ABSTRACT
Complex regional pain syndrome (CRPS) is a chronic pain condition with extensive symptoms. Motor imagery (MI) is a common therapy in chronic pain, however it is unknown whether patients with CRPS have a subjective MI deficit. In this single-center observational study, 120 patients were recruited, comprising 40 participants each with CRPS, chronic limb pain (CLP), and healthy individuals.. In this single-center observational study, 120 patients were recruited, comprising 40 participants each with CRPS, chronic limb pain (CLP), and healthy individuals. Participants completed the Motor Imagery Questionnaire - Revised Second (MIQ-RS) once on each side to assess their subjective kinesthetic (KMI) and visual (VMI) MI abilities. The study was performed in the University Hospital of Nîmes (France). The total MIQ-RS score and KMI and VMI subscores were compared between groups and between healthy and painful sides.. There was no difference between or within groups (p>0.05; 95% CI) for all scores. Bayesian analysis suggested that CLP patients had the same MI abilities as healthy individuals and between their healthy and painful sides for the KMI score (BF01>3). This approach allowed us to conclude that if a difference exists for CRPS patients, it is very small. Motor imagery training appears to be effective in rehabilitation programs for patients with CRPS, but not in improving a deficit in the ability to perform motor imagery. Future studies should aim to assess MI abilities by a subjective or objective method in a longitudinal study and observe changes in these abilities.
INTRODUCTION
Complex regional pain syndrome (CRPS) is a chronic pain condition characterized by multiple categories of symptoms with sensory, motor, vasomotor, and sudomotor disorders 1–3. The long duration of symptoms represents a high cost to the health care system 4,5. The mechanism of CRPS is not fully understood, although there is an international consensus to explain the symptoms by both central and peripheral processes2,6. Some studies appear to show changes in brain structure and function on both affected and unaffected sides, but the results are not consistent 7–10. However, abnormal neuroplasticity11 in the central nervous system has been demonstrated, accounting for reductions in higher-order motor control and dysfunction of intentional movements 12, and altered body perception called “neglect-like syndrome” 13,14. Furthermore, a reduced activation of cortical areas during the motor imagery (MI) task in the affected side 15 has been observed, and changes in motor areas such as the premotor cortex that is involved in MI abilities 7,16,17.
MI is a dynamic state whereby subjects mentally simulate specific actions 18, and precedes motor execution in the premotor cortex 19. Subjective MI abilities can be assessed using questionnaires, most often assessing two distinct parts of subjective MI abilities: visual motor imagery (VMI) and kinesthetic motor imagery (KMI) 20,21. VMI can be considered as external imagery (performing the movement as a viewer) and KMI as internal imagery (performing the movement in the first person with kinesthetic sensations). Movement is impaired early in CPRS and as a part of motor planning. Currently, MI training is a recommended technique and effective therapy in chronic pain and CRPS 22–26. Therefore, assessment of MI abilities using a simple questionnaire could be an interesting strategy in the clinical management of patients. Furthermore, patients with chronic low back pain pain (CLBP) appear to have impaired subjective motor skills 27, and chronic pain seems to be associated to changes in primary motor cortex 28. In addition, people with chronic pain as chronic limb pain shows impairments in Laterality judgment performance 29–31. Thus, it is difficult to determine whether the changes observed, particularly in MI tasks, are specific to CRPS or to chronic pain. 28,32
The aim of this study was to compare MI abilities between patients with CRPS, patients with chronic limb pain (CLP) and healthy subjects. We hypothesized that individuals with CRPS will have lower specific nerve impulse abilities than controls and at least equal to those of individuals with pain in other limbs. Secondarily, we aimed to compare MI abilities of participants’ affected side to their unaffected side in both chronic pain groups. We hypothesized that the painful side would be more affected than the healthy side 2,14,33.
MATERIALS AND METHOD
Design
This was a prospective, single-center, cross-sectional study conducted at the University Hospital of Nîmes (France). The study was approved by the local ethics committees (2020-A02281-38) and registered on clinicaltrials.org (NCT04703348). All patients received an information letter and provided written consent.
Participants and setting
Patients with CRPS were recruited from January 2021 to October 2022 in the pain medicine department (CHU Nîmes, France). Healthy participants were recruited from hospital employees using a poster campaign. Patients with CLP were recruited in the physical medicine and rehabilitation department and were included if they experienced limb pain lasting more than 3 months including musculoskeletal disorders, chronic post-traumatic pain, or post-surgical pain regardless of etiology.
