.8 (7.0)
<0.0001 n.s.
38.1 (6.7)
29.3 (4.3)
0.01
DNO: Devic’s neuromyelitis optica; MTR: magnetization transfer ratio; SD: standard deviation; n.s.: not significant. a For statistical analysis, see the text.
Cervical cord lesions were identified by the same observer on the fast-STIR scans. From the two GE images, with and without the saturation pulse, MTR maps were derived as described previously (Filippi et al., 2000). Then, MTR histograms of the entire cervical cord were derived from all subjects (Filippi et al., 2000). For each histogram, the following measures were considered: the relative peak height (i.e., the proportion of pixels at the most common MTR value), the peak location (i.e., the most common MTR value), and the average MTR. Statistical analysis Changes in blood oxygenation level dependent (BOLD) contrast associated with the performance of the motor tasks were assessed on a pixel-by-pixel basis, using the general linear model (Friston et al., 1995) and the theory of Gaussian fields (Worsley and Friston, 1995). Specific effects were tested by applying appropriate linear contrasts. Significant hemodynamic changes for each contrast were assessed using t statistical parametric maps (SPMt). The intragroup activations and comparisons between groups were investigated using a random-effect analysis (Friston et al., 1999), with a one- or two-sample t test performed as appropriate. We report activations below a threshold of P < 0.05 corrected for multiple comparisons. MP-RAGE images from each subject were coregistered with the corresponding fMRI data sets and normalized into the same standard space. Then, the fMRI results were superimposed on these high-resolution images, and using cluster analysis on a patient-by-patient basis (Poline et al., 1997), we evaluated the spatial extent and the coordinates of the centers of activations of brain areas with significantly different relative activations at group analysis. Differences of these fMRI metrics between DNO patients and healthy volunteers were assessed using a two-tailed Student t test for unpaired data. To assess the correlation of BOLD changes with clinical and MRIderived metrics, these quantities were entered into the SPM design matrix using basic models and linear regression analysis (Friston et al., 1999).
Results Structural MRI All healthy volunteers and six DNO patients had normal brain MRI dual-echo scans. Two DNO patients had one small brain T2 visible lesion each, and the remaining two patients had a few
1065
nonspecific T2 abnormalities, possibly related to aging. No abnormalities were seen on any of the cervical cord scans from healthy volunteers. Cervical cord fast-STIR scans were abnormal in eight patients with DNO. Five of these patients had single lesions extending throughout the entire cervical cord to the upper segment of the dorsal cord. Two patients had a single lesion extending from C2 to C6 and the other patient had a single lesion located at C5 – C6. The remaining two patients had abnormalities of the dorsal cord (one patient had a lesion at D3 – D4 and one lesion at D12 – L1) on MRI scans performed before this study was initiated. In Table 1, MTR histogram-derived metrics of the cervical cord from DNO patients and healthy volunteers are reported. Compared to controls, DNO patients had significantly lower average cervical cord MTR and MTR histogram peak location. Functional MRI During fMRI acquisition, all subjects performed the tasks correctly and no additional movements were noted. In Figs. 1 and 2, the activated areas in healthy volunteers and DNO patients during the performance of the two tasks are shown on a rendered brain. Task#1 Compared to controls, DNO patients had more significant activations of several areas in the ipsilateral hemisphere, including the primary sensorimotor cortex (SMC) (SPM space coordinates: 26, 24, and 56), the superior frontal sulcus (SFS) (SPM space coordinates: 24, 2, and 54), the rolandic operculum (SPM space coordinates: 52, 2, and 14), the precuneus (SPM space coordinates: 12, 66, and 44), the superior parietal lobule (SPM space coordinates: 36, 44, and 52), the secondary sensorimotor cortex (SII) (SPM space coordinates: 64, 38, and 34), the MT/V5 complex (SPM space coordinates: 46, 56, and 12), the temporal lobe in a region in the superior temporal gyrus (SPM space coordinates: 44, 8, and 16), and in another region in the parahippocampal gyrus (SPM space coordinates: 28, 14, and 26). They also had increased recruitment of areas in the contralateral hemisphere, including the primary SMC (SPM space coordinates: 16, 28, and 62), the posterior lobe of the cerebellum (SPM space coordinates: 6, 62, and 26), and the fusiform gyrus (SPM space coordinates: 38, 4, and 26) (Fig. 3). There were no areas significantly less activated in DNO patients than in controls. The two groups did not differ in terms of the spatial extent and coordinates of the centers of activations of all the areas with significantly different relative activations at random effect analysis. In DNO patients, the relative activation of the ipsilateral postcentral gyrus was correlated with cervical cord average MTR (r = 0.85, P < 0.001) (Fig. 4) and MTR peak position (r = 0.83, P < 0.001). Task#2 Compared to controls, DNO patients had more significant activations of several areas in the ipsilateral hemisphere, including the postcentral gyrus (SPM space coordinates: 24, 54, and 66), the precuneus (SPM space coordinates: 16, 70, and 56), the cuneus (SPM space coordinates: 6, 80, and 34), and the primary SMC (SPM space coordinates: 36, 14, and 58). They also had increased recruitment of areas in the contralateral hemisphere, including the precuneus (SPM space coordinates: 20, 70, and 52), the inferior parietal lobule (IPL) (SPM space coordinates:
1066
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
Fig. 4. Correlations between relative activation of the ipsilateral postcentral gyrus and cervical cord average MTR in patients with Devic’s neuromyelitis optica.
34, 46, 50 and 52, 38, 40), the middle frontal gyrus (MFG) (SPM space coordinates: 52, 4, and 40), the MT/V5 complex (SPM space coordinates: 50, 64, and 16), and the inferior frontal gyrus (IFG) (SPM space coordinates: 48, 0, and 26). There were no areas significantly less activated in DNO patients than in controls. The two groups did not differ in terms of the spatial extent and coordinates of the centers of activations of all the areas with significantly different relative activations at random effect analysis. In DNO patients, cervical cord average MTR was correlated with the relative activation of the contralateral postcentral gyrus (r = 0.84, P < 0.001). Cervical cord MTR histogram peak position was correlated with the activity of the contralateral postcentral gyrus (r = 0.76, P < 0.001), the contralateral MFG (r = 0.76, P < 0.001), and the ipsilateral IPL (r = 0.83, P < 0.001). In DNO patients, no correlation was found between fMRI metrics (both tasks) and (a) clinical (EDSS score, 9-HPT score, finger-tapping rate, disease duration, and time since last myelitis attack) and (b) conventional MRI measures (presence, length, and volume of cord lesions). The results of the between group comparisons did not change when the two patients without fast-STIR lesions in the cervical cord were removed from the analysis.
