Advertisement

Exploring the Relationships Between Altered Body Perception, Limb Position Sense, and Limb Movement Sense in Complex Regional Pain Syndrome

Open AccessPublished:August 09, 2018DOI:https://doi.org/10.1016/j.jpain.2018.07.008

      Highlights

      • Alterations of senses of limb position and movement are observed in complex regional pain syndrome.
      • These alterations are not related to alterations in perception of the painful limb.
      • These 2 body representations should be assessed separately in rehabilitation.

      Abstract

      Chronic pain is often accompanied by patient-reported distorted body perception and an altered kinesthesia (referring to the senses of limb position and limb movement), but the association between these deficits is unknown. The objectives of this study were to assess body perception and the senses of limb position and limb movement in complex regional pain syndrome (CRPS) and to test whether these variables are related to each other and to pain intensity. Thirteen patients with upper limb CRPS (mean pain intensity, 4.2 ± 2.4 out of 10) and 13 controls were recruited. Body perception was self-reported with a questionnaire, and the senses of limb position (task 1) and of limb movement (task 2) were assessed with a robotic system combined with a 2D virtual reality display. The results showed altered kinesthesia in the patients with CRPS compared with controls (all P < .05). Moreover, in the CRPS group, greater pain intensity was associated with lower performance on task 2 (r = -.60; P < .05). Although alterations in participants’ sense of limb position and limb movement were associated with each other (r = -.70, P < .01), they were not related to the altered body perception (all P > .26). Therefore, the results suggest that kinesthesia and body perception should be considered and evaluated separately in patients with CRPS.

      Perspective

      Senses of limb position and movement rely on sensorimotor integration. Both are altered in complex regional pain syndrome. However, they are not related to the subjective perception of the painful limb, and thus they should be assessed separately in rehabilitation.

      Key words

      Perceiving the size, shape, position, and movement of our limbs is essential to help us interact adequately with our environment. An extensive literature supports the idea that chronic pain conditions are often accompanied by various distortions in body perception, which can include changes in perception of the size, shape, and temperature of the painful limb.
      • Lewis JS
      • Kersten P
      • McPherson KM
      • Taylor GJ
      • Harris N
      • McCabe CS
      • Blake DR
      Wherever is my arm? Impaired upper limb position accuracy in complex regional pain syndrome.
      • Moseley GL
      Distorted body image in complex regional pain syndrome.
      ,
      • Moseley GL
      I can't find it! Distorted body image and tactile dysfunction in patients with chronic back pain.
      • Peltz E
      • Seifert F
      • Lanz S
      • Müller R
      Maihöfner C: Impaired hand size estimation in CRPS.
      ,
      • Valenzuela-Moguillansky C
      Pain and disturbances in body awareness.
      In addition, alterations of the sense of limb position (assessed with the limb in a static posture) have been observed.
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      • Lewis JS
      • Kersten P
      • McPherson KM
      • Taylor GJ
      • Harris N
      • McCabe CS
      • Blake DR
      Wherever is my arm? Impaired upper limb position accuracy in complex regional pain syndrome.
      ,
      • Sheeran L
      • Sparkes V
      • Caterson B
      • Busse-Morris M
      • van Deursen R
      Spinal position sense and trunk muscle activity during sitting and standing in non-specific chronic low back pain: Classification analysis.
      For example, individuals with complex regional pain syndrome (CRPS) have been shown to overestimate the angular position of their painful wrist on active and passive movements, but passive movement elicited the greatest disparity.
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      Moreover, alterations of the sense of position have been associated with the severity of motor deficits.
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      • Hamacher D
      • Hamacher D
      • Krowicki M
      • Schega L
      Gait variability in chronic back pain sufferers with experimentally diminished visual feedback: A pilot study.
      However, only 1 previous study has assessed the sense of limb movement during active movement and found that individuals with chronic low back pain tend to overestimate their trunk flexion compared with pain-free controls.
      • Roosink M
      • McFadyen BJ
      • Hébert LJ
      • Jackson PL
      • Bouyer LJ
      • Mercier C
      Assessing the perception of trunk movements in military personnel with chronic non-specific low back pain using a virtual mirror.
      Importantly, in that study, the flexion-extension movement was continuous, to ensure that judgment relied on a continuous comparison between sensory input and motor output rather than on a comparison of final (static) postures.
      • Roosink M
      • McFadyen BJ
      • Hébert LJ
      • Jackson PL
      • Bouyer LJ
      • Mercier C
      Assessing the perception of trunk movements in military personnel with chronic non-specific low back pain using a virtual mirror.
      The observed alterations of body perception and kinesthesia (ie, the senses of limb position and limb movement
      • Proske U
      • Gandevia SC
      The kinaesthetic senses.
      ) in various populations of individuals with chronic pain raises the question of whether these 3 variables are related. Surprisingly, no study to date has assessed such associations in chronic pain. It is particularly relevant to study these potential associations in a CRPS population because in CRPS, pain is accompanied by sensorimotor and autonomic dysfunction, abnormal body perception,
      • Lewis JS
      • Kersten P
      • McCabe CS
      • McPherson KM
      • Blake DR
      Body perception disturbance: A contribution to pain in complex regional pain syndrome (CRPS).
      and altered sense of limb position.
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      • Lewis JS
      • Kersten P
      • McPherson KM
      • Taylor GJ
      • Harris N
      • McCabe CS
      • Blake DR
      Wherever is my arm? Impaired upper limb position accuracy in complex regional pain syndrome.
      However, to the best of our knowledge, no study has assessed the sense of limb movement during active movement in this population, despite the fact that motor deficits are frequently observed.
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      • Maihöfner C
      • Baron R
      • DeCol R
      • Binder A
      • Birklein F
      • Deuschl G
      • Handwerker HO
      • Schattschneider J
      The motor system shows adaptive changes in complex regional pain syndrome.
      ,
      • Schilder JC
      • Schouten AC
      • Perez RS,
      • Huygen FJ,
      • Dahan A
      • Noldus LP
      • van Hilten JJ
      • Marinus J
      Motor control in complex regional pain syndrome: A kinematic analysis.
      Therefore, the aim of this study was to assess body perception and kinesthesia in individuals with CRPS and to assess whether body perception, sense of limb position, and sense of limb movement are related to one another.
      Our first specific objective was to assess the sense of limb position, sense of limb movement during active movement, and body perception in patients with CPRS. We hypothesised that the senses of limb position and movement would be altered compared to controls and to normative data. Body perception was self-reported using the Bath Body Perception Disturbances Scale
      • Lewis JS
      • McCabe CS
      Correcting the body in mind: body perception disturbance in complex regional pain syndrome (CRPS) and rehabilitation approaches.
      and the senses of limb position and movement were objectively measured using a robotic system combined with virtual reality. The second specific objective was to test whether these 3 variables are related to one another and to pain intensity.

