Journal Logo

Research papers

Cerebral activation during motor imagery in complex regional pain syndrome type 1 with dystonia

Gieteling, Esther W.a,c,*; van Rijn, Monique A.b; de Jong, Bauke M.a,c; Hoogduin, Johannes M.c; Renken, Remcoc; van Hilten, Jacobus J.b; Leenders, Klaus L.a

Author Information
doi: 10.1016/j.pain.2007.04.029
  • Free


1. Introduction

Complex Regional Pain Syndrome type 1 (CRPS-1), also called Reflex Sympathetic Dystrophy (RSD), is a syndrome predominantly characterised by a variety of sensory, autonomic and trophic features (Veldman et al., 1993; Schwartzman et al., 2006). Symptoms include pain, oedema, hyperhydrosis and impaired function. Growing evidence indicates that CRPS-1 may also include some form of movement disorder, like tremor and dystonia (Schwartzman and Kerrigan, 1990; van Hilten et al., 2005; Geyer and Bressman, 2006).

Dystonia is characterised by involuntary sustained muscle contractions, causing twisting and repetitive movements or abnormal postures (Fahn et al., 1998; Geyer and Bressman, 2006). Functional brain imaging with PET and fMRI has provided valuable new insights into the role of altered activation of basal ganglia-cortical networks during execution of movement in focal and generalized dystonia. Underactivation of the primary motor cortex and overactivation of the somatosensory cortex, prefrontal-, premotor- and parietal cortical regions have been reported. Other studies showed an overactivation of the primary motor cortex and underactivation of the premotor cortex (Playford et al., 1998; Detante et al., 2004; Dresel et al., 2006). These inconsistencies may result from a combination of differences in scanning procedures, tasks and a varying degree of dystonia during motor execution. However, a distinct pathophysiology underlying different forms of dystonia may equally be an option (Dresel et al., 2006).

The pathophysiology of CRPS-1 itself remains a controversial issue (Verdugo and Ochoa, 2000). Some features of CRPS-1 favour a spinal aetiology (van Hilten et al., 2005), others indicate a cerebral reorganization in both the sensory and motor domain (Ribbers et al., 2002; Maihofner et al., 2003). A recent fMRI study in CRPS, focussing on mechanical hyperalgesia, reported alterations in nociceptive, cognitive and motor processing (Maihofner et al., 2005). Hitherto, no functional imaging study in dystonia of CRPS-1 during movement execution has been performed.

In overt dystonic movement, altered sensory feedback (in particular noxious input) during painful movements is a potential confounder in functional brain imaging, which can be reduced by motor imagery, i.e., mental rehearsal of a motor act without overt movement. Previous work suggests that the volume of brain activation differs between execution and motor imagery, but the distribution of cerebral activity tends to be partly similar (Hanakawa et al., 2003; Ehrsson et al., 2003). Motor imagery is used in sport to improve performance. A positive effect of motor imagery in the rehabilitation of CRPS-1 patients has been established (Moseley, 2004a, 2005b).

In this study we aimed to explore the distribution of cerebral activations in CRPS-1 patients with tonic dystonia during both motor execution and imagining of movement, of affected as well as unaffected limbs. Based on the clinical resemblance with other forms of dystonia, we hypothesised to find a functional alteration in regions supporting a primary motor function, and in circuitry associated with higher-order motor control, particularly in the parietal cortex.

2. Subjects and methods

2.1. Subjects

Eight patients (seven female, mean age 46.4, SD 6.0 years) with CRPS-1 related dystonia from the Leiden University Medical Center and 17 healthy volunteers (15 female, age 42.9, SD 9.2 years), matched for age, were studied. All patients fulfilled the officially accepted diagnostic criteria for CRPS-1 of the IASP (Merskey and Bogduk, 1994). Revisions of these diagnostic criteria are under consideration (Harden and Bruehl, 2006). Furthermore, the presence of tonic dystonia in at least the right upper extremity was obligatory. The presence of dystonia in the CRPS-affected limb was based on the presence of prolonged muscle contractions resulting in abnormal postures or movements of the limb. In upper limb dystonia such involuntary muscle contractions resulted in typical flexion postures of the hand and fingers. In all patients a tonic stretch reflex was present. Passive stretching of the affected digits induced increased flexor activity. The dystonic limb was also always the CRPS-1-affected limb. In five patients the right leg was also affected, in one the left leg and in one both legs (Table 1). Patients in which the left arm was affected by CRPS-1 were excluded. Right handedness was obligatory and was assessed according to the Dutch Handedness Questionnaire (van Strien, 1992). The ability to perform mental imagery was assessed by the Vividness of Movement Imagery Questionnaire (VMIQ) (Isaac et al., 1986). Further assessments included a neurological examination. The T1 weighted MRI scan did not show pathology. Informed written consent was obtained and the study was approved by the Medical Ethical Committees of the Groningen and Leiden University Medical Centers.

Table 1:
CRPS-1 Patient characteristics

2.2. Tasks

Subjects performed four tasks: execution and imagination of flexion/extension movements of the separate right and left wrist. The movements were performed in a vertical plane and paced at 0.5Hz by a visual stimulus: in response to each stimulus one self-paced extension–flexion cycle was made. The forearms were positioned in pronation on pillows. All tasks were preceded by a rest condition. All conditions had a duration of 30s. Three runs were performed lasting 12min each. In each run, 12 response blocks were scheduled. For each subject, tasks were presented in a random, but balanced, order. Subjects were monitored and videotaped during scanning. Special attention was paid to voluntary and involuntary movements of wrists. Prior to the experiment, subjects practised the tasks outside the MRI scanner. They were instructed to imagine moving their wrist freely. Their limbs were not within their field of view. After data acquisition all patients reported pain during the movement execution task. None of the patients reported pain during the imagined movements. Pain was not formally quantified by standardized questionnaires.

2.3. Functional imaging

Subjects were scanned using a 3T Philips MRI scanner (Best, The Netherlands). The following pulse sequence parameters were used: single shot EPI; 46 slices; 3.5mm slice thickness; no gap; 224×224 mm field of view; 64×64 scan matrix; transverse slice orientation; repetition time 3000ms; echo time 35ms; flip angle 90°. Three runs of 240 brain volumes each were acquired, i.e., 10 volumes per 30s condition block. In addition, a T1-weighted whole brain anatomical image was acquired (resolution 1×1×1mm).

2.4. Statistical analysis

Spatial pre-processing and statistical analysis (random effects) was carried out using Statistical Parametric Mapping (version SPM2) (Friston et al., 1995). The functional images were realigned, normalised and subsequently smoothed with an isotropic Gaussian filter using an 8mm Full Width-Half Maximum Gaussian kernal. Head movements were more frequently seen in the patient group. This reflects the difficulties the patients had in performing the tasks. Optionally, the estimated head-motion parameters were used as covariates as described by Friston et al. (Friston et al., 1996). Only minor differences were encountered in this comparison, indicating the presence of task-related head movements. However, areas affected by motion did not show overlap with areas activated during the tasks, demonstrating that the group result was not caused by head motion. Note, for the reported results motion correction has been applied. In the first analysis, the movement and imagining conditions of each hand were compared with rest for each single subject (first level analysis). For within-group analysis, for the patient and control group, respectively, these contrasts were tested using a one-sample T-test (second level analysis). For the larger control group (17 subjects), a high threshold at voxel level was used (P<0.05 Family-wise Error (FWE); extent threshold ≥5 voxels). This avoided overlap of activations which would have blurred regional identification when thresholded at P<0.001 (uncorrected). For patients (8 subjects) a threshold at P<0.001 (uncorrected) was used. This was done as to look for the spatial distribution of a circuitry with general resemblance to that observed in controls. For a between-group analysis (i.e., patients compared to controls) a two-sample T-test was used (P<0.001 (uncorrected)). Resulting clusters for both tests were considered significant at P<0.05 (cluster-level corrected for whole brain volume).

3. Results

Findings on the VMIQ questionnaire (patients 110±52, control subjects 83±34; two-sample T-test, 2-tailed P=0.133) revealed no significant differences.

In both controls and patients, activated areas during imagining showed a symmetrical distribution and activation of the sensorimotor cortex was not observed. Activation in the patient group appeared to be less robust than in the controls for the four different tasks (Fig. 1 and Table 2). By visual inspection, it already was apparent that the patients’ activation networks were most affected during imagining of motor performance with the affected hand. But direct comparison of the figures only allows observation of a trend, partly because of differences in population size (17 controls versus 8 patients). The statistical significance of differences between the two groups can only be established by formal between-group analysis. Such statistically significant differences between patients and controls were only obtained during imagining right hand movements. No statistically significant between-group differences were obtained for the three other tasks.

Fig. 1:
Areas of activation in controls (n = 17) and patients (n = 8). Four different tasks compared with rest condition, projected on a template rendered brain image. P uncorrected <0.01, extent threshold ≥5 voxels.
Table 2:
Areas activated during motor execution and imagining in controls and patients, compared with rest

During imagining of right hand movements three regions were significantly less activated in patients compared with controls. No increases were seen in the patients during this task. These foci of reduced activation were distributed ipsilateral over both the middle frontal gyrus, comprising the prefrontal cortex and premotor cortex (P corrected-cluster-level 0.030, cluster size 186 voxels), and the anterior part of the insular cortex adjoining the superior temporal gyrus and the inferior frontal gyrus (P corrected-cluster-level 0.010, cluster size 242 voxels). In the hemisphere contralateral to the imaginary moving hand, the postcentral gyrus and inferior parietal cortex were less activated in the patient group (P corrected-cluster-level 0.030, cluster size 186 voxels), (Fig. 2 and Table 3).

Fig. 2:
Areas with reduced activation in patients compared with controls during imagining of movement of the right hand. Areas projected on glass brain and transversal slices of a mean template image. P < 0.05 (volume corrected, cluster-level). (1) Ipsilateral middle frontal gyrus and premotor cortex, (2) ipsilateral anterior part of insular cortex, superior temporal gyrus and inferior frontal gyrus, (3) contralateral inferior parietal cortex and postcentral gyrus.
Table 3:
Areas significantly less activated in patients than in controls during imagining of right hand movements

4. Discussion

Both execution and imagining of the motor task in our healthy volunteers result in activation of previously described cerebral circuitry (Deiber et al., 1998; Hanakawa et al., 2003; Ehrsson et al., 2003). In CRPS-1 patients, however, the motor imagining paradigm revealed a conspicuously reduced activation of cortical networks that subserve imagining of movement of the affected limb, in comparison with controls. Only two other imaging studies have reported an imagining paradigm in dystonia patients. In a preliminary PET study idiopathic torsion dystonia patients showed a similar activation pattern as controls (Ceballos-Baumann et al., 1994). In a recent fMRI study on post-stroke dystonia, mental representations of movement of the affected hand resulted in overactivity of the ipsilateral inferior parietal cortex, insula, bilateral prefrontal cortex and other cortical areas (Lehericy et al., 2004). Part of these regions are also involved in our CRPS-1 patients, although we found underactivity. This difference may be due to differences in the underlying pathophysiology or the application of different scanning paradigms.

4.1. Imagining of movement

All patients recalled the ability to perform the imagining task with their affected limb. Moreover, the ability of patients for imagining movement in general did not significantly differ from the controls, as we inferred from the VMIQ questionnaire. This demonstrated the ability of the patients to perform a movement imagery task, although we have no behavioural characteristics to answer the question to what extent imagining movement of the affected hand was correct. In CRPS-1 (without dystonia), prolonged time for imagining moving the affected body part, when compared to the not-affected body part, has previously been explained by a distorted cortical correlate of the body scheme (Schwoebel et al., 2002; Moseley, 2004c). Such a distorted body image has also been inferred from referred sensations (McCabe et al., 2003), mislocalization of tactile stimulation (Maihofner et al., 2006b) and the experience of an increased size of the affected limb (Moseley, 2005a).

The aforementioned behavioural findings are consistent with the idea that cortical reorganization plays an important role in CRPS-1 (Ribbers et al., 2002; Maihofner et al., 2003). We think it likely that changes in body scheme, associated with changes in cortical representation, play an important role in our results. By using sensory stimuli, cerebral reorganization in CRPS-1 has also been demonstrated to occur in the nociceptive (or sensory) domain as well as in cognitive and motor domains (Maihofner et al., 2005; Pleger et al., 2006; Maihofner et al., 2006a). A recently published case-report even mentioned a temporal increase in pain and swelling of the affected hand in a CRPS-patient after imagined movements (Moseley, 2004b). Although this was not mentioned by our patients and the report was limited to one patient, it implies that symptoms of CRPS-1 may be mediated by cortical mechanisms associated with (imagined) movement of the affected body part.

In order to comprehend the specific distribution of the regional decreases in our movement execution and movement imagining study, it is important to consider how the functions of these regions may provide a logical combination in the context of pain and disturbed motor control. The premotor cortex is related to planning and organization of movement (Hoshi and Tanji, 2004). Both impaired activation and enhanced activation in the premotor cortex during a motor task in different forms of dystonia have been reported (for an overview, see Dresel et al., 2006). The inferior parietal cortex is an association area, which receives information of different sensory modalities and thus holds a strategic position in processing space perception, body scheme and so in linking sensation to motor control (Mesulam, 1998). The posterior part of the sensory cortex which constituted the anterior border of the parietal activation in our group is particularly involved in the proprioceptic sensation related to limb movement (de Jong et al., 2002; Naito et al., 2005).

The posterior insula is known as a secondary motor area. For the anterior part of the insula, however, a relation with motor control is less obvious (Flynn et al., 1999). The latter has a dominant role in autonomic regulation. The paralimbic cortex in the anterior superior temporal sulcus, along with anterior insula and orbitofrontal cortex, has been proposed to provide an interface between limbic cortex in the medial temporal lobes and frontoparietal association cortices (Mesulam, 1998). The anterior part of the insula and superior temporal sulcus are situated between the posterior cortex and frontal cortices, thus linking multimodal perception of stimuli and executive processes (Laurens et al., 2005). The anterior insula is involved in pathways that are critical for mental processing of pain related experiences in patients with an amputated hand during imagining of painful finger movements (Rosen et al., 2001) and in a pin-prick hyperalgesia study in CRPS patients (Maihofner et al., 2005).

In view of the long disease duration (mean 11 years) of our patients, long-lasting pain may have induced changes on efficient motor control, reflected by a decreased activation in circuitry providing limbic access to higher-order motor control.

4.2. Execution of movement

Motor execution of the affected side in patients showed only nonsignificant differences between patients and controls. At first sight, this may look strange. On the other hand, the between-group comparison may not have reached statistical significance as a consequence of higher intersubject variability in the activated areas during execution. Such variability may be caused by differences in effort to move. Our most plausible explanation is the altered sensory feedback in the patients, as they experience pain when performing the movement task. This sensory feedback may result in activation of cortical areas, scaling down the areas activated in relation with the execution of movement. However, the fact that we did not find a consistently increased activation in pain-associated circuitry makes us reluctant to provide a definitive explanation in this matter.

Findings on motor execution in our patients with dystonia of CRPS-1 are different from those found in other causes of dystonia, although studies on dystonia, in general, have reported contradictory findings. This can often be attributed to methodological differences, but a different pathophysiology for different forms of dystonia may also be an option. In the case of CRPS-1, a different pathophysiology of dystonia is likely.

In conclusion, patients with CRPS-1 and dystonia displayed areas with decreased activation during imagining tasks that are involved in planning of movement, multimodal sensorimotor integration, autonomic function and pain. Pain may profoundly alter the cerebral organization of movement by functional interaction between these regions.


Remco Renken received a personal grant from the Hazewinkel-Beringer foundation.

The first author had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.


Ceballos-Baumann AO, Marsden CD, Passingham RE, Stephan KM, Frackowiak RSJ, Brooks DJ. Cerebral activation with performing and imagining movement in idiopathic torsion dystonia (ITD): a PET study. Neurology. 1994;44(Suppl. 2):A338.
de Jong BM, Leenders KL, Paans AM. Right parieto-premotor activation related to limb-independent antiphase movement. Cereb Cortex. 2002;12:1213-1217.
Deiber MP, Ibanez V, Honda M, Sadato N, Raman R, Hallett M. Cerebral processes related to visuomotor imagery and generation of simple finger movements studied with positron emission tomography. Neuroimage. 1998;7:73-85.
Detante O, Vercueil L, Thobois S, Broussolle E, Costes N, Lavenne F, et al. Globus pallidus internus stimulation in primary generalized dystonia: a H215O PET study. Brain. 2004;127:1899-1908.
Dresel C, Haslinger B, Castrop F, Wohlschlaeger AM, Ceballos-Baumann AO. Silent event-related fMRI reveals deficient motor and enhanced somatosensory activation in orofacial dystonia. Brain. 2006;129:36-46.
Ehrsson HH, Geyer S, Naito E. Imagery of voluntary movement of fingers, toes, and tongue activates corresponding body-part-specific motor representations. J Neurophysiol. 2003;90:3304-3316.
Fahn S, Bressman SB, Marsden CD. Classification of dystonia. Adv Neurol. 1998;78:1-10.
Flynn FG, Benson DF, Ardila A. Anatomy of the insula – functional and clinical correlates. Aphasiology. 1999;13:55-78.
Friston KJ, Holmes AP, Worsley KJ, Poline JB, Frith C, Frackowiak RSJ. Statistical Parametric Maps in functional imaging: a general linear approach. Hum Brain Mapp. 1995;2:189-210.
Friston KJ, Williams S, Howard R, Frackowiak RS, Turner R. Movement-related effects in fMRI time-series. Magn Reson Med. 1996;35:346-355.
Geyer HL, Bressman SB. The diagnosis of dystonia. Lancet Neurol. 2006;5:780-790.
Hanakawa T, Immisch I, Toma K, Dimyan MA, Van Gelderen P, Hallett M. Functional properties of brain areas associated with motor execution and imagery. J Neurophysiol. 2003;89:989-1002.
Harden RN, Bruehl SP. Diagnosis of complex regional pain syndrome: signs, symptoms, and new empirically derived diagnostic criteria. Clin J Pain. 2006;22:415-419.
Hoshi E, Tanji J. Functional specialization in dorsal and ventral premotor areas. Prog Brain Res. 2004;143:507-511.
Isaac A, Marks DF, Russel DG. An instrument for assessing imagery of movement: the vividness of movement imagery questionnaire (VMIQ). J Ment Imagery. 1986;10:23-30.
Laurens KR, Kiehl KA, Liddle PF. A supramodal limbic-paralimbic-neocortical network supports goal-directed stimulus processing. Hum Brain Mapp. 2005;24:35-49.
Lehericy S, Gerardin E, Poline JB, Meunier S, Van de Moortele PF, Le Bihan D, et al. Motor execution and imagination networks in post-stroke dystonia. Neuroreport. 2004;15:1887-1890.
Maihofner C, Forster C, Birklein F, Neundorfer B, Handwerker HO. Brain processing during mechanical hyperalgesia in complex regional pain syndrome: a functional MRI study. Pain. 2005;114:93-103.
Maihofner C, Handwerker HO, Birklein F. Functional imaging of allodynia in complex regional pain syndrome. Neurology. 2006a;66:711-717.
Maihofner C, Handwerker HO, Neundorfer B, Birklein F. Patterns of cortical reorganization in complex regional pain syndrome. Neurology. 2003;61:1707-1715.
Maihofner C, Neundorfer B, Birklein F, Handwerker HO. Mislocalization of tactile stimulation in patients with complex regional pain syndrome. J Neurol. 2006b;253:772-779.
McCabe CS, Haigh RC, Halligan PW, Blake DR. Referred sensations in patients with complex regional pain syndrome type 1. Rheumatology (Oxford). 2003;42:1067-1073.
Merskey H, Bogduk N., 1994. Complex regional pain syndromes. In: Merskey H, Bogduk N, editors., Classification of chronic pain. IASP Press, Seattle, pp. 40-43.
Mesulam MM. From sensation to cognition. Brain. 1998;121(Pt 6):1013-1052.
Moseley GL. Graded motor imagery is effective for long-standing complex regional pain syndrome: a randomised controlled trial. Pain. 2004a;108:192-198.
Moseley GL. Imagined movements cause pain and swelling in a patient with complex regional pain syndrome. Neurology. 2004b;62:1644.
Moseley GL. Why do people with complex regional pain syndrome take longer to recognize their affected hand? Neurology. 2004c;62:2182-2186.
Moseley GL. Distorted body image in complex regional pain syndrome. Neurology. 2005a;65:773.
Moseley GL. Is successful rehabilitation of complex regional pain syndrome due to sustained attention to the affected limb? A randomised clinical trial. Pain. 2005b;114:54-61.
Naito E, Roland PE, Grefkes C, Choi HJ, Eickhoff S, Geyer S, et al. Dominance of the right hemisphere and role of area 2 in human kinesthesia. J Neurophysiol. 2005;93:1020-1034.
Playford ED, Passingham RE, Marsden CD, Brooks DJ. Increased activation of frontal areas during arm movement in idiopathic torsion dystonia. Mov Disord. 1998;13:309-318.
Pleger B, Ragert P, Schwenkreis P, Forster AF, Wilimzig C, Dinse H, et al. Patterns of cortical reorganization parallel impaired tactile discrimination and pain intensity in complex regional pain syndrome. Neuroimage. 2006;32:503-510.
Ribbers GM, Mulder T, Geurts AC, den Otter RA. Reflex sympathetic dystrophy of the left hand and motor impairments of the unaffected right hand: impaired central motor processing? Arch Phys Med Rehabil. 2002;83:81-85.
Rosen G, Hugdahl K, Ersland L, Lundervold A, Smievoll AI, Barndon R, et al. Different brain areas activated during imagery of painful and non-painful ‘finger movements’ in a subject with an amputated arm. Neurocase. 2001;7:255-260.
Schwartzman RJ, Alexander GM, Grothusen J. Pathophysiology of complex regional pain syndrome. Expert Rev Neurother. 2006;6:669-681.
Schwartzman RJ, Kerrigan J. The movement disorder of reflex sympathetic dystrophy. Neurology. 1990;40:57-61.
Schwoebel J, Coslett HB, Bradt J, Friedman R, Dileo C. Pain and the body schema: effects of pain severity on mental representations of movement. Neurology. 2002;59:775-777.
van Hilten JJ, Blumberg H, Schwartzman RJ., 2005. Movement disorders and dystrophy pathophysiology and measurement. In: Wilson PR, Stanton-Hicks M, Harden RN, editors., CRPS current diagnosis and therapy. IASP Press, Seattle, pp. 119-137.
van Strien JW. Classification of left- and right handed subjects. Ned Tijdschr Psychol. 1992;47:88-92.
Veldman PH, Reynen HM, Arntz IE, Goris RJ. Signs and symptoms of reflex sympathetic dystrophy: prospective study of 829 patients. Lancet. 1993;342:1012-1016.
Verdugo RJ, Ochoa JL. Abnormal movements in complex regional pain syndrome: assessment of their nature. Muscle Nerve. 2000;23:198-205.

Complex regional pain syndrome type 1; Dystonia; fMRI; Imaging; Movement execution; Movement imagining

© 2008 Lippincott Williams & Wilkins, Inc.