Abstract
Complex regional pain syndrome (CRPS) is a dehabilitating chronic condition occurring with peripheral lesions. There is growing consensus for a central contribution to CRPS. Although the nature of this central body representation disorder is increasingly debated, it has been repeatedly argued that CRPS results in motor neglect of the affected side. The present article describes a comprehensive and quantitative case report demonstrating that: (1) not all patients with chronic CRPS exhibit decreased spatial attention for the affected side and (2) patients may actually exhibit a substantial, broad and reliable attentional bias toward the painful side, akin to spatial neglect for the healthy side. This unexpected result agrees with the idea that patients can be hyper-attentive toward their pathological side as a manifestation of lowered pain threshold, allodynia and kinesiophobia.
1
Introduction
Complex regional pain syndrome (CRPS) is a lateralized chronic pain condition that usually appears after a traumatic and noxious event such as fracture or surgery and is characterized by severe and disproportionate pain concerning a joint and its neighborhood. CRPS patients show body schema abnormalities . Galer et al. observed underuse of the pathological limb, which they related to a kind of motor extinction: the arm can move satisfyingly only if the patient is paying a lot of attention to it. The pathological limb movements were also described as being hypokinetic, bradykinetic and hypometric , which had been described in spatial neglect . Twenty years ago, Galer et al. chose the term “neglect-like” to qualify the motor symptoms they observed in CRPS patients: poor motor function and motor neglect complaints expressed by patients. The authors explicitly did not intend to “suggest that the symptoms and signs seen in our patients are analogous to the classic hemispatial neglect that develops following stroke” . Nevertheless, this terminology paved the way for a long series of publications focused on the question of spatial neglect .
A considerable amount of theoretical elaborations on this issue have been published , but relatively few experimental studies are available. Among these contributions, the very notion that CRPS involves spatial neglect had been discussed or challenged. For example, it was proposed that these symptoms could be regarded more like a learned underuse than neglect-like symptoms . Legrain et al. proposed that a top-down attentional bias could be responsible for greater weight given to somatic or nociceptive input, thereby leading to an amplified perception of pain.
Keeping in mind the prototypical picture of spatial neglect , CRPS patients obviously do not present such a profound attentional bias regarding all sensory and motor modalities. The most commonly evoked feature of CRPS that has been associated with CRPS is motor neglect (e.g., ). Several aspects of motor neglect have been described: arm underuse, movement reductions and motor extinction. Arm underuse corresponds to patients’ total or partial lack of spontaneous movement in the absence of actual motor deficit. Movement reduction can be viewed as a minor form of underuse. Decreased movement amplitude (hypometria), increased latency (hypokinesia) or duration (bradykinesia) have been described in neglect patients acting with their healthy arm toward the left . One yet milder variant of this symptom is motor extinction: although movements can be performed normally in unimanual condition for both hands, bimanual movements unveil a progressive decrease of quality, amplitude and frequency on the pathological side . Clinically, this symptom is commonly explored by using a simple finger-tapping task.
As for neglect patients, CRPS patients show modifications of spatial reference frames. The Sumitani et al. study showed a visual straight-ahead deviation for the 36 CRPS patients they examined. However, surprisingly, this deviation was found toward the painful side, which is the opposite of the implicit hypothesis of “neglect-like” behavior of the pathological side, which would have implied a deviation toward the healthy side, as found in neglect. However, a bias in the perception of the visual subjective body midline has also been found to be independent of the side of pain. Reinersmann et al. observed that in the dark, CRPS patients perceived their visual body midline as shifted toward the left, independent of the actual side of pain, whereas Kolb et al. found no difference between CRPS patients and pain control patients.
Another similarity between neglect after stroke and the spatial cognition disorder in CRPS is the therapeutic effect of prism adaptation. Indeed, prism adaptation is one of the most widely used rehabilitation methods for neglect and also one of the most effective . It is also an effective method for CRPS rehabilitation . Prism adaptation for CRPS patients alleviates pain and restores motor ability and range of motion. However, determining the direction of the prismatic displacement is not straightforward, because whether CRPS is associated with neglect or over-representation of the painful limb is unclear.