      Methods

      Participants

      Thirteen patients with unilateral CRPS (diagnosed according to the Budapest clinical criteria
      • Harden RN
      • Bruehl S
      • Perez RS
      • Birklein F
      • Marinus J
      • Maihofner C
      • Lubenow T
      • Buvanendran A
      • Mackey S
      • Graciosa J
      • Mogilevski M
      • Ramsden C
      • Chont M
      • Vatine JJ
      Validation of proposed diagnostic criteria (the “Budapest criteria”) for Complex Regional Pain Syndrome.
      by an anesthesiologist at the Center of Expertise in Chronic Pain, Quebec) and 13 healthy controls matched for sex, age, and self-reported laterality were recruited over a 1-year period in the Quebec City area. CRPS participants were recruited from the outpatient clinic at the Center of Expertise in Chronic Pain. Controls were recruited from Laval University. Both CRPS type 1 and type 2 (referring, respectively, to the absence or presence of a peripheral nerve injury) were included, given that in the Budapest criteria, the clinical diagnosis is similar for both types, and the clinical utility of these subgroups is controversial.
      • Harden RN
      • Bruehl S
      • Perez RS
      • Birklein F
      • Marinus J
      • Maihofner C
      • Lubenow T
      • Buvanendran A
      • Mackey S
      • Graciosa J
      • Mogilevski M
      • Ramsden C
      • Chont M
      • Vatine JJ
      Validation of proposed diagnostic criteria (the “Budapest criteria”) for Complex Regional Pain Syndrome.
      Participants were excluded if they had any motor impairment interfering with the task performance, which necessitated 80° shoulder abduction and forward movements with an amplitude of 20 cm and the weight of the arm fully supported. Exclusion criteria for controls were the presence of acute upper limb (UL) pain in the last 3 months or of chronic UL pain in the last year. Finally, the presence of noncorrected visual impairments was an exclusion criterion for both groups. Two patients with CRPS were excluded from the study due to motor impairments and noncorrected visual deficits.
      All participants provided written informed consent before enrollment. This study was approved by the local Ethical Review Board (Institut de réadaptation en déficience physique de Québec, no. 2014-395) and conformed with the Declaration of Helsinki.
      In the CRPS group, a brief history of each patient's condition, including information about the circumstances and the timing of CRPS onset, pain manifestations, pain treatments (pharmaceutical and nonpharmaceutical), and comorbidities, was obtained from a semistructured interview. Patients were asked to indicate the anatomic location of their pain and to rate their pain intensity over the last 24 hours on an 11-point numerical rating scale ranging from 0 = no pain to 10 = worst pain imaginable.

      Procedure

      The CRPS patients and controls participated in a single session lasting approximately 2 hours. For the CRPS group, each session began with as assessment of body perception using a questionnaire. The sense of limb position (task 1) and the sense of limb movement (task 2) were assessed successively in all groups with the KINARM Exoskeleton Lab (BKIN Technologies, Kingston, ON, Canada), a robotized bilateral exoskeleton allowing movements of the shoulder (horizontal abduction-adduction) and the elbow (flexion-extension) joints to move participants’ ULs in the horizontal plane (Fig. 1). In task 1, ULs were simply obstructed from view. In task 2, the robot was interfaced with a 2D virtual environment allowing replacement of the participant's UL by a virtual UL (presented with an appropriate perception of depth; Fig. 1B). Joint angular positions for both the shoulder and elbow joints were obtained from KINARM motor encoders and sampled at 1 kHz, and the position of the index finger was computed in real time. Data processing was conducted with MATLAB R2011b (MathWorks, Natick, MA).
      Fig 1
      Figure 1Experimental setup. (A) The robotized exoskeleton is fitted to the anthropometric characteristics of the participant. (B) The 2D virtual environment involves the projection of virtual upper limbs on a mirror (47’’) by a television. The upper limbs are fully supported by the exoskeleton and are hidden from the participant's view.

      Task 1: Sense of Limb Position

      This task assessed the sense of limb position at rest. The robot passively moved a UL to 1 of 4 predefined positions in the ipsilateral hemispace. The participant then had to reproduce the position with the other UL (ie, the second UL being the mirror image of the first UL). Both ULs were obstructed from the participant's view (Fig. 3A).
      This task is a KINARM Standard test (arm position matching task
      • Scott S
      • Brown I
      Method and apparatus for assessing proprioceptive function.
      ) that has been used in a previous study to assess alterations in the sense of limb position in stroke patients.
      • Dukelow SP
      • Herter TM
      • Moore KD
      • Demers MJ
      • Glasgow JI
      • Bagg SD
      • Norman KE
      • Scott SH
      Quantitative assessment of limb position sense following stroke.
      The 4 (nonvisible) targets are spread on a 2 × 2 grid at 20-cm intervals on the ipsilateral hemispace. Each position is repeated 6 times in a pseudorandomized order (total of 24 trials). Both ULs were tested in all participants in a random order. The precision of the robot (position error = 1.5 mm) and reliability of the measures for this task (intraclass coefficient correlation = .86, P < .00001
      • Dukelow SP
      • Herter TM
      • Moore KD
      • Demers MJ
      • Glasgow JI
      • Bagg SD
      • Norman KE
      • Scott SH
      Quantitative assessment of limb position sense following stroke.
      ) were very good.

      Task 2: Sense of Limb Movement

      Task 2 consisted of judging whether the movement made by a virtual UL (anchored to the participant's UL movement) was greater or smaller than the participant's actual movement. For each trial, the robot moved the UL to a starting position, and then a visual cue indicated to the participant that he or she needed to move his or her hand forward at a comfortable speed (Video 1, Supplementary Material). Across trials, the movements of the virtual UL displayed in real time were smaller, greater, or identical (no scaling) (Fig. 2) to the participants’ actual limb movements. Importantly, the actual and virtual ULs were always aligned at the beginning of a movement, and the virtual UL was disappearing before the end of a movement, constraining the participant to base his or her judgment on the movement and not on the final position. After each trial, the participant reported whether the virtual UL's movement was “greater” or “smaller” than his or her own movement (2-alternative forced choice paradigm).
      Fig 2
      Figure 2Scaling of movements of the virtual upper limb (UL) in task 2. The amplitude (and thus the velocity) of the virtual UL was scaled in real time to appear smaller or greater than the participant's movements. The size and the starting position of the virtual UL (35° for the shoulder and 115° for the elbow) were similar in all conditions. Participants were seeing the virtual UL exclusively. Blue bars, red dots, and green dots denote the actual position of the participant's UL, the position of the elbow joint, and the computed position of the index, respectively.
      For this task, only the painful limb was tested in the CRPS group, because testing both arms took too long and was too tiring for these patients. However, both arms were tested in the controls, in random order.
      Familiarization Trials
      Before the experimental tasks described above, a 2-step familiarization procedure was performed. First, participants practiced forward movements at a comfortable speed without scaling of the virtual UL (5 trials). Then the movement of the virtual UL was augmented 2.75 times (.364 for smaller movements), and each trial was repeated twice. These scaling factors were greater than those used in the experimental trials and were used to make sure that the participant understood the task correctly.
      Experimental Trials
      The participants were exposed to 13 scaling factors. The scaling was applied to the angular rotation of the elbow and the shoulder joints. The scaling factors ranged from 1.25 to 2.5 times the rotation angle of the elbow and shoulder joints (ie, 1.25, 1.5, 1.75, 2, 2.25, and 2.5) for the larger movements. The inverses of these factors (eg, 1/1.5 = .667) were used for smaller movements. For the identical condition, the scaling was set to 1. Each scaling factor was repeated 8 times, for a total of 104 trials (performed in a pseudorandomized order). No feedback on performance was provided, to minimize potential learning.