Discussion This study demonstrates that functional cortical or subcortical changes do occur in patients with DNO and that such changes might have an adaptive role in limiting the clinical impact of cord damage. Previous fMRI studies already assessed the movement-associated patterns of cortical activations in patients with cord damage of different origins (Cramer et al., 2001; Curt et al., 2002; Mikulis et al., 2002; Sabbah et al., 2002). In general, when patients and controls were contrasted, all these studies disclosed an abnormal location of
the activated areas and a more widespread recruitment of the ‘‘classical’’ sensorimotor network, including the primary SMC, the supplementary motor area (SMA), the prefrontal cortex, the thalami, and the cerebellum. The novelty of this study was to assess the movement-associated cortical changes in patients with DNO, a neurological condition that can be viewed as a ‘natural model’ to elucidate the role of cord damage in eliciting cortical functional reorganization. In addition, to avoid the influence of different task performance on fMRI findings, we carefully selected patients with a preserved function of the tested limbs. When DNO patients and controls were contrasted for the performance of the two motor tasks (with the dominant and the nondominant hands), they showed an increased recruitment of several areas of the sensorimotor network (primary SMC, postcentral gyrus, MFG, rolandic operculum, secondary SII, precuneus, and cerebellum) and of several other areas mainly in the temporal and occipital lobes, such as MT/V5, the fusiform gyrus, the cuneus, and the parahippocampal gyrus. An increased recruitment of the contra- or ipsilateral primary SMC has already been demonstrated in patients with myelitis (Bruehlmeier et al., 1998; Cramer et al., 2001; Curt et al., 2002) and has been related to the time elapsed between the onset of clinical symptoms and evaluation (Green et al., 1999), the level of spinal cord injury (Bruehlmeier et al., 1998), and the degree of daily hand activity (Cramer et al., 2001). This is the first report of a bilateral recruitment of the primary SMC in patients with cord damage during the performance of finger movements with the clinically unaffected dominant hand. Since ipsilateral pathways contribute to the control of hand movements in normal adults (Kim et al., 1993) and their recruitment seems to increase with increased task complexity (Wexler et al., 1997), it is likely that the reported ipsilateral SMC activation might represent a compensatory mechanism with the potential to facilitate motor unit firing. However, contrary to previous PET (Bruehlmeier et al., 1998) and fMRI (Lotze et al., 1999) studies, which also reported a shifted location of the primary motor hand area in patients with cord injury, the present study showed that the somatotopic location of the hand areas in both hemispheres was preserved in patients with DNO. This discrepancy might be due to the different nature and location of cord pathology since the previous two studies evaluated patients with traumatic cord injuries, mainly affecting the thoracic cord (Bruehlmeier et al., 1998; Lotze et al., 1999). That, in an attempt to limit the functional consequences of cord damage, DNO patients might tend to recruit regions that are usually activated in healthy individuals when performing complex or novel tasks is also supported by the finding of an increased activation (during the performance of both tasks) of regions of the sensorimotor network in the frontal (MFG and IFG) and in the parietal lobes (postcentral gyrus, precuneus, and IPL). These regions are reciprocally interconnected and have extensive projections to the primary SMC and the spinal cord (Rizzolatti et al., 1997; Rizzolatti and Luppino, 2001). Contrary to a previous study of patients with myelitis (Curt et al., 2002), patients with DNO did not have an increased activation of the SMA, which is an important component of the cortical motor network, contributing to movement preparation, coordination, and execution. Since efferent from the SMA projects directly to the brainstem and the spinal cord (Rouiller et al., 1994; Weilke et al., 2001), a plausible explanation for the lack of an increased SMA recruitment in DNO patients
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
might be related to the severe damage of these projections in the cord of these patients. DNO patients also showed an increased recruitment of areas in the temporal and occipital lobes and of another region in the posterior lobe of the cerebellum. Interestingly, the posterior lobe of the cerebellum has been related to motor imagery (Grafton et al., 1996) and motor learning (Sakai et al., 1998). The posterior lobe of the cerebellum also has projections to regions of the parietal cortex involved in the processing of sensorial information (Ehrsson et al., 2002). The regions in the occipital and temporal lobes, which were found to be more activated in DNO patients than in controls, are all part of the visual system. As a consequence, the increased recruitment of these areas in DNO patients might be the result of cross-modal plasticity, as it has been previously shown in blind people (Cohen et al., 1997). Admittedly, only one of the patients we studied was blind, but nonetheless all the others had had severe optic neuritis, which resulted in marked reduction of visual acuity. Intriguingly, the transfer of somatosensory information to the occipital cortex is thought to be mediated by connections between parietal and visual associative areas (Bruce et al., 1981), which were significantly activated in our DNO patients. In addition, the lateral temporal cortex and the parahippocampal gyrus (as well as the parietal cortex) are considered sites of heteromodal neurons where modality-specific sensory inputs are bound into a multimodal representation (Calvert, 2001). Therefore, the increased recruitment of these regions in patients with DNO might be the result of their enhanced firing secondary to the severe cord damage of afferent and efferent pathways to and from the primary unimodal cortices. Consistently with the interpretation that the detected fMRI changes may have an adaptive role in limiting the clinical consequences of DNO-related cord injury, our results demonstrate that, at least for some of the activated cortical areas, during the performance of both motor tasks, there is a strong correlation between the extent of fMRI activations and the extent of cervical cord damage. Widespread and marked abnormalities of cervical cord MTR histogram metrics in patients with DNO have already been demonstrated by a previous study (Filippi et al., 1999). The present one confirms and extends the previous findings by showing that these abnormalities might have a role in eliciting adaptive functional changes. It is also worth noting that cord MTR metrics were found to be correlated with increased recruitment of the sensorimotor network and not with the increased recruitment of the visual regions. One might speculate that an MTR assessment of the optic nerve damage (Inglese et al., 2002) might disclose a similar relationship between the activation of the visual network and the severity of the optic nerve structural changes.
Acknowledgments This study was supported by a grant from Fondazione Italiana Sclerosi Multipla (FISM/2002/R/28).