We targeted several main questions in this case study. First, do CRPS patients systematically exhibit neglect? If so, do they neglect the healthy or the pathological side? Is their deficit limited to motor symptoms for the affected limb or do they expand to perceptual neglect? Do they exhibit a reliable, spatial frame of reference bias similar to spatial neglect patients? To address these questions, we assessed one CRPS patient for motor neglect with 2 kinematic tasks, for perceptual neglect with line-bisection and mental number bisection, and examined spatial reference frames with visual and manual straight-ahead tasks.
2
Material and methods
2.1
Patient
A 50-year-old woman had CRPS type 1 of the left hand due to surgery to remove a benign cyst 3 years before. Intense pain on the back and palm of the hand and the wrist rapidly developed. The patient presented the following symptoms fulfilling the Budapest criteria for CRPS diagnosis: continuing pain that was disproportionate to the inciting event, intense allodynia on the back of the hand, sudomotor changes, motor dysfunction (tremor) and trophic changes (hair loss). She went through several types of therapeutics with no sustainable effect; therapy included the usual analgesic medications step 1, 2 and 3; physiotherapy; occupational therapy; and trans-cutaneous electro-stimulation. She had stopped working; the psychological impact of her disability was huge and her sleep was severely affected. She experienced permanent pain, rated from 60 to 80/100 on a visual analog scale, and showed a pain behavior protecting her pathological hand.
2.2
Spatial frames of reference
Spatial frames of reference were collected following the Rode et al. procedure .
2.2.1
Visual straight-ahead
The task consisted of 10 trials for each condition: 100 cm and 200 cm away. The VSA1m involved 3 sessions of 10 trials each and the VSA2m, 2 sessions of 10 trials.
2.2.2
Manual/proprioceptive straight-ahead
The patient was asked to point 10 times with her right hand and 10 times with her left hand in darkness in the “straight-ahead” position in the direction of an imaginary line dividing her body into 2 equivalent halves. Measurement precision was estimated at ± 0.5 degrees. The test involved 3 sessions of 10 trials each.
For the 2 tasks, we calculated the mean ± SE of the 3 sessions in 2 ways: from the 3 sessions’ means (SE1) and from the 30 measures (SE2). We compared the pooled 30 measures to the theoretical value zero by Student t -test.
2.3
Spatial cognition
2.3.1
Line-bisection task
The patient was comfortably seated in front of a table with a A4 sheet of paper laying on the table and aligned with her body axis on which a centered 200-mm long and 2-mm thick line was figured. The patient was asked to mark what she thought to be the middle of the line without making a calculation. The experimental session consisted in 10 bisections with each hand. The distance was calculated by measuring the distance in millimeters between the reported point and the objective midline. A leftward error was signed negatively and a rightward error positively. The results were analyzed by comparing the mean to zero by Student t -test.
2.3.2
Testing for line-bisection reference frames
The patient was comfortably seated in front of a table on which there was a higher shelf containing the sheet. The patient’s hand lay on the table. The patient performed all bisections with her right hand while her left hand had various positions indicated by the examiner from extreme left to extreme right on the table, and the sheets of paper had also different position from left to right including the middle on the shelf ( Figs. 1 and 3 ). In short, the left unseen hand could lie at the left or the right of each line, which could be at the right or the left or aligned with the patient’s body axis. The test involved 6 trials for each combination, in a random order. Two-way Anova was computed to disentangle the contribution of left hand position with respect to the line (right and left) and test sheet location with respect to body axis (left, centre, right) and to test for potential interactions.
2.3.3
Mental number line-bisection task
Pairs of numbers were read aloud to the participant, who was then asked to state the number that she intuitively perceived to be at the middle of each interval (without making calculations). There was no time limit for responding, and number pairs were repeated at the participant’s request. The numbers ranged from 2 to 98, and the intervals used were 9 (e.g., 13 and 22), 16, 25, 36, 49, and 64 . The experimental session consisted of 72 trials. The numerical distance was calculated by subtracting the reported (subjective) midpoint number from the objective-midpoint number of each number interval. Since the mental number line is implicitly represented from left to right in one’s internal representation, underestimation of the midpoint number would indicate a leftward shift of the subjective center in internal space, and overestimation of the midpoint number would indicate a rightward shift. The results were analyzed by comparing the mean to zero by Student t -test.