      Assessment of Body Perception in the CRPS Group

      Body perception of the affected UL in CRPS participants was assessed using the French version of the Bath Body Perception Disturbances Scale.
      • Lewis JS
      • McCabe CS
      Correcting the body in mind: body perception disturbance in complex regional pain syndrome (CRPS) and rehabilitation approaches.
      This scale, which was specifically developed for a CRPS population based on a qualitative study assessing perceptual abnormalities of the painful limb,
      • Lewis JS
      • Kersten P
      • McCabe CS
      • McPherson KM
      • Blake DR
      Body perception disturbance: A contribution to pain in complex regional pain syndrome (CRPS).
      includes questions regarding the sense of disowning the body part, impairment of the perceived limb position, and attention to and hostile feelings about the painful limb (Fig. 5A). The question about the desire to amputate the painful limb and the drawing part of the questionnaire were not included. Participants were required to rate 4 questions from 0 to 10 on a Likert scale (Fig. 5A) and to provide 4 “yes”/”no” responses in the fifth question to assess perturbations in the perception of the painful limb about size, temperature, pressure, and weight. The English version has been shown to have a variable consistency ranging from poor
      • Bean DJ
      • Johnson MH
      • Heiss-Dunlop W
      • Kydd RR
      Extent of recovery in the first 12 months of complex regional pain syndrome type-1: A prospective study.
      to good
      • Bean DJ
      • Johnson MH
      • Heiss-Dunlop W
      • Kydd RR
      Extent of recovery in the first 12 months of complex regional pain syndrome type-1: A prospective study.
      • Lewis JS
      • Schweinhardt P
      Perceptions of the painful body: The relationship between body perception disturbance, pain and tactile discrimination in complex regional pain syndrome.
      and adequate interrater reliability.
      • Lewis JS
      • Schweinhardt P
      Perceptions of the painful body: The relationship between body perception disturbance, pain and tactile discrimination in complex regional pain syndrome.
      Translation of the questionnaire was performed using a forward and backward translation, and the translated version was reviewed by an expert committee. The French version has not yet been validated, however.

      Outcome Measures

      Task 1
      Mean absolute distance errors in the x- and y-axes across trials were obtained using Dexterit-E software (arm position matching task
      • Scott S
      • Brown I
      Method and apparatus for assessing proprioceptive function.
      , version 3.4.2).
      Task 2
      The sense of limb movement was evaluated as described by Roosink et al.
      • Roosink M
      • McFadyen BJ
      • Hébert LJ
      • Jackson PL
      • Bouyer LJ
      • Mercier C
      Assessing the perception of trunk movements in military personnel with chronic non-specific low back pain using a virtual mirror.
      • Roosink M
      • Robitaille N
      • McFadyen BJ
      • Hébert LJ
      • Jackson PL
      • Bouyer LJ
      • Mercier C
      Real-time modulation of visual feedback on human full-body movements in a virtual mirror: Development and proof-of-concept.
      First, results obtained with the 2-alternative forced choice paradigm (greater = 1, smaller = 0) were plotted against the 13 scaling factors (log-transformed to be on a linear scale). Second, a sigmoid curve (1), with initial values XY.50 = 0, constraints YMAX = 1 and YMIN = 0, and a variable slope (m), was fitted to the data using Prism 6 for Windows (GraphPad Software, La Jolla, CA):
      Y=Y(MIN)+Y(MAX)Y(MIN)1+10(X(Y0.50)X)m.
      (1)


      Finally, 3 data points were interpolated from each curve (XY0.25, XY0.50, and XY0.75) and used to determine the point of subjective equivalence (PSE) (2) and the just noticeable difference (JND) (3):
      PSE=XY0.50
      (2)


      and
      JND=X(Y0.75)X(Y0.25)2.
      (3)


      The PSE corresponds to the scaling factors for which the participant equally answered that the virtual UL movement was “smaller” or “greater”. In theory, the PSE is equal to 0; that is, there is a 50% chance of responding smaller or greater when in fact no scaling has been applied. When the PSE is 0, there is no alteration of the sense of limb movement. A PSE >0 indicates that the participant overestimates his or her own movement, whereas a PSE <0 indicates that the participant underestimates his or her own movement. The JND refers to the ability to discriminate between different levels of scaled feedback. The higher the slope and the smaller the JND, the better a participant can discriminate between different levels of scaling factors.
      The percentage of accurate responses was also calculated for each level of scaling, except for trials from the identical condition (ie, no scaling). Measuring accuracy may appear redundant, considering that the PSE and JND are included as variables; however, in participants with a very low percentage of accurate responses, it can be impossible to fit a psychophysical curve and thus to calculate the PSE and JND. Therefore, the percentage of accurate responses allows analysis of all participants’ data, including the most impaired participants.
      Finally, the mean velocity between the beginning of the movement and the disappearance of the virtual UL was calculated for each participant in each trial. Because participants were instructed to perform the task at a comfortable speed, the patients with CRPS could have moved more slowly than controls owing to pain or fear of movement. Because movement velocity could be a confounding variable, this aspect was controlled for in statistical analyses.
      Body Perception Disturbances Scale
      A total score for the first 4 questions was computed corresponding to the sum of their numerical Likert ratings, and then the total sum of the 4 responses in the fifth question (yes/no questions: yes, 1; no, 0) was added. A higher score indicates greater disturbances in body perception, with 44 being the maximum possible score.

      Statistical Analysis

      Results are reported as mean ± SD or as median (range). The independent t test (2-tailed) was used to compare groups for age. Descriptive analyses (mean ± standard deviation) were used for the Body Perception Disturbances Scale.
      Errors in task 1 were analyzed using 2 [group (CRPS or control)] × 2 [error direction (x-axis or y-axis)] repeated-measures analyses of variance. Individual data from each participant were compared with age- and sex-matched normative data available for that KINARM Standard test (based on a group of 461 healthy participants, including 214 males and 247 females age 18–93 years).
      For task 2, analyses of covariance with the movement velocity as a covariate were performed to determine the percentage of accurate responses, PSE, and JND. Pearson coefficients were used to test whether body perception and the senses of limb position and limb movement were related with one another and with pain intensity. Statistical analyses were performed with R version 3.1.2 (R Institute for Statistical Computing, Vienna, Austria).

      Results

      Population

      The CRPS and control groups were similar in terms of mean age (CRPS, 56.1 ± 9.2 years; controls, 50.8 ± 13.8; P = .31), sex distribution (10 females in each group), and distribution of laterality (11 right-handed in each group). For the CRPS group, clinical characteristics and results (Body Perception Disturbances Scale score, sense of limb position, and sense of limb movement) for each participant are reported in Table 1. All patients in the CRPS group was receiving analgesics, and all but 1 (patient 4) were receiving physiotherapy and occupational therapy.
      Table 1Clinical Characteristics and Results of the CRPS Group
      PatientClinical characteristicsResults
      CRPS subtypeHandednessAffected sideEtiologyTime since diagnosis (months)Pain intensityBody perception, total BPDSSense of limb position, mean errorSense of limb movement
      Accurate response, %PSEJND
      1Type IIRightNDNerve compression725224.792.08.091
      2Type IRightNDHand surgery236.5104.9
      Asterisks indicate participants who obtained abnormal scores compared with age and sex-matched controls.4,32
      53
      3Type IRightNDFall37253.283.10.17
      4Type IRightDHand surgery33113.479-.01
      5Type IRightNDWrist surgery5799.2
      Asterisks indicate participants who obtained abnormal scores compared with age and sex-matched controls.4,32
      26
      6Type ILeftNDHand surgery44227.1
      Asterisks indicate participants who obtained abnormal scores compared with age and sex-matched controls.4,32
      80-.01.19
      7Type IRightDFall30
      One patient did not report pain during the last 24 hours but typically experienced pain and reported a pain level of 4/10 by the end of the experience.
      83.992.03.09
      8Type IRightDWrist surgery7.52132.386.16.08
      9Type IRightNDWrist fracture153163.577.15.22
      10Type IRightDFall44293.779.17.16
      11Type IRightDHand surgery1.51.5124.2
      Asterisks indicate participants who obtained abnormal scores compared with age and sex-matched controls.4,32
      91.11.09
      12Type ILeftDHand fracture48185.3
      Asterisks indicate participants who obtained abnormal scores compared with age and sex-matched controls.4,32
      91-.09.10
      13Type IRightDFall43171.895-.05.03
      Mean ± SD12.7 ± 20.64.2 ± 2.416.3 ± 6.64.4 ± 2.178.7 ± 19.2.056 ± .09.12 ± .06
      Abbreviations: CRPS, complex regional pain syndrome; BPDS, Body Perception Disturbances Scale; PSE, point of subjective equivalence; JND, just noticeable difference; ND, nondominant; D, dominant; SD, standard deviation.
      low asterisk Asterisks indicate participants who obtained abnormal scores compared with age and sex-matched controls.
      • Dukelow SP
      • Herter TM
      • Moore KD
      • Demers MJ
      • Glasgow JI
      • Bagg SD
      • Norman KE
      • Scott SH
      Quantitative assessment of limb position sense following stroke.
      • Scott S
      • Brown I
      Method and apparatus for assessing proprioceptive function.
      One patient did not report pain during the last 24 hours but typically experienced pain and reported a pain level of 4/10 by the end of the experience.

      Task 1: Sense of Limb Position

      Because there was no statistical difference between the dominant and the nondominant arms in the controls (t(12) = .89, P = .40) and between the painful and nonpainful ULs in the CRPS group (t(12) = 1.5, P = .15), statistical analyses were performed on the mean of both arms. Although this absence of difference for CRPS patients might be surprising at first sight, it needs to be kept in mind that this task was bilateral, with 1 arm matching the position of the other.
      Fig. 3A shows examples of the performance of 2 representative participants from each group. Fig. 3B shows the mean errors on the x- and y-axes for each group. Errors were found to be larger in the CRPS group compared with controls (mean, 4.4 ± 1.9 cm vs 3.1 ± 1.7 cm, F(1,24) = 5.1, P = .03, ŋp = .19) and to be larger on the x-axis compared with the y-axis (4.6 ± 1.8 cm vs 2.8 ± 1.5 cm, F(1,24) = 54.4, P < .001, ŋp = .68). However, no significant interaction between the group and the error direction was observed (F(1,24) = .46, P = .51).
      Fig 3
      Figure 3Errors in task 1. (A) Individual data for 2 representative participants from each group. The green squares represent the position of 1 upper limb (UL) moved passively by the robot, and the superimposed blue dashed squares represent the matching positions with the contralateral UL. (B) Mean errors for the x- and y-axes for the CRPS and control groups. Error bars represent the standard error of the mean.
      The errors in the x- and y-axes were also compared with normative data (age and sex-matched controls
      • Dukelow SP
      • Herter TM
      • Moore KD
      • Demers MJ
      • Glasgow JI
      • Bagg SD
      • Norman KE
      • Scott SH
      Quantitative assessment of limb position sense following stroke.
      • Scott S
      • Brown I
      Method and apparatus for assessing proprioceptive function.
      ) for each participant in the CRPS and control groups. Abnormal scores were recorded in 5 patients in the CRPS group, but in no controls.

      Task 2: Sense of Limb Movement

      In task 2, for the control group, a significant difference was found between the dominant and the nondominant UL for the percentage of accurate responses (P = .04) and for the JND (P = .006). No statistical difference was observed for the PSE (P = .17). Therefore, statistical analyses were performed on only 1 UL for the control group. The UL included in the analysis for each control participant was selected to maintain a comparable proportion of dominant and nondominant ULs in both groups. In the CRPS group, 7 patients had CRPS in the dominant limb, and the other 6 had CRPS in the nondominant limb.
      On average, the patients with CRPS had significantly slower in limb movements than controls (mean, .34 ± .09 m·s−1 vs .45 ± .17 m·s−1, t(24) = 2.08, P = .047). The movement velocity influenced neither the PSE (F(1,19) = 1.1, P = .30) nor the percentage of accurate responses (F(1,19) = 1.9, P = .17); however, it was positively associated with the JND (F(1,19) = 8.6, P = .009, ŋp = .31), meaning that it was more difficult to discriminate between levels of scaling when the movement was fast. To control for the potential impact of movement velocity on outcomes, this was included as a covariate in the statistical analyses.
      Percentage of Accurate Responses
      A trend toward a difference in response accuracy between groups was observed (F(1,21) = 4.15, P = .055, ŋp = .18). The mean percentage of accurate responses was 78 ± 20% in the CRPS group and 88 ± 8% in the control group.
      PSE and JND
      For the CRPS group, 2 participants had to be excluded from these analyses because the low percentage of their accurate responses (26% and 53%) precluded fitting the psychophysical curve. Interestingly, these 2 participants scored outside of the normative values in task 1. As shown by the psychophysical curves in Fig. 4, the mean PSE was similar for the CRPS and control groups (.05 ± .09 vs .04 ± .07, F(1,21) = .09, P = .76); however, the JND was higher (ie, the slope was lower) for the CRPS group compared with the controls (.12 ± .05 vs .11 ± .06, F(1,21) = 5.12, P = .035, ŋp = .21), demonst-rating impaired ability to discriminate between different levels of scaling in the CRPS group.
      Fig 4
      Figure 4Grand average psychophysical curves for the CRPS (red line and dots) and control (black line and crosses) groups. The blue line indicates X0.50 (point of subjective equivalence), and the green lines indicate XY0.25 and XY0.75. The solid line represents the X points for the CRPS group, and the dashed line represents the X point for the controls. The PSE corresponds to the XY0.50 points, and the JND corresponds the difference between the XY0.75 and XY0.25 points divided by 2.

      Body Perception (Body Perception Disturbance Scale)

      As shown in Fig. 5B, there was a high variability between CRPS participants for the first 4 questions of the Body Perception Disturbance Scale. For the fifth question (focusing on perceptual changes), participants reported changes in the size (n = 6), temperature (n = 9), pressure (n = 8), and weight (n = 6) of the painful limb. The total score of the questionnaire was 16.3 ± 6.6. Note, however, that a posteriori analysis showed low internal consistency of the body perception disturbances scale in our study (Cronbach's α = .53, 95% confidence interval = .23–.84).
      Fig 5
      Figure 5Body Perception Disturbance Scale. (A) The 4 questions of the Body Perception Disturbance Scale. (b): Individual data for the CRPS group. Each dot corresponds to a CRPS participant. The vertical red bars represent the mean values of the CRPS group.

      Correlation Analyses

      The correlation coefficients between body perception, the senses of limb position and movement, and pain intensity for the CRPS and the control groups are presented in Table 2. A lower percentage of accurate responses in task 2 (sense of limb movement) was strongly associated with a poorer ability to discriminate between different levels of scaling factors (JND task 2; r = -.94, P < .0001) and with larger errors in task 1 (sense of limb position; r = -.71; P = .006). As shown in Table 2, similar associations were also found to be significant in controls (r = -.63, P = .018 and r = -.94, P < .0001, respectively). Moreover, greater pain intensity in the CRPS group was associated with a lower percentage of accurate responses in task 2 (sense of limb movement; r = -.60, P = .027). However, no significant correlations were observed between body perception and the senses of limb position and movement (see Table 2 for Pearson's coefficients and P values).
      Table 2Pearson's Coefficients Between Body Perception, Senses of Limb Position and Movement, and Pain Level in the CRPS Group
      Sense of limb positionSense of limb movementBody perception
      Mean errorAccurate responsesPSEJNDTotal score BPDS
      Control group
      Mean error-.63.15.57
      (P = .018)(P = .63)(P = .069)
      Accurate response-.71.03-.92
      (P = .006)(P = .92)(P < .0001)
      PSE-.30-.40-.19
      (P = .36)(P = .22)(P = .56)
      JND.32 )-.94.28
      (P = .33(P < .0001)(P = .39)
      BPDS-.15.30.21.27
      (P = .61)(P = .26)(P = .53)(P = .41)
      Pain intensity.47-.60-.28.15.30
      (P = .15)(P = .027)(P = .85)(P = .47)(P = .31)
      CRPS group
      Abbreviations: BPDS, Body Perception Disturbance Scale; PSE, point of subjective equivalence; JND, just noticeable difference; CRPS, complex regional pain syndrome.
      Note. Bold type indicates a significant correlation.

      Discussion

      Whereas previous CRPS studies have focused primarily on body perception and the sense of limb position, the 2 novel aims of this study were to investigate the sense of limb movement during active movement in CRPS and to identify any associations between body perception, the senses of limb movement and limb position, and pain in the CRPS affected limb. Our results show that the senses of limb position and limb movement are altered in patients with CRPS compared with healthy controls. Interestingly, although alterations in the senses of limb position and limb movement are associated, they are not related to the participants’ perceptions of their painful limbs, as assessed by the Bath Body Perception Scale. Because task 1 required bilateral movements of the ULs (ie, the painful limb was always involved), and task 2 tested only the painful limb in the CRPS group, our data do not allow us to assess the senses of limb position and limb movement in the nonpainful limb. Thus, our discussion focuses only on our findings in the painful limb.
      In accordance with previous studies showing deficits in proprioception in persons with CRPS,
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      • Lewis JS
      • Kersten P
      • McPherson KM
      • Taylor GJ
      • Harris N
      • McCabe CS
      • Blake DR
      Wherever is my arm? Impaired upper limb position accuracy in complex regional pain syndrome.
      our data also demonstrate that limb position (task 1) is altered in CRPS. The patients with CRPS made more errors than controls in the arm position matching task. Moreover, we report for the first time alterations in the sense of limb movement during active movement (task 2), characterized by a lower percentage of accurate responses and a poorer capacity to discriminate between different levels of scaled feedback in the CRPS group compared with controls. Previous studies have suggested that perception of the body and its movements are biased in the presence of chronic pain—for example, showing that the subjective body midline in CRPS is shifted toward the painful side
      • Sumitani M
      • Shibata M
      • Iwakura T
      • Matsuda Y
      • Sakaue G
      • Inoue T
      • Mashimo T
      • Miyauchi S
      Pathologic pain distorts visuospatial perception.
      or that trunk flexion movements are overestimated in chronic low back pain.
      • Roosink M
      • McFadyen BJ
      • Hébert LJ
      • Jackson PL
      • Bouyer LJ
      • Mercier C
      Assessing the perception of trunk movements in military personnel with chronic non-specific low back pain using a virtual mirror.
      In contrast with these observations, we did not find a perceptive bias in the assessment of kinesthesia in CRPS. Indeed, no interaction between group and error direction was observed in task 1, which does not indicate a bias toward larger errors on the x-axis in the CRPS group relative to controls. Moreover, the CRPS patients did not overestimate their own movements (as measured by the PSE in task 2).
      Interestingly, a lower percentage of accurate responses in task 2 was associated with a higher rate of errors in task 1, suggesting that the observed deficits in the senses of limb position and limb movement in CRPS rely on similar processes. Our results are in line with the idea that body representations are blurred in the presence of pain, as suggested by behavioral and neuroimaging studies reporting alterations in motor and sensory cortical areas in CRPS (for a literature review, see Swart et al
      • Swart CM.
      • Stins JF
      • Beek PJ
      Cortical changes in complex regional pain syndrome (CRPS).
      ). Indeed, referred sensations,
      • Maihöfner C
      • Neundörfer B
      • Birklein F
      • Handwerker HO
      Mislocalization of tactile stimulation in patients with complex regional pain syndrome.
      • McCabe CS
      • Haigh RC
      • Halligan PW
      • Blake DR
      Referred sensations in patients with complex regional pain syndrome type 1.
      alterations in the primary and secondary somatosensory areas,
      • Di Pietro F
      • McAuley JH
      • Parkitny L
      • Lotze M
      • Wand BM
      • Moseley GL
      • Stanton TR
      Primary somatosensory cortex function in complex regional pain syndrome: A systematic review and meta-analysis.
      and expansion of the motor areas
      • Maihöfner C
      • Baron R
      • DeCol R
      • Binder A
      • Birklein F
      • Deuschl G
      • Handwerker HO
      • Schattschneider J
      The motor system shows adaptive changes in complex regional pain syndrome.
      of the painful limb are observed in CRPS. Interestingly, these alterations have been shown to be positively related to pain level,
      • Maihöfner C
      • Baron R
      • DeCol R
      • Binder A
      • Birklein F
      • Deuschl G
      • Handwerker HO
      • Schattschneider J
      The motor system shows adaptive changes in complex regional pain syndrome.
      • Maihöfner C
      • Neundörfer B
      • Birklein F
      • Handwerker HO
      Mislocalization of tactile stimulation in patients with complex regional pain syndrome.
      ,
      • Pleger B
      • Ragert P
      • Schwenkreis P
      • Förster AF
      • Wilimzig C
      • Dinse H
      • Nicolas V
      • Maier C
      • Tegenthoff M
      Patterns of cortical reorganization parallel impaired tactile discrimination and pain intensity in complex regional pain syndrome.
      consistent with our observation that performance in task 2 was negatively associated with pain intensity. Taken together, these results suggest that alterations of sensorimotor cortical areas could explain the deficits in the senses of limb position and limb movement during active movement in CRPS.
      One important new finding from our data is that the alteration of kinesthesia in CRPS was not related to the reported alterations of body perception (eg, perceived changes in the size, temperature, pressure, and weight or changes in feeling of ownership of the painful limb), suggesting that these variables are generated by independent processes. This result can be interpreted in line with the classical model of body representation that suggests that at least 2 distinct and independent body representations govern our motor action: body image and body schema.
      • Paillard J
      Body schema and body image - a double dissociation in deafferented patients.
      • Schwoebel J
      • Coslett HB
      Evidence for multiple, distinct representations of the human body.
      ,
      • de Vignemont F
      Body schema and body image–pros and cons.
      Whereas body image and body schema share similar somatosensory and parietal areas
      • Dijkerman HC
      • de Haan EH
      Somatosensory processes subserving perception and action.
      • Naito E
      • Morita T
      • Amemiya K
      Body representations in the human brain revealed by kinesthetic illusions and their essential contributions to motor control and corporeal awareness.
      ,
      • Pitron V
      • De Vignemont F
      Beyond differences between the body schema and the body image: Insights from body hallucinations.
      and are often impaired in pathological conditions,
      • Pitron V
      • De Vignemont F
      Beyond differences between the body schema and the body image: Insights from body hallucinations.
      they are relatively independent.
      • Dijkerman HC
      • de Haan EH
      Somatosensory processes subserving perception and action.
      • Pitron V
      • De Vignemont F
      Beyond differences between the body schema and the body image: Insights from body hallucinations.
      ,
      • Schwoebel J
      • Coslett HB
      Evidence for multiple, distinct representations of the human body.
      • de Vignemont F
      Body schema and body image–pros and cons.
      Body schema depends on online sensorimotor integration
      • Dijkerman HC
      • de Haan EH
      Somatosensory processes subserving perception and action.
      • de Vignemont F
      Body schema and body image–pros and cons.
      ,

      Wolpert DM, Ghahramani Z, Jordan MI: An internal model for sensorimotor integration. Science 269:1880-1882, 1995

      and has been shown to be altered in CRPS.
      • Schwoebel J
      • Friedman R
      • Duda N
      • Coslett HB
      Pain and the body schema: Evidence for peripheral effects on mental representations of movement.
      Interestingly, the largest deficits in the sense of limb position in CRPS were found to be positively related to the degree of motor deficits.
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      Therefore, we would suggest that the observed alterations in the senses of limb position and limb movement reflect a deficit in body schema in CRPS. On the other hand, body image, which refers to the “conscious awareness of one's own body,”
      • Paillard J
      Body schema and body image - a double dissociation in deafferented patients.
      seems to be related to the insula,
      • Dijkerman HC
      • de Haan EH
      Somatosensory processes subserving perception and action.
      an area involved in emotional processing,
      • Gogolla N
      The insular cortex.
      agency,
      • Farrer C
      • Franck N
      • Georgieff N
      • Frith CD
      • Decety J
      • Jeannerod M
      Modulating the experience of agency: A positron emission tomography study.
      and ownership.
      • Karnath HO
      • Baier B
      • Nägele T
      Awareness of the functioning of one's own limbs mediated by the insular cortex?.
      In our study, alterations of body perception could be a part of a deficit in the body image. In CRPS, disturbances in body perception are characterized by feeling the painful limb as a foreign body part, a pronounced dislike and denial of the painful limb, and a high desire to amputate it.
      • Lewis JS
      • Kersten P
      • McCabe CS
      • McPherson KM
      • Blake DR
      Body perception disturbance: A contribution to pain in complex regional pain syndrome (CRPS).
      Lewis et al
      • Lewis JS
      • Kersten P
      • McCabe CS
      • McPherson KM
      • Blake DR
      Body perception disturbance: A contribution to pain in complex regional pain syndrome (CRPS).
      suggested that such disturbances could interfere with the body schema and consequently with motor control. Although an extensive literature shows that body schema
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      • Lewis JS
      • Kersten P
      • McPherson KM
      • Taylor GJ
      • Harris N
      • McCabe CS
      • Blake DR
      Wherever is my arm? Impaired upper limb position accuracy in complex regional pain syndrome.
      ,
      • Moseley GL
      Why do people with complex regional pain syndrome take longer to recognize their affected hand?.
      • Schwoebel J
      • Friedman R
      • Duda N
      • Coslett HB
      Pain and the body schema: Evidence for peripheral effects on mental representations of movement.
      and body image
      • Lewis JS
      • Kersten P
      • McCabe CS
      • McPherson KM
      • Blake DR
      Body perception disturbance: A contribution to pain in complex regional pain syndrome (CRPS).
      • Moseley GL
      I can't find it! Distorted body image and tactile dysfunction in patients with chronic back pain.
      are altered in CRPS, here we demonstrate for the first time that these alterations might be independent. More work is needed to clarify the mechanistic underpinnings, however.
      Some limitations of the present study need to be highlighted. First, the sense of limb position in task 1 was assessed using both upper limbs (with 1 arm matching the position of the other), which made it difficult to dissociate proprioception in the painful and nonpainful limbs. However, the advantage of this task was access to normative data, which allowed us to compare to a wide group of age- and sex-matched controls.
      • Dukelow SP
      • Herter TM
      • Moore KD
      • Demers MJ
      • Glasgow JI
      • Bagg SD
      • Norman KE
      • Scott SH
      Quantitative assessment of limb position sense following stroke.
      • Scott S
      • Brown I
      Method and apparatus for assessing proprioceptive function.
      Another limitation was that the CRPS participants were significantly slower in task 2 compared with controls; however, ANCOVA was used to statistically control for that difference. Moreover, it is important to note that faster movements were related to lower performance, and thus it is unlikely that the difference in velocity between the groups would explain the difference in performance. Rather, the velocity difference between the groups could have resulted in underestimation of the deficit in the CRPS patients. Furthermore, our ability to measure the performance in the most severely impaired patients was limited by the low percentage of accurate responses in task 2. Two patients were excluded in task 2. A wider range of scaling factors would be needed to successfully fit psychophysical curves in these individuals. These 2 patients were those with the poorest performance in task 1 and with the greatest pain intensity. The limited sample size and the heterogeneity of the CRPS population makes it difficult to identify specific factors explaining why these patients had the poorest performance. No clear difference in their clinical profiles compared with the other CRPS participants was noted; Table 1 presents data for each CRPS participant to allow comparisons of their performance and clinical characteristics. Moreover, high variability was observed in the Bath Body Perception Disturbances scale, suggesting that disturbances in body perception in CRPS are heterogeneous. Such intersubject variability should normally facilitate the observation of correlations between variables, but no correlation was found for this specific variable. However, the metrologic properties of the French version of this scale have not yet been established, and our sample size was limited. Moreover, in accordance with a previous study,
      • Bean DJ
      • Johnson MH
      • Heiss-Dunlop W
      • Kydd RR
      Extent of recovery in the first 12 months of complex regional pain syndrome type-1: A prospective study.
      we found low internal consistency of the Bath Body Perception Disturbances scale, suggesting that this scale is not sufficiently consistent for measuring alterations in body perception. Thus, the lack of correlation for this variable needs to be interpreted very carefully, because it might result simply from an inability to reliably assess this construct in the study sample. However, as demonstrated by Lewis et al,
      • Lewis JS
      • Kersten P
      • McCabe CS
      • McPherson KM
      • Blake DR
      Body perception disturbance: A contribution to pain in complex regional pain syndrome (CRPS).
      the nature and extent of body perception impairments in CRPS are highly variable across individuals. Whereas other scales measuring body image focus mainly on the size and shape of the painful limb,
      • Moseley GL
      Distorted body image in complex regional pain syndrome.
      • Peltz E
      • Seifert F
      • Lanz S
      • Müller R
      Maihöfner C: Impaired hand size estimation in CRPS.
      the Bath Body Perception Disturbances scale takes feelings about and attention to the painful limb into account, possibly explaining the low internal consistency. Finally, the pain intensity reported by CRPS patients was low compared with previous studies assessing body perception and sense of limb position in this population
      • Bank PJ
      • Peper CL
      • Marinus J
      • Beek PJ
      • van Hilten JJ
      Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
      • Lewis JS
      • Kersten P
      • McCabe CS
      • McPherson KM
      • Blake DR
      Body perception disturbance: A contribution to pain in complex regional pain syndrome (CRPS).
      ,
      • Lewis JS
      • Kersten P
      • McPherson KM
      • Taylor GJ
      • Harris N
      • McCabe CS
      • Blake DR
      Wherever is my arm? Impaired upper limb position accuracy in complex regional pain syndrome.
      ; however, deficits in body perception and kinesthesia were observed in our series, suggesting that alterations are present even in less severe cases of CRPS.

      Conclusions

      The senses of limb position and limb movement during active movement are altered and associated in CRPS, suggesting a blurred representation, but not a bias, in perception of the painful limb. The strong correlation found between the 2 tasks suggests that the deficits observed stem from similar underlying processes. Interestingly, alterations in kinesthesia were not related to the participants’ reported perceptions of the painful limb, suggesting independent processes in the alterations of body schema and body image in CRPS. From a clinical perspective, these data suggest that these 2 body representations should be evaluated separately in CRPS, and that interventions aimed at improving body image will not necessarily impact body schema and vice versa; however, more reliable assessment methods of body image in CRPS are needed.

      Acknowledgments

      We thank Nicolas Robitaille, PhD, for developing and implementing the task. The technical development for that study was supported by an ENGAGE grant from Natural Sciences and Engineering Research Council of Canada (EGP477404-14) and performed in collaboration with BKIN Technologies, which provided technical support. Support was also provided by an operating grant from the Canadian Institutes of Health Research (MOP-125869). C.B. was supported by fellowships from the Centre interdisciplinaire de recherche en réadaptation et en intégration sociale, Centre thématique de recherche en neurosciences of Laval University, and Fonds de recherche Nature et Technologies. C. Mercier is supported by a salary award from Fonds de recherche Québec-Santé. C. McCabe is supported by the Florence Nightingale Foundation as a Clinical Professor in Nursing.

      Appendix. Supplementary materials

      References

        • Bank PJ
        • Peper CL
        • Marinus J
        • Beek PJ
        • van Hilten JJ
        Motor dysfunction of complex regional pain syndrome is related to impaired central processing of proprioceptive information.
        J Pain. 2013; 14: 1460-1474
        • Bean DJ
        • Johnson MH
        • Heiss-Dunlop W
        • Kydd RR
        Extent of recovery in the first 12 months of complex regional pain syndrome type-1: A prospective study.
        Eur J Pain. 2016; 20: 884-894
        • Dijkerman HC
        • de Haan EH
        Somatosensory processes subserving perception and action.
        Behav Brain Sci. 2007; 30: 189-201
        • Dukelow SP
        • Herter TM
        • Moore KD
        • Demers MJ
        • Glasgow JI
        • Bagg SD
        • Norman KE
        • Scott SH
        Quantitative assessment of limb position sense following stroke.
        Neurorehabil Neural Repair. 2010; 24: 178-187
        • Farrer C
        • Franck N
        • Georgieff N
        • Frith CD
        • Decety J
        • Jeannerod M
        Modulating the experience of agency: A positron emission tomography study.
        Neuroimage. 2003; 18: 324-333
        • Gogolla N
        The insular cortex.
        Curr Biol. 2017; 27: R580-R586
        • Hamacher D
        • Hamacher D
        • Krowicki M
        • Schega L
        Gait variability in chronic back pain sufferers with experimentally diminished visual feedback: A pilot study.
        J Mot Behav. 2016; 48: 205-208
        • Harden RN
        • Bruehl S
        • Perez RS
        • Birklein F
        • Marinus J
        • Maihofner C
        • Lubenow T
        • Buvanendran A
        • Mackey S
        • Graciosa J
        • Mogilevski M
        • Ramsden C
        • Chont M
        • Vatine JJ
        Validation of proposed diagnostic criteria (the “Budapest criteria”) for Complex Regional Pain Syndrome.
        Pain. 2010; 150: 268-274
        • Karnath HO
        • Baier B
        • Nägele T
        Awareness of the functioning of one's own limbs mediated by the insular cortex?.
        J Neurosci. 2005; 25: 7134-7138
        • Lewis JS
        • Kersten P
        • McCabe CS
        • McPherson KM
        • Blake DR
        Body perception disturbance: A contribution to pain in complex regional pain syndrome (CRPS).
        Pain. 2007; 133: 111-119
        • Lewis JS
        • Kersten P
        • McPherson KM
        • Taylor GJ
        • Harris N
        • McCabe CS
        • Blake DR
        Wherever is my arm? Impaired upper limb position accuracy in complex regional pain syndrome.
        Pain. 2010; 149: 463-469
        • Lewis JS
        • McCabe CS
        Correcting the body in mind: body perception disturbance in complex regional pain syndrome (CRPS) and rehabilitation approaches.
        Pract Pain Manag. 2010; 10: 60-66
        • Lewis JS
        • Schweinhardt P
        Perceptions of the painful body: The relationship between body perception disturbance, pain and tactile discrimination in complex regional pain syndrome.
        Eur J Pain. 2012; 16: 1320-1330
        • Maihöfner C
        • Baron R
        • DeCol R
        • Binder A
        • Birklein F
        • Deuschl G
        • Handwerker HO
        • Schattschneider J
        The motor system shows adaptive changes in complex regional pain syndrome.
        Brain. 2007; 130: 2671-2687
        • Maihöfner C
        • Neundörfer B
        • Birklein F
        • Handwerker HO
        Mislocalization of tactile stimulation in patients with complex regional pain syndrome.
        J Neurol. 2006; 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
        • Moseley GL
        Distorted body image in complex regional pain syndrome.
        Neurology. 2005; 65: 773
        • Moseley GL
        I can't find it! Distorted body image and tactile dysfunction in patients with chronic back pain.
        Pain. 2008; 140: 239-243
        • Moseley GL
        Why do people with complex regional pain syndrome take longer to recognize their affected hand?.
        Neurology. 2004; 62: 2182-2186
        • Naito E
        • Morita T
        • Amemiya K
        Body representations in the human brain revealed by kinesthetic illusions and their essential contributions to motor control and corporeal awareness.
        Neurosci Res. 2016; 104: 16-30
        • Paillard J
        Body schema and body image - a double dissociation in deafferented patients.
        in: Gantchev GN Mori S Massion J Motor Control Today and Tomorrow. Academic Publishing House, Sofia1999: 197-214
        • Peltz E
        • Seifert F
        • Lanz S
        • Müller R
        Maihöfner C: Impaired hand size estimation in CRPS.
        J Pain. 2011; 12: 1095-1101
        • Di Pietro F
        • McAuley JH
        • Parkitny L
        • Lotze M
        • Wand BM
        • Moseley GL
        • Stanton TR
        Primary somatosensory cortex function in complex regional pain syndrome: A systematic review and meta-analysis.
        J Pain. 2013; 14: 1001-1018
        • Pitron V
        • De Vignemont F
        Beyond differences between the body schema and the body image: Insights from body hallucinations.
        Conscious Cogn. 2017; 53: 115-121
        • Pleger B
        • Ragert P
        • Schwenkreis P
        • Förster AF
        • Wilimzig C
        • Dinse H
        • Nicolas V
        • Maier C
        • Tegenthoff M
        Patterns of cortical reorganization parallel impaired tactile discrimination and pain intensity in complex regional pain syndrome.
        Neuroimage. 2006; 32: 503-510
        • Proske U
        • Gandevia SC
        The kinaesthetic senses.
        J Physiol. 2009; 587: 4139-4146
        • Roosink M
        • McFadyen BJ
        • Hébert LJ
        • Jackson PL
        • Bouyer LJ
        • Mercier C
        Assessing the perception of trunk movements in military personnel with chronic non-specific low back pain using a virtual mirror.
        PLoS One. 2015; 10e0120251
        • Roosink M
        • Robitaille N
        • McFadyen BJ
        • Hébert LJ
        • Jackson PL
        • Bouyer LJ
        • Mercier C
        Real-time modulation of visual feedback on human full-body movements in a virtual mirror: Development and proof-of-concept.
        J Neuroeng Rehabil. 2015; 12: 1-10
        • Schilder JC
        • Schouten AC
        • Perez RS,
        • Huygen FJ,
        • Dahan A
        • Noldus LP
        • van Hilten JJ
        • Marinus J
        Motor control in complex regional pain syndrome: A kinematic analysis.
        Pain. 2012; 153: 805-812
        • Schwoebel J
        • Coslett HB
        Evidence for multiple, distinct representations of the human body.
        J Cogn Neurosci. 2005; 17: 543-553
        • Schwoebel J
        • Friedman R
        • Duda N
        • Coslett HB
        Pain and the body schema: Evidence for peripheral effects on mental representations of movement.
        Brain. 2001; 124: 2098-2104
        • Scott S
        • Brown I
        Method and apparatus for assessing proprioceptive function.
        U.S. Patent. 2013; 8: 277-396
        • Sheeran L
        • Sparkes V
        • Caterson B
        • Busse-Morris M
        • van Deursen R
        Spinal position sense and trunk muscle activity during sitting and standing in non-specific chronic low back pain: Classification analysis.
        Spine (Phila Pa 1976). 2012; 37: E486-E495
        • Sumitani M
        • Shibata M
        • Iwakura T
        • Matsuda Y
        • Sakaue G
        • Inoue T
        • Mashimo T
        • Miyauchi S
        Pathologic pain distorts visuospatial perception.
        Neurology. 2007; 68: 152-154
        • Swart CM.
        • Stins JF
        • Beek PJ
        Cortical changes in complex regional pain syndrome (CRPS).
        Eur J Pain. 2009; 13: 902-907
        • Valenzuela-Moguillansky C
        Pain and disturbances in body awareness.
        Cogn Process. 2012; 13: S30
        • de Vignemont F
        Body schema and body image–pros and cons.
        Neuropsychologia. 2010; 48: 669-680
      1. Wolpert DM, Ghahramani Z, Jordan MI: An internal model for sensorimotor integration. Science 269:1880-1882, 1995