References Bruce, C., Desimone, R., Gross, C.G., 1981. Visual properties of neurons in a polysensory area in superior temporal sulcus of the macaque. J. Neurophysiol. 46, 369 – 384.
1067
Bruehlmeier, M., Dietz, V., Leenders, K., Roelcke, U., Missimer, J., Curt, A., 1998. How does the human brain deal with a spinal cord injury? Eur. J. Neurosci. 10, 3918 – 3922. Calvert, G.A., 2001. Crossmodal processing in the human brain: insights from functional neuroimaging studies. Cereb. Cortex 11, 1110 – 1123 (review). Cohen, L.G., Celnik, P., Pascual-Leone, A., Corwell, B., Falz, L., Dambrosia, J., Honda, M., Sadato, N., Gerloff, C., Catala, M.D., Hallett, M., 1997. Functional relevance of cross-modal plasticity in blind humans. Nature 389, 180 – 183. Cramer, S.C., Fray, E., Tievsky, A., Parker, R.A., Riskind, P.N., Stein, M.C., Wedeen, V., Rosen, B.R., 2001. Changes in motor cortex activation after recovery from spinal cord inflammation. Mult. Scler. 7, 364 – 370. Curt, A., Alkadhi, H., Crelier, G.R., Boendermaker, S.H., Hepp-Reymond, M.C., Kollias, S.S., 2002. Changes of non-affected upper limb cortical representation in paraplegic patients as assessed by fMRI. Brain 125, 2567 – 2578. Ehrsson, H.H., Kuhtz-Buschbeck, J.P., Forssberg, H., 2002. Brain regions controlling nonsynergistic versus synergistic movement of the digits: a functional magnetic resonance imaging study. J. Neurosci. 22, 5074 – 5080. Filippi, M., Rocca, M.A., Moiola, L., Martinelli, V., Ghezzi, A., Capra, R., Salvi, F., Comi, G., 1999. MRI and magnetization transfer imaging changes in the brain and cervical cord of patients with Devic’s neuromyelitis optica. NeuroImage 15, 537 – 546. Filippi, M., Bozzali, M., Horsfield, M.A., Rocca, M.A., Sormani, M.P., Iannucci, G., Colombo, B., Comi, G., 2000. A conventional and magnetization transfer MRI study of the cervical cord in patients with MS. Neurology 53, 1705 – 1710. Filippi, M., Rocca, M.A., Falini, A., Caputo, D., Ghezzi, A., Colombo, B., Scotti, G., Comi, G., 2002. Correlations between structural CNS damage and functional MRI changes in primary progressive MS. Neurology 54, 207 – 213. Friston, K.J., Holmes, A.P., Poline, J.B., Grasby, P.J., Williams, S.C., Frackowiak, R.S., Turner, R., 1995. Analysis of fMRI time-series revisited. NeuroImage 2, 45 – 53. Friston, K.J., Holmes, A.P., Price, C.J, Buchel, C., Worsley, K.J., 1999. Multisubject fMRI studies and conjunction analyses. NeuroImage 10, 385 – 396. Grafton, S.T., Arbib, M.A., Fadiga, L., Rizzolatti, G., 1996. Localization of grasp representations in humans by positron emission tomography. 2. Observation compared with imagination. Exp. Brain Res. 112, 103 – 111. Green, J.B., Sora, E., Bialy, Y., Ricamato, A., Thatcher, R.W., 1998. Cortical sensorimotor reorganization after spinal cord injury: an electroencephalographic study. Neurology 50, 1115 – 1121. Green, J.B., Sora, E., Bialy, Y., Ricamato, A., Thatcher, R.W., 1999. Cortical motor reorganization after paraplegia: an EEG study. Neurology 53, 736 – 743. Herndon, R.M., 1997. Handbook of neurologic rating scales. Demos Vermande, New York. Inglese, M., Ghezze, A., Bianchi, S., Gerevini, S., Sormani, M.P., Martinelli, V., Comi, G., Filippi, M., 2002. Irreversible disability and tissue loss in multiple sclerosis: a conventional and magnetization transfer magnetic resonance imaging study of the optic nerves. Arch. Neurol. 59, 250 – 255. Kim, S.G., Ashe, J., Hendrich, K., Ellermann, J.M., Merkle, H., Ugurbil, K., Georgopoulos, A.P., 1993. Functional magnetic resonance imaging of motor cortex: hemispheric asymmetry and handedness. Science 261, 615 – 617. Kurtzke, J.F., 1983. Rating neurological impairment in multiple sclerosis: en expanded disability status scale (EDSS). Neurology 33, 1444 – 1452. Lotze, M., Laubis-Herrmann, U., Topka, H., Erb, M., Grodd, W., 1999. Reorganization in the primary motor cortex after spinal cord injury—A functional magnetic resonance (fMRI) study. Restor. Neurol. Neurosci. 14, 183 – 187.
1068
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
Mikulis, D.J., Jurkiewicz, M.T., McIlroy, W.E., Staines, W.R., Rickards, L., Kalsi-Ryan, S., Crawley, A.P., Fehlings, M.G., Verrier, M.C., 2002. Adaptation in the motor cortex following cervical spinal cord injury. Neurology 58, 794 – 801. Poline, J.B., Worsley, K.J., Evans, A.C., Friston, K.J., 1997. Combining spatial extent and peak intensity to test for activations in functional imaging. NeuroImage 5, 83 – 96. Rizzolatti, G., Luppino, G., 2001. The cortical motor system. Neuron 31, 889 – 901 (review). Rizzolatti, G., Fogassi, L., Gallese, V., 1997. Parietal cortex: from sight to action. Curr. Opin. Neurobiol. 7, 562 – 567 (review). Rouiller, E.M., Babalian, A., Kazennikov, O., et al., 1994. Transcallosal connections of the distal forelimb representations of the primary and supplementary motor cortical areas in macaque monkeys. Exp. Brain Res. 102, 227 – 243. Sabbah, P., de Schonen, V., Leveque, C., Gay, S., Pfefer, F., Nioche, C., Sarrazin, J.L., Barouti, H., Tadie, M., Cordoliani, Y., 2002. Sensorimo-
tor cortical activity in patients with complete spinal cord injury: a functional magnetic resonance imaging study. J. Neurotrauma 19, 53 – 60. Sakai, K., Takino, R., Hikosaka, O., Miyauchi, S., Sasaki, Y., Putz, B., Fujimaki, N., 1998. Separate cerebellar areas for motor control. NeuroReport 9, 2359 – 2363. Weilke, F., Spiegel, S., Boecker, H., von Einsiedel, H.G., Conrad, B., Schwaiger, M., Erhard, P., 2001. Time-resolved fMRI of activation patterns in M1 and SMA during complex voluntary movement. J. Neurophysiol. 85, 1858 – 1863. Wexler, B.E., Fulbright, R.K., Lacadie, C.M., et al., 1997. An fMRI study of the human cortical motor system response to increasing functional demands. Magn. Reson. Imaging 15, 385 – 396. Wingerchuk, D.M., Hogancamp, W.F., O’Brien, P.C., Weinshenker, B.G., 1999. The clinical course of neuromyelitis optica (Devic’s syndrome). Neurology 53, 1107 – 1114. Worsley, K.J., Friston, K.J., 1995. Analysis of fMRI time-series revisited again. NeuroImage 2, 173 – 181.