2.4
Motor extinction task
2.4.1
Apparatus and procedure
The patient sat in a comfortable arm-chair facing a table, hands lying on the table with the palms down. In this resting position, she was blindfolded and asked to listen to a sound at 120 beats per min during 10 s and to remember it. Then, after a “go” signal from the experimenter, she was asked to begin tapping with her index finger on the table at the previous heard frequency for 30 s. For this tapping test, 2 hand positions were performed in different blocks: distant hands (28 cm from each other) and hands closer to the sagittal axis (5 cm from each other). For each position, movements were recorded by using 3 conditions in the following order: right index finger alone, left index finger alone and both fingers simultaneously. These recordings were done twice.
2.4.2
Movement recording and data processing and analysis
We used an optoelectronic VICON MX GIGANET system composed of 8 infrared stroboscopes and 100-Hz infrared cameras to record the 3D motion of passive markers during the tapping test. One passive infrared reflecting marker was placed on the nail of the index finger. Each data acquisition began with the “go” signal from the experimenter and ended automatically after 30 s. After recording and 3D reconstruction, the spatial positions of each marker were filtered by using a Butterworth filter at a 6-Hz cutoff frequency. The spatial position of the index nail marker was used to compute movement kinematic parameters.
The 2 runs performed by the patient in each condition were averaged to generate an average run that was used to compute the following parameters. First, the index finger marker was used to compute the maximum amplitude of each movement and the time interval between taps. Second, a linear regression was computed for each parameter (amplitude and period) to explore temporal trends over the recording period. Moreover, for each parameter, a supplementary linear regression was computed on subtractive data between the unimanual versus bimanual conditions, to test whether trends observed in the bimanual condition significantly differed from those observed in the baseline unimanual condition.
We also used the mean run to calculate a “detrended” value for each tap to allow for comparing means independently of the tendency. For this purpose, we first calculated an estimated value for each observed value (of amplitude and period) with the formula x estimated = x observed × a +b (ax + b being the regression observed on this variable). Then we calculated the detrended value with the formula x dentreded = run’s mean − x estimated + x observed . We then compared the detrended amplitude or period of different conditions by Student t -test for independent samples.
2.5
Reach-to-grasp task
2.5.1
Apparatus and procedure
The patient sat facing the same table as previously. In the resting position, the hand was on the table with the thumb and the index finger held in pinch-grip position on the starting position, positioned about 5 cm from the subject’s trunk along her sagittal axis. She was asked to reach and grasp a glass located on the table, 40 cm away from the starting point, at 20° of her sagittal axis on the left or right hemispace. Each trial began with a 2-s alert signal followed by a “go” signal from the experimentor, both triggering data acquisition and indicating the movement direction. Both hands were tested in separate blocks. In each block, 2 sessions were performed separately by using a small glass (4.3-cm diameter) on the left side and a large glass (5.3-cm diameter) on the right side or vice versa. In each session, 11 movements toward each glass were performed in a pseudo-random order. Thus, in this experimental procedure, 88 movements were recorded (2 hands × 2 positions × 2 glasses × 11 repetitions).
2.5.2
Movement recording and data processing and analysis
We used the same motion-capture system. Four passive infrared reflecting markers were placed on the following sites: the nails of the thumb and the index finger, and the styloid process of the radius at the wrist. The time-course of wrist velocity, wrist acceleration, thumb-index grip aperture and grip aperture velocity were derived from these data. The movement parameters described below were determined on these profiles by a semi-automatic procedure with manual verification trial by trial. We examined the 5 following parameters, which were the most relevant to whether CRPS is systematically accompanied by motor neglect:
- •
reaction time calculated as the delay (in ms) between the “go” signal and the release of the start position;
- •
peak velocity calculated as the maximum wrist velocity value during the reach-to-grasp task;
- •
maximum grip aperture (MGA) calculated as the maximal 3D distance between the thumb and index markers;
- •
MGA time calculated as the time the MGA reached its maximum value from movement onset (The movement end was determined visually on the grip aperture profile as the point of achievement of a stable grip on the glass, before it was lifted);
- •
movement time thus computed as the time between the onset and the end of the movement.
A three-way Anova was used to assess how 3 factors (hand, hemispace and size of the glass) influenced the movement components. The significance level was set at P ≤ 0.05 for all analyses.