Inclusion criteria were: age above 18, less than 150 minutes of moderate-to-intense physical activity per week, and education until A level of baccalaureate / high school diploma or equivalent. CRPS patients had to have a CRPS diagnosis by validated Budapest criteria 22,34,35, and CLP patients’ CLP must have arisen secondary to trauma, surgery, or musculoskeletal disorders. Additionally, patients were excluded if CRPS was secondary to stroke, stellate block injection performed 3 weeks before the questionnaire, presence of central neurological disease, patients with chronic fibromyalgia or low back pain, patient pregnancy/parturition/breastfeeding, visual blindness, amputation, or previous experience with MI practice.
Intervention
Questionnaires were originally planned to be performed during consultation. However, due to the Covid-19 pandemic, some participants (60%) completed the questionnaire over a video conference with an investigator. The questionnaires were audio recorded on REDCap© (online questionnaire) 36. Patients completed the MIQ-RS (Movement Imagery Questionnaire - Revised Second Edition) twice, starting with the right side and followed by the left side (non-randomized running order based on the difference in laterality between the upper and lower limb) with a pause if necessary.
Age, sex, body mass index (BMI), upper and lower dominant limb, education level, and physical activity level were recorded.
Outcome measures
The MIQ-RS is a validated questionnaire for assessing subjective MI 37,38, with a validated French translation 39, but it has not yet been used in patients with CRPS. It is a practitioner-administered questionnaire (done by 2 administrators), comprising 14 tasks in which patients first perform a movement such as a knee raise, then imagine it visually and kinesthetically. Patients evaluate their abilities on a 7-point Likert scale ranging from “very easy to see/feel” (1 point) to “very hard to see/feel” (7 points). A total score and two sub scores (KMI and VMI) are calculated on a scale of 7. There is no official cutoff available. The primary outcome was the difference in MIQ-RS total score and sub scores between the groups; the secondary outcome was the difference in scores between the healthy and painful side in the CRPS and CLP groups.
Data analysis
No published results were available in this population to perform a sample size calculation.
Based on the assumption that CRPS patients experience abnormal plasticity of the central nervous system, we used a population of stroke patients 40 to estimate the expected deficit (score of 29 for the VMI, and 25 for the KMI with standard deviations around 10 for both measures). The expected scores in the control group were based on the study by Loison et al. 39 (40 for the VMI and 33 for the KMI with standard deviations around 10). Thus, the inclusion of 40 patients seemed reasonable to highlight the smaller of the two expected differences (25 vs 33 for the KMI score with a SD=10) with a power of 90% and an alpha risk at 5%.
MI abilities were compared between the three groups to explore the effect of pain on subjective abilities to determine whether MI was different in CRPS or whether pain could account for this difference. The mean scores of the total MIQ-RS and both sub scores were recorded. Then, an intragroup comparison of MI abilities was performed between the painful side and the healthy side for CRPS and CLP patients to address whether pain affected the subjective ability to perform MI.
We performed a second analysis with a Bayesian statistical approach to test for an effect of CRPS on MI subjective abilities 41–43. This approach determines the degree of evidence to reject the hypothesis. A Bayes factor between 5 and 10 is considered moderate evidence 42,44.
JASP version 0.16.4© software was used to perform statistical analyses. Because of a skewed distribution, a non-parametric approach was performed. Quantitative variables were expressed as median and IQR (Inter Quartile Range), and compared between groups (CRPS vs CLP vs healthy) using a Kruskal-Wallis test. We performed a chi-square test to compare homogeneity for categorical data. A nonparametric signed rank test was used for the within-group comparison. Statistical analyses were performed at the conventional two-tailed α level of 0.05.
RESULTS
A total of 129 participants were screened for inclusion, and 120 patients were retained after exclusion. Each group consisted of 40 patients; the CRPS group comprised 31 women (78%), and the CLP group 23 women (58%) compared to 28 women (70%) for the Healthy group (Table 1).
Primary objective results
There were no statistical differences between the three groups for the MIQ-RS total score (Figure 1) (95%CI p>0.05), the KMI score (95%CI p>0.05), or the VMI score (95%CI p>0.05) (Table 2).
Secondary objective
CRPS Group
There was no statistical difference between the painful side and the healthy side for the MIQ-RS total (Figure 2) (95%CI p>0.05), KMI score (95%CI p>0.05) and VMI score (95%CI p>0.05). Results are summarized in Table 3.
Chronic Limb Pain Group
There was also no statistical difference between the painful side and the healthy side for the MIQ-RS total, KMI and VMI score (Table 4) in the CLP group.
Complementary analysis
To allow us to better interpret the results, we performed a second analysis with a Bayesian statistical approach to test for the presence or absence of an effect of CRPS on MI subjective abilities. 41–43. This approach gives us the ability to determine the degree of evidence to reject the hypothesis. A Bayes factor between 5 and 10 is considered moderate evidence 42,44. Based on this observation, we can state that the CLP group and the healthy group have the same subjective MI abilities (BF01=4.259 for the MIQ-RS KMI score; BF01=4.159 for the MIQ-RS KMI score; BF01=4.304 for the MIQ-RS VMI score), and the CLP patients have the same subjective KMI abilities in their healthy and painful side (BF01=3.381 for the MIQ-RS KMI score).
For the other objectives, we cannot interpret the results, probably due to a lack of power.
DISCUSSION
There was no statistical difference in subjective MI abilities between the CRPS, CLP, and healthy groups, and between the healthy and painful sides for the CRPS and CLP groups. Furthermore, CLP patients did not appear to be impaired in their subjective MI abilities compared with healthy individuals and in their affected limb compared with their healthy limb. The secondary analysis with the Bayesian method allows us not to consider a significant degree of evidence to reject the null hypothesis in the presence of a Bayesian factor lower than 3. The finding of no difference with an inferential model and the absence of similarities leads us to take precautions in conclusion. This result seems to be due to a lack of power.
Our results could be explained by some limitations of our study, such as the non-random order in which the questionnaire was completed and the lack of matching between the groups, due to the low prevalence of the CRPS and our strict selection criteria. Indeed, education and level of physical activity were statistically different between the groups and are known to modify motor imagery abilities 39. Although the MIQ-RS is a commonly used questionnaire 45, and seems to be a reliable and valid tool 37–39, it is not well balanced between upper and lower extremity questions, with 10 of the 14 questions focused on upper extremity MI abilities. Another limitation of this questionnaire is the assessment of the lower limbs. Indeed, the assessment was done by lifting the leg, whereas pain could occur when the patient put weight on it. A third limitation of the questionnaire was that some questions were asked while sitting, others while standing, and some questions did not specify the position. Finally, a systematic review proposed a threshold of 3 levels of capacity according to brain lesions, and despite abnormal central plasticity in CRSD patients, we could not use this classification 46.
The selection criteria were very strict to avoid selection or interview bias. The STROBE checklist 47,48 shows a very low level of bias. The strict selection criteria may have made the population non-representative of that commonly encountered in rehabilitation, particularly for education and physical activity level 49–51.
Our study only looked at subjective and not objective abilities in MI, however, the KMI subscore seems to correlate well with motor cortex activation in stroke patients 52. Motor cortex excitability during MI in patients with CRPS does not appear to differ from that of healthy persons 53, and our results seem to confirm these observations. Furthermore, neuroimaging techniques studying central nervous system dysfunction in CRPS patients are inconclusive or show bilateral disinhibition of the primary motor cortex 54. This change may be more consistent with altered body perception 13,14 than with MI disorders in CRPS patients. In futures studies, it could be interesting to correlate subjective MI abilities with medical imaging, such as electroencephalograms, to measure motor cortex activity and connectivity during the MI task, as has been done previously with healthy individuals. 55.
Chronic pain alters the central nervous system in different ways depending on the condition 28,56. A previous study using the Revised Movement Imagery Questionnaire (MIQ-R) showed impaired KMI and VMI abilities for CLBP patients 27. Patients with CLBP showed high levels of kinesiophobia, pain catastrophizing and low levels of coping compared with healthy participants. Psychological factors were not assessed in our study and could explain this difference. Pain catastrophizing impacts motor activity cortex 57,58. However, in CRPS patients, the action observation seems to increase symptoms 59 and epidemiological studies show presence of psychological symptoms 60. Thus, evidence suggest that the patients included in our study are not different in terms of psychological factors.
MI training appears to be an effective treatment for improving neuronal excitability or synapse conductance in healthy and pathological individuals 61 and KMI has