A functional MRI study of movement-associated cortical changes in patients with Devic’s neuromyelitis optica M.A. Rocca, a,b F. Agosta, a,b D.M. Mezzapesa, a,b A. Falini, c V. Martinelli, b F. Salvi, d R. Bergamaschi, e G. Scotti, c G. Comi, b and M. Filippi a,b,* a
Neuroimaging Research Unit, Scientific Institute and University Ospedale San Raffaele, Milan, Italy Department of Neurology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy c Department of Neuroradiology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy d Department of Clinical Neurology, Ospedale Bellaria, University of Bologna, Bologna, Italy e Department of Neurology, Istituto Mondino, Pavia, Italy b
Received 22 July 2003; revised 7 October 2003; accepted 7 October 2003
Movement-associated cortical changes have been shown in several neurological conditions and were found to be associated to the extent of brain and cord damage. Devic’s neuromyelitis optica (DNO) is characterized by a severe involvement of the cord and optic nerve, with sparing of the brain. To assess the actual role of cord pathology on the pattern of movement-associated cortical recruitment, we obtained functional magnetic resonance imaging (fMRI) from patients with DNO and investigated whether the extent of brain activation is correlated with the extent of cervical cord damage. We studied 10 right-handed DNO patients and 15 sex- and agematched healthy controls. The MRI assessment consisted of the following: (a) fMRI during repetitive flexion extension of the last four fingers of the right and left hand, (b) brain and cervical cord conventional MRI, and (c) cervical cord magnetization transfer (MT) MRI. Compared to controls and for both tasks, DNO patients had an increased recruitment of several regions of the sensorimotor network (primary sensorimotor cortex, postcentral gyrus, middle frontal gyrus, rolandic operculum, secondary sensorimotor cortex, precuneus, and cerebellum) and of several other regions mainly in the temporal and occipital lobes, such as MT/V5, the fusiform gyrus, the cuneus, and the parahippocampal gyrus. For both tasks, strong correlations (r values ranging from 0.76 to 0.85) were found between relative activations of cortical sensorimotor areas and the severity of cervical cord damage. This study shows an abnormal pattern of movement-associated cortical activations in patients with DNO, which extends beyond the ‘classical’ sensorimotor network and also involves visual areas devoted to motion processing. The correlation found between fMRI changes and the extent of cord damage suggests that such functional cortical changes might have an adaptive role in limiting the clinical outcome of DNO structural pathology. D 2003 Elsevier Inc. All rights reserved. Keywords: Devic’s neuromyelitis optica; Functional magnetic resonance imaging; Cortical changes
Introduction Several neurophysiological (Green et al., 1998, 1999), positron emission tomography (PET) (Bruehlmeier et al., 1998), and functional magnetic resonance imaging (fMRI) (Cramer et al., 2001; Curt et al., 2002; Lotze et al., 1999; Mikulis et al., 2002; Sabbah et al., 2002) studies have shown that the human brain is capable of extensive reorganization after spinal cord damage of different etiology, including traumatic spinal cord injury (Bruehlmeier et al., 1998; Green et al., 1998, 1999; Lotze et al., 1999; Mikulis et al., 2002) and acute (Cramer et al., 2001) and chronic (Filippi et al., 2002) demyelination. Although a possible association between the extent of cervical cord damage and the degree of movementassociated cortical reorganization has been suggested in patients with multiple sclerosis (MS) (Filippi et al., 2002), the actual role of cord damage in eliciting functional changes of the brain has not been fully elucidate yet since tissue damage in MS is not limited to the cord but it affects diffusely the entire central nervous system (CNS). To elucidate whether and to which extent spinal cord damage in the absence of—or with only minimal—brain damage affects the pattern of movement-associated cortical activations, we investigated, using fMRI, patients with Devic’s neuromyelitis optica (DNO) during the performance of simple motor tasks with their clinically unaffected upper limbs. DNO is a condition characterized by a severe cord and optic nerve damage, with no or only minimal involvement of the brain (Wingerchuk et al., 1999). We also attempted to collect evidence for a potential adaptive role of cortical reorganization in these patients by investigating the correlation between the extent of brain functional recruitment and the extent of cervical cord damage, measured with conventional and magnetization transfer (MT) MRI. Patients and methods
* Corresponding author. Neuroimaging Research Unit Department of Neurology, Scientific Institute and University Ospedale San Raffaele, Via Olgettina, 60, 20132 Milan, Italy. Fax: +39-2-2643-3054. E-mail address: [email protected] (M. Filippi). Available online on ScienceDirect (www.sciencedirect.com.) 1053-8119/$ - see front matter D 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2003.10.013
Patients We studied 10 right-handed individuals with a diagnosis of DNO according to the criteria of Wingerchuk et al. (1999). There
1062
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
were six women and four men. Their mean age was 48.1 years (range = 31 – 68 years), median disease duration was 11.3 years (range = 2 – 25 years), and median Expanded Disability Status Scale (Kurtzke, 1983) score was 4.0 (range = 1.0 – 6.5). All patients had a multiphasic involvement of the optic nerve and spinal cord. At the time of the MRI exams, six patients were on immunosuppressive treatment. Fifteen right-handed healthy volunteers with no previous history of neurological dysfunction and a normal neurological exam (nine women and six men, mean age = 48.3 years, range = 34 – 62 years) served as controls. All subjects were assessed clinically by a single neurologist who was unaware of
the MRI results. Local ethical committee approval and written informed consent for each subject were obtained before study initiation. Functional assessment Motor functional assessment was performed for all the subjects on the same day of MRI acquisition using the ninehole peg test (9-HPT) and the maximum finger-tapping frequency (Herndon, 1997). The maximum finger-tapping rate was observed for two 30-s trial periods outside the magnet and the
Fig. 1. Cortical activations on a rendered brain from right-handed healthy subjects (A, C, and E) and patients with Devic’s neuromyelitis optica (B, D, and F) during the performance of a simple motor task with their clinically unimpaired and fully normal functioning right hands (within-group analysis; one-sample t test). Activated foci are shown with a significance threshold set at P < 0.05, corrected for multiple comparisons (color-coded t values). See text for further details.