2
Material and methods
2.1
Patient
A 50-year-old woman had CRPS type 1 of the left hand due to surgery to remove a benign cyst 3 years before. Intense pain on the back and palm of the hand and the wrist rapidly developed. The patient presented the following symptoms fulfilling the Budapest criteria for CRPS diagnosis: continuing pain that was disproportionate to the inciting event, intense allodynia on the back of the hand, sudomotor changes, motor dysfunction (tremor) and trophic changes (hair loss). She went through several types of therapeutics with no sustainable effect; therapy included the usual analgesic medications step 1, 2 and 3; physiotherapy; occupational therapy; and trans-cutaneous electro-stimulation. She had stopped working; the psychological impact of her disability was huge and her sleep was severely affected. She experienced permanent pain, rated from 60 to 80/100 on a visual analog scale, and showed a pain behavior protecting her pathological hand.
2.2
Spatial frames of reference
Spatial frames of reference were collected following the Rode et al. procedure .
2.2.1
Visual straight-ahead
The task consisted of 10 trials for each condition: 100 cm and 200 cm away. The VSA1m involved 3 sessions of 10 trials each and the VSA2m, 2 sessions of 10 trials.
2.2.2
Manual/proprioceptive straight-ahead
The patient was asked to point 10 times with her right hand and 10 times with her left hand in darkness in the “straight-ahead” position in the direction of an imaginary line dividing her body into 2 equivalent halves. Measurement precision was estimated at ± 0.5 degrees. The test involved 3 sessions of 10 trials each.
For the 2 tasks, we calculated the mean ± SE of the 3 sessions in 2 ways: from the 3 sessions’ means (SE1) and from the 30 measures (SE2). We compared the pooled 30 measures to the theoretical value zero by Student t -test.
2.3
Spatial cognition
2.3.1
Line-bisection task
The patient was comfortably seated in front of a table with a A4 sheet of paper laying on the table and aligned with her body axis on which a centered 200-mm long and 2-mm thick line was figured. The patient was asked to mark what she thought to be the middle of the line without making a calculation. The experimental session consisted in 10 bisections with each hand. The distance was calculated by measuring the distance in millimeters between the reported point and the objective midline. A leftward error was signed negatively and a rightward error positively. The results were analyzed by comparing the mean to zero by Student t -test.
2.3.2
Testing for line-bisection reference frames
The patient was comfortably seated in front of a table on which there was a higher shelf containing the sheet. The patient’s hand lay on the table. The patient performed all bisections with her right hand while her left hand had various positions indicated by the examiner from extreme left to extreme right on the table, and the sheets of paper had also different position from left to right including the middle on the shelf ( Figs. 1 and 3 ). In short, the left unseen hand could lie at the left or the right of each line, which could be at the right or the left or aligned with the patient’s body axis. The test involved 6 trials for each combination, in a random order. Two-way Anova was computed to disentangle the contribution of left hand position with respect to the line (right and left) and test sheet location with respect to body axis (left, centre, right) and to test for potential interactions.
2.3.3
Mental number line-bisection task
Pairs of numbers were read aloud to the participant, who was then asked to state the number that she intuitively perceived to be at the middle of each interval (without making calculations). There was no time limit for responding, and number pairs were repeated at the participant’s request. The numbers ranged from 2 to 98, and the intervals used were 9 (e.g., 13 and 22), 16, 25, 36, 49, and 64 . The experimental session consisted of 72 trials. The numerical distance was calculated by subtracting the reported (subjective) midpoint number from the objective-midpoint number of each number interval. Since the mental number line is implicitly represented from left to right in one’s internal representation, underestimation of the midpoint number would indicate a leftward shift of the subjective center in internal space, and overestimation of the midpoint number would indicate a rightward shift. The results were analyzed by comparing the mean to zero by Student t -test.
2.4
Motor extinction task
2.4.1
Apparatus and procedure
The patient sat in a comfortable arm-chair facing a table, hands lying on the table with the palms down. In this resting position, she was blindfolded and asked to listen to a sound at 120 beats per min during 10 s and to remember it. Then, after a “go” signal from the experimenter, she was asked to begin tapping with her index finger on the table at the previous heard frequency for 30 s. For this tapping test, 2 hand positions were performed in different blocks: distant hands (28 cm from each other) and hands closer to the sagittal axis (5 cm from each other). For each position, movements were recorded by using 3 conditions in the following order: right index finger alone, left index finger alone and both fingers simultaneously. These recordings were done twice.
2.4.2
Movement recording and data processing and analysis
We used an optoelectronic VICON MX GIGANET system composed of 8 infrared stroboscopes and 100-Hz infrared cameras to record the 3D motion of passive markers during the tapping test. One passive infrared reflecting marker was placed on the nail of the index finger. Each data acquisition began with the “go” signal from the experimenter and ended automatically after 30 s. After recording and 3D reconstruction, the spatial positions of each marker were filtered by using a Butterworth filter at a 6-Hz cutoff frequency. The spatial position of the index nail marker was used to compute movement kinematic parameters.
The 2 runs performed by the patient in each condition were averaged to generate an average run that was used to compute the following parameters. First, the index finger marker was used to compute the maximum amplitude of each movement and the time interval between taps. Second, a linear regression was computed for each parameter (amplitude and period) to explore temporal trends over the recording period. Moreover, for each parameter, a supplementary linear regression was computed on subtractive data between the unimanual versus bimanual conditions, to test whether trends observed in the bimanual condition significantly differed from those observed in the baseline unimanual condition.
We also used the mean run to calculate a “detrended” value for each tap to allow for comparing means independently of the tendency. For this purpose, we first calculated an estimated value for each observed value (of amplitude and period) with the formula x estimated = x observed × a +b (ax + b being the regression observed on this variable). Then we calculated the detrended value with the formula x dentreded = run’s mean − x estimated + x observed . We then compared the detrended amplitude or period of different conditions by Student t -test for independent samples.
2.5
Reach-to-grasp task
2.5.1
Apparatus and procedure
The patient sat facing the same table as previously. In the resting position, the hand was on the table with the thumb and the index finger held in pinch-grip position on the starting position, positioned about 5 cm from the subject’s trunk along her sagittal axis. She was asked to reach and grasp a glass located on the table, 40 cm away from the starting point, at 20° of her sagittal axis on the left or right hemispace. Each trial began with a 2-s alert signal followed by a “go” signal from the experimentor, both triggering data acquisition and indicating the movement direction. Both hands were tested in separate blocks. In each block, 2 sessions were performed separately by using a small glass (4.3-cm diameter) on the left side and a large glass (5.3-cm diameter) on the right side or vice versa. In each session, 11 movements toward each glass were performed in a pseudo-random order. Thus, in this experimental procedure, 88 movements were recorded (2 hands × 2 positions × 2 glasses × 11 repetitions).
2.5.2
Movement recording and data processing and analysis
We used the same motion-capture system. Four passive infrared reflecting markers were placed on the following sites: the nails of the thumb and the index finger, and the styloid process of the radius at the wrist. The time-course of wrist velocity, wrist acceleration, thumb-index grip aperture and grip aperture velocity were derived from these data. The movement parameters described below were determined on these profiles by a semi-automatic procedure with manual verification trial by trial. We examined the 5 following parameters, which were the most relevant to whether CRPS is systematically accompanied by motor neglect:
- •
reaction time calculated as the delay (in ms) between the “go” signal and the release of the start position;
- •
peak velocity calculated as the maximum wrist velocity value during the reach-to-grasp task;
- •
maximum grip aperture (MGA) calculated as the maximal 3D distance between the thumb and index markers;
- •
MGA time calculated as the time the MGA reached its maximum value from movement onset (The movement end was determined visually on the grip aperture profile as the point of achievement of a stable grip on the glass, before it was lifted);
- •
movement time thus computed as the time between the onset and the end of the movement.
A three-way Anova was used to assess how 3 factors (hand, hemispace and size of the glass) influenced the movement components. The significance level was set at P ≤ 0.05 for all analyses.
Related posts:
Validation of a novel activity monitor in impaired, slow-walking, crutch-supported patients
Care-related pain and discomfort in children with motor disabilities in rehabilitation centres
Pelvic floor muscle reflex activity during coughing – an exploratory and reliability study
An update on the pathophysiology of osteoarthritis
Does the physiatrist replicate the functions of the neurologist or orthopedist?
Takotsubo cardiomyopathy as a reversible complication of intrathecal baclofen withdrawal
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