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
mean frequency to the nearest 0.5 Hz entered the analysis. To avoid differences in task execution between DNO patients and healthy volunteers, patients were asked to perform a task during fMRI acquisition only if they had a functional score within two standard deviations (SD) from the mean score obtained from controls. Experimental design Using a block design (ABAB), where five periods of activation were alternated with six periods of rest (each period of activation and rest consisting of five measurements), the subjects were
1063
scanned while performing two simple motor tasks, each of them consisting of 60 measurements. Task 1 consisted of repetitive flexion extension of the last four fingers of the right hand moving together. Task 2 was identical to Task 1, but performed with the left hand. The tasks were paced by a metronome at a 1-Hz frequency. Patients and controls were trained before performing the experiments. Subjects were instructed to keep their eyes closed during fMRI acquisition and were monitored visually during scanning to ensure accurate task performance and to assess for additional (e.g., mirror) movements. After functional assessment, seven patients were asked to perform both tasks, one only for Task 1, and two only for Task 2.
Fig. 2. Cortical activations on a rendered brain from right-handed healthy subjects (A, C, and E) and patients with Devic’s neuromyelitis optica (B, D, and F) during the performance of a simple motor task with their clinically unimpaired and fully normal functioning upper left hands (within-group analysis; onesample t test). Activated foci are shown with a significance threshold set at P < 0.05, corrected for multiple comparisons (color-coded t values). See text for further details.
1064
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
fMRI acquisition Brain MRI scans were obtained using a magnet operating at 1.5 T (Vision, Siemens, Erlangen, Germany). Sagittal T1-weighted images were acquired to define the anterior – posterior commissural (AC – PC) plane. Functional MR images were acquired using a T2*-weighted single-shot echo-planar imaging (EPI) sequence (TE = 66 ms, flip angle = 90j, matrix size = 128 128, field of view = 256 256 mm, TR = 3 s). Twenty-four axial slices, parallel to the AC – PC plane, with a thickness of 5 mm covering the whole brain were acquired during each measurement. Shimming was performed for the entire brain using an autoshim routine, which yielded satisfactory magnetic field homogeneity. Structural MRI acquisition On the same occasion and using the same magnet, the following additional sequences were obtained: (1) brain dual-echo turbo spin echo (TSE) (TR/TE = 3300/16, 98; echo train length = 5, 24 axial slices, 5-mm thickness) with the same orientation as the fMRI data set; (2) brain sagittal 3D T1-weighted magnetization prepared rapid acquisition gradient echo (MP-RAGE) (TR/TE = 11.4/4.4, FA = 15j, FOV = 256 256, matrix size = 256 256, slab thickness 160 mm, voxel size = 1 1 1 mm3); (3) cervical cord fast short-tau inversion recovery (STIR) (TR = 2288, TE = 60, TI = 110, ETL =
11, FOV = 280 280 mm, matrix size = 264 512, number of signal averages = 4, 8 contiguous sagittal slices, 3-mm thickness); and (4) cervical cord 2D GE (TR/TE = 640/12, flip angle = 20j, 20 contiguous axial slices, 5-mm thickness), with and without an offresonance radio frequency (RF) saturation pulse (offset frequency = 1.5 kHz, Gaussian envelope duration = 16.4 ms, flip angle = 500j). FMRI analysis All image postprocessing was performed on an independent computer workstation (Sun Sparcstation, Sun Microsystems, Mountain View, CA). FMRI data were analyzed using the statistical parametric mapping (SPM99) software developed by Friston et al. (1995). Before statistical analysis, all images were realigned to the first one to correct for subject motion, spatially normalized into the standard space of SPM, and smoothed with a 10-mm 3DGaussian filter. Structural MRI postprocessing All the structural MRI analysis was performed by a single experienced observer, unaware to whom the scans belonged and blinded to the fMRI results. Whenever present, brain lesions were identified on the proton-density (PD)-weighted scans. T2-weighted images were always used to increase confidence in lesion detection.
Fig. 3. Relative cortical activations in patients with Devic’s neuromyelitis optica during a simple motor task with the right hand in comparison to healthy volunteers (color-coded t values). (A) Contralateral fusiform gyrus, posterior lobe of the cerebellum, and ipsilateral parahippocampal gyrus. (B) Ipsilateral superior temporal gyrus. (C) Ipsilateral rolandic operculum, secondary sensorimotor cortex, and MT/V5 complex. (D) Ipsilateral precuneus. (E) Ipsilateral superior frontal sulcus, precentral gyrus, and superior parietal lobule. (F) Bilateral primary sensorimotor cortex.
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068 Table 1 Cervical cord MTR histogram-derived metrics from healthy volunteers and patients with DNO
Average MTR [%] MTR histogram peak height MTR histogram peak location
Healthy volunteers (SD)
DNO patients (SD)
Pa
40.4 (1.4) 50.0 (10.5)
33.2 (2.4) 42.8 (7.0)
<0.0001 n.s.
38.1 (6.7)
29.3 (4.3)
0.01
DNO: Devic’s neuromyelitis optica; MTR: magnetization transfer ratio; SD: standard deviation; n.s.: not significant. a For statistical analysis, see the text.
Cervical cord lesions were identified by the same observer on the fast-STIR scans. From the two GE images, with and without the saturation pulse, MTR maps were derived as described previously (Filippi et al., 2000). Then, MTR histograms of the entire cervical cord were derived from all subjects (Filippi et al., 2000). For each histogram, the following measures were considered: the relative peak height (i.e., the proportion of pixels at the most common MTR value), the peak location (i.e., the most common MTR value), and the average MTR. Statistical analysis Changes in blood oxygenation level dependent (BOLD) contrast associated with the performance of the motor tasks were assessed on a pixel-by-pixel basis, using the general linear model (Friston et al., 1995) and the theory of Gaussian fields (Worsley and Friston, 1995). Specific effects were tested by applying appropriate linear contrasts. Significant hemodynamic changes for each contrast were assessed using t statistical parametric maps (SPMt). The intragroup activations and comparisons between groups were investigated using a random-effect analysis (Friston et al., 1999), with a one- or two-sample t test performed as appropriate. We report activations below a threshold of P < 0.05 corrected for multiple comparisons. MP-RAGE images from each subject were coregistered with the corresponding fMRI data sets and normalized into the same standard space. Then, the fMRI results were superimposed on these high-resolution images, and using cluster analysis on a patient-by-patient basis (Poline et al., 1997), we evaluated the spatial extent and the coordinates of the centers of activations of brain areas with significantly different relative activations at group analysis. Differences of these fMRI metrics between DNO patients and healthy volunteers were assessed using a two-tailed Student t test for unpaired data. To assess the correlation of BOLD changes with clinical and MRIderived metrics, these quantities were entered into the SPM design matrix using basic models and linear regression analysis (Friston et al., 1999).
Results Structural MRI All healthy volunteers and six DNO patients had normal brain MRI dual-echo scans. Two DNO patients had one small brain T2 visible lesion each, and the remaining two patients had a few
1065
nonspecific T2 abnormalities, possibly related to aging. No abnormalities were seen on any of the cervical cord scans from healthy volunteers. Cervical cord fast-STIR scans were abnormal in eight patients with DNO. Five of these patients had single lesions extending throughout the entire cervical cord to the upper segment of the dorsal cord. Two patients had a single lesion extending from C2 to C6 and the other patient had a single lesion located at C5 – C6. The remaining two patients had abnormalities of the dorsal cord (one patient had a lesion at D3 – D4 and one lesion at D12 – L1) on MRI scans performed before this study was initiated. In Table 1, MTR histogram-derived metrics of the cervical cord from DNO patients and healthy volunteers are reported. Compared to controls, DNO patients had significantly lower average cervical cord MTR and MTR histogram peak location. Functional MRI During fMRI acquisition, all subjects performed the tasks correctly and no additional movements were noted. In Figs. 1 and 2, the activated areas in healthy volunteers and DNO patients during the performance of the two tasks are shown on a rendered brain. Task#1 Compared to controls, DNO patients had more significant activations of several areas in the ipsilateral hemisphere, including the primary sensorimotor cortex (SMC) (SPM space coordinates: 26, 24, and 56), the superior frontal sulcus (SFS) (SPM space coordinates: 24, 2, and 54), the rolandic operculum (SPM space coordinates: 52, 2, and 14), the precuneus (SPM space coordinates: 12, 66, and 44), the superior parietal lobule (SPM space coordinates: 36, 44, and 52), the secondary sensorimotor cortex (SII) (SPM space coordinates: 64, 38, and 34), the MT/V5 complex (SPM space coordinates: 46, 56, and 12), the temporal lobe in a region in the superior temporal gyrus (SPM space coordinates: 44, 8, and 16), and in another region in the parahippocampal gyrus (SPM space coordinates: 28, 14, and 26). They also had increased recruitment of areas in the contralateral hemisphere, including the primary SMC (SPM space coordinates: 16, 28, and 62), the posterior lobe of the cerebellum (SPM space coordinates: 6, 62, and 26), and the fusiform gyrus (SPM space coordinates: 38, 4, and 26) (Fig. 3). There were no areas significantly less activated in DNO patients than in controls. The two groups did not differ in terms of the spatial extent and coordinates of the centers of activations of all the areas with significantly different relative activations at random effect analysis. In DNO patients, the relative activation of the ipsilateral postcentral gyrus was correlated with cervical cord average MTR (r = 0.85, P < 0.001) (Fig. 4) and MTR peak position (r = 0.83, P < 0.001). Task#2 Compared to controls, DNO patients had more significant activations of several areas in the ipsilateral hemisphere, including the postcentral gyrus (SPM space coordinates: 24, 54, and 66), the precuneus (SPM space coordinates: 16, 70, and 56), the cuneus (SPM space coordinates: 6, 80, and 34), and the primary SMC (SPM space coordinates: 36, 14, and 58). They also had increased recruitment of areas in the contralateral hemisphere, including the precuneus (SPM space coordinates: 20, 70, and 52), the inferior parietal lobule (IPL) (SPM space coordinates:
1066
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
Fig. 4. Correlations between relative activation of the ipsilateral postcentral gyrus and cervical cord average MTR in patients with Devic’s neuromyelitis optica.
34, 46, 50 and 52, 38, 40), the middle frontal gyrus (MFG) (SPM space coordinates: 52, 4, and 40), the MT/V5 complex (SPM space coordinates: 50, 64, and 16), and the inferior frontal gyrus (IFG) (SPM space coordinates: 48, 0, and 26). There were no areas significantly less activated in DNO patients than in controls. The two groups did not differ in terms of the spatial extent and coordinates of the centers of activations of all the areas with significantly different relative activations at random effect analysis. In DNO patients, cervical cord average MTR was correlated with the relative activation of the contralateral postcentral gyrus (r = 0.84, P < 0.001). Cervical cord MTR histogram peak position was correlated with the activity of the contralateral postcentral gyrus (r = 0.76, P < 0.001), the contralateral MFG (r = 0.76, P < 0.001), and the ipsilateral IPL (r = 0.83, P < 0.001). In DNO patients, no correlation was found between fMRI metrics (both tasks) and (a) clinical (EDSS score, 9-HPT score, finger-tapping rate, disease duration, and time since last myelitis attack) and (b) conventional MRI measures (presence, length, and volume of cord lesions). The results of the between group comparisons did not change when the two patients without fast-STIR lesions in the cervical cord were removed from the analysis.
Discussion This study demonstrates that functional cortical or subcortical changes do occur in patients with DNO and that such changes might have an adaptive role in limiting the clinical impact of cord damage. Previous fMRI studies already assessed the movement-associated patterns of cortical activations in patients with cord damage of different origins (Cramer et al., 2001; Curt et al., 2002; Mikulis et al., 2002; Sabbah et al., 2002). In general, when patients and controls were contrasted, all these studies disclosed an abnormal location of
the activated areas and a more widespread recruitment of the ‘‘classical’’ sensorimotor network, including the primary SMC, the supplementary motor area (SMA), the prefrontal cortex, the thalami, and the cerebellum. The novelty of this study was to assess the movement-associated cortical changes in patients with DNO, a neurological condition that can be viewed as a ‘natural model’ to elucidate the role of cord damage in eliciting cortical functional reorganization. In addition, to avoid the influence of different task performance on fMRI findings, we carefully selected patients with a preserved function of the tested limbs. When DNO patients and controls were contrasted for the performance of the two motor tasks (with the dominant and the nondominant hands), they showed an increased recruitment of several areas of the sensorimotor network (primary SMC, postcentral gyrus, MFG, rolandic operculum, secondary SII, precuneus, and cerebellum) and of several other areas mainly in the temporal and occipital lobes, such as MT/V5, the fusiform gyrus, the cuneus, and the parahippocampal gyrus. An increased recruitment of the contra- or ipsilateral primary SMC has already been demonstrated in patients with myelitis (Bruehlmeier et al., 1998; Cramer et al., 2001; Curt et al., 2002) and has been related to the time elapsed between the onset of clinical symptoms and evaluation (Green et al., 1999), the level of spinal cord injury (Bruehlmeier et al., 1998), and the degree of daily hand activity (Cramer et al., 2001). This is the first report of a bilateral recruitment of the primary SMC in patients with cord damage during the performance of finger movements with the clinically unaffected dominant hand. Since ipsilateral pathways contribute to the control of hand movements in normal adults (Kim et al., 1993) and their recruitment seems to increase with increased task complexity (Wexler et al., 1997), it is likely that the reported ipsilateral SMC activation might represent a compensatory mechanism with the potential to facilitate motor unit firing. However, contrary to previous PET (Bruehlmeier et al., 1998) and fMRI (Lotze et al., 1999) studies, which also reported a shifted location of the primary motor hand area in patients with cord injury, the present study showed that the somatotopic location of the hand areas in both hemispheres was preserved in patients with DNO. This discrepancy might be due to the different nature and location of cord pathology since the previous two studies evaluated patients with traumatic cord injuries, mainly affecting the thoracic cord (Bruehlmeier et al., 1998; Lotze et al., 1999). That, in an attempt to limit the functional consequences of cord damage, DNO patients might tend to recruit regions that are usually activated in healthy individuals when performing complex or novel tasks is also supported by the finding of an increased activation (during the performance of both tasks) of regions of the sensorimotor network in the frontal (MFG and IFG) and in the parietal lobes (postcentral gyrus, precuneus, and IPL). These regions are reciprocally interconnected and have extensive projections to the primary SMC and the spinal cord (Rizzolatti et al., 1997; Rizzolatti and Luppino, 2001). Contrary to a previous study of patients with myelitis (Curt et al., 2002), patients with DNO did not have an increased activation of the SMA, which is an important component of the cortical motor network, contributing to movement preparation, coordination, and execution. Since efferent from the SMA projects directly to the brainstem and the spinal cord (Rouiller et al., 1994; Weilke et al., 2001), a plausible explanation for the lack of an increased SMA recruitment in DNO patients
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
might be related to the severe damage of these projections in the cord of these patients. DNO patients also showed an increased recruitment of areas in the temporal and occipital lobes and of another region in the posterior lobe of the cerebellum. Interestingly, the posterior lobe of the cerebellum has been related to motor imagery (Grafton et al., 1996) and motor learning (Sakai et al., 1998). The posterior lobe of the cerebellum also has projections to regions of the parietal cortex involved in the processing of sensorial information (Ehrsson et al., 2002). The regions in the occipital and temporal lobes, which were found to be more activated in DNO patients than in controls, are all part of the visual system. As a consequence, the increased recruitment of these areas in DNO patients might be the result of cross-modal plasticity, as it has been previously shown in blind people (Cohen et al., 1997). Admittedly, only one of the patients we studied was blind, but nonetheless all the others had had severe optic neuritis, which resulted in marked reduction of visual acuity. Intriguingly, the transfer of somatosensory information to the occipital cortex is thought to be mediated by connections between parietal and visual associative areas (Bruce et al., 1981), which were significantly activated in our DNO patients. In addition, the lateral temporal cortex and the parahippocampal gyrus (as well as the parietal cortex) are considered sites of heteromodal neurons where modality-specific sensory inputs are bound into a multimodal representation (Calvert, 2001). Therefore, the increased recruitment of these regions in patients with DNO might be the result of their enhanced firing secondary to the severe cord damage of afferent and efferent pathways to and from the primary unimodal cortices. Consistently with the interpretation that the detected fMRI changes may have an adaptive role in limiting the clinical consequences of DNO-related cord injury, our results demonstrate that, at least for some of the activated cortical areas, during the performance of both motor tasks, there is a strong correlation between the extent of fMRI activations and the extent of cervical cord damage. Widespread and marked abnormalities of cervical cord MTR histogram metrics in patients with DNO have already been demonstrated by a previous study (Filippi et al., 1999). The present one confirms and extends the previous findings by showing that these abnormalities might have a role in eliciting adaptive functional changes. It is also worth noting that cord MTR metrics were found to be correlated with increased recruitment of the sensorimotor network and not with the increased recruitment of the visual regions. One might speculate that an MTR assessment of the optic nerve damage (Inglese et al., 2002) might disclose a similar relationship between the activation of the visual network and the severity of the optic nerve structural changes.
Acknowledgments This study was supported by a grant from Fondazione Italiana Sclerosi Multipla (FISM/2002/R/28).
References Bruce, C., Desimone, R., Gross, C.G., 1981. Visual properties of neurons in a polysensory area in superior temporal sulcus of the macaque. J. Neurophysiol. 46, 369 – 384.
1067
Bruehlmeier, M., Dietz, V., Leenders, K., Roelcke, U., Missimer, J., Curt, A., 1998. How does the human brain deal with a spinal cord injury? Eur. J. Neurosci. 10, 3918 – 3922. Calvert, G.A., 2001. Crossmodal processing in the human brain: insights from functional neuroimaging studies. Cereb. Cortex 11, 1110 – 1123 (review). Cohen, L.G., Celnik, P., Pascual-Leone, A., Corwell, B., Falz, L., Dambrosia, J., Honda, M., Sadato, N., Gerloff, C., Catala, M.D., Hallett, M., 1997. Functional relevance of cross-modal plasticity in blind humans. Nature 389, 180 – 183. Cramer, S.C., Fray, E., Tievsky, A., Parker, R.A., Riskind, P.N., Stein, M.C., Wedeen, V., Rosen, B.R., 2001. Changes in motor cortex activation after recovery from spinal cord inflammation. Mult. Scler. 7, 364 – 370. Curt, A., Alkadhi, H., Crelier, G.R., Boendermaker, S.H., Hepp-Reymond, M.C., Kollias, S.S., 2002. Changes of non-affected upper limb cortical representation in paraplegic patients as assessed by fMRI. Brain 125, 2567 – 2578. Ehrsson, H.H., Kuhtz-Buschbeck, J.P., Forssberg, H., 2002. Brain regions controlling nonsynergistic versus synergistic movement of the digits: a functional magnetic resonance imaging study. J. Neurosci. 22, 5074 – 5080. Filippi, M., Rocca, M.A., Moiola, L., Martinelli, V., Ghezzi, A., Capra, R., Salvi, F., Comi, G., 1999. MRI and magnetization transfer imaging changes in the brain and cervical cord of patients with Devic’s neuromyelitis optica. NeuroImage 15, 537 – 546. Filippi, M., Bozzali, M., Horsfield, M.A., Rocca, M.A., Sormani, M.P., Iannucci, G., Colombo, B., Comi, G., 2000. A conventional and magnetization transfer MRI study of the cervical cord in patients with MS. Neurology 53, 1705 – 1710. Filippi, M., Rocca, M.A., Falini, A., Caputo, D., Ghezzi, A., Colombo, B., Scotti, G., Comi, G., 2002. Correlations between structural CNS damage and functional MRI changes in primary progressive MS. Neurology 54, 207 – 213. Friston, K.J., Holmes, A.P., Poline, J.B., Grasby, P.J., Williams, S.C., Frackowiak, R.S., Turner, R., 1995. Analysis of fMRI time-series revisited. NeuroImage 2, 45 – 53. Friston, K.J., Holmes, A.P., Price, C.J, Buchel, C., Worsley, K.J., 1999. Multisubject fMRI studies and conjunction analyses. NeuroImage 10, 385 – 396. Grafton, S.T., Arbib, M.A., Fadiga, L., Rizzolatti, G., 1996. Localization of grasp representations in humans by positron emission tomography. 2. Observation compared with imagination. Exp. Brain Res. 112, 103 – 111. Green, J.B., Sora, E., Bialy, Y., Ricamato, A., Thatcher, R.W., 1998. Cortical sensorimotor reorganization after spinal cord injury: an electroencephalographic study. Neurology 50, 1115 – 1121. Green, J.B., Sora, E., Bialy, Y., Ricamato, A., Thatcher, R.W., 1999. Cortical motor reorganization after paraplegia: an EEG study. Neurology 53, 736 – 743. Herndon, R.M., 1997. Handbook of neurologic rating scales. Demos Vermande, New York. Inglese, M., Ghezze, A., Bianchi, S., Gerevini, S., Sormani, M.P., Martinelli, V., Comi, G., Filippi, M., 2002. Irreversible disability and tissue loss in multiple sclerosis: a conventional and magnetization transfer magnetic resonance imaging study of the optic nerves. Arch. Neurol. 59, 250 – 255. Kim, S.G., Ashe, J., Hendrich, K., Ellermann, J.M., Merkle, H., Ugurbil, K., Georgopoulos, A.P., 1993. Functional magnetic resonance imaging of motor cortex: hemispheric asymmetry and handedness. Science 261, 615 – 617. Kurtzke, J.F., 1983. Rating neurological impairment in multiple sclerosis: en expanded disability status scale (EDSS). Neurology 33, 1444 – 1452. Lotze, M., Laubis-Herrmann, U., Topka, H., Erb, M., Grodd, W., 1999. Reorganization in the primary motor cortex after spinal cord injury—A functional magnetic resonance (fMRI) study. Restor. Neurol. Neurosci. 14, 183 – 187.
1068
M.A. Rocca et al. / NeuroImage 21 (2004) 1061–1068
Mikulis, D.J., Jurkiewicz, M.T., McIlroy, W.E., Staines, W.R., Rickards, L., Kalsi-Ryan, S., Crawley, A.P., Fehlings, M.G., Verrier, M.C., 2002. Adaptation in the motor cortex following cervical spinal cord injury. Neurology 58, 794 – 801. Poline, J.B., Worsley, K.J., Evans, A.C., Friston, K.J., 1997. Combining spatial extent and peak intensity to test for activations in functional imaging. NeuroImage 5, 83 – 96. Rizzolatti, G., Luppino, G., 2001. The cortical motor system. Neuron 31, 889 – 901 (review). Rizzolatti, G., Fogassi, L., Gallese, V., 1997. Parietal cortex: from sight to action. Curr. Opin. Neurobiol. 7, 562 – 567 (review). Rouiller, E.M., Babalian, A., Kazennikov, O., et al., 1994. Transcallosal connections of the distal forelimb representations of the primary and supplementary motor cortical areas in macaque monkeys. Exp. Brain Res. 102, 227 – 243. Sabbah, P., de Schonen, V., Leveque, C., Gay, S., Pfefer, F., Nioche, C., Sarrazin, J.L., Barouti, H., Tadie, M., Cordoliani, Y., 2002. Sensorimo-
tor cortical activity in patients with complete spinal cord injury: a functional magnetic resonance imaging study. J. Neurotrauma 19, 53 – 60. Sakai, K., Takino, R., Hikosaka, O., Miyauchi, S., Sasaki, Y., Putz, B., Fujimaki, N., 1998. Separate cerebellar areas for motor control. NeuroReport 9, 2359 – 2363. Weilke, F., Spiegel, S., Boecker, H., von Einsiedel, H.G., Conrad, B., Schwaiger, M., Erhard, P., 2001. Time-resolved fMRI of activation patterns in M1 and SMA during complex voluntary movement. J. Neurophysiol. 85, 1858 – 1863. Wexler, B.E., Fulbright, R.K., Lacadie, C.M., et al., 1997. An fMRI study of the human cortical motor system response to increasing functional demands. Magn. Reson. Imaging 15, 385 – 396. Wingerchuk, D.M., Hogancamp, W.F., O’Brien, P.C., Weinshenker, B.G., 1999. The clinical course of neuromyelitis optica (Devic’s syndrome). Neurology 53, 1107 – 1114. Worsley, K.J., Friston, K.J., 1995. Analysis of fMRI time-series revisited again. NeuroImage 2, 173 – 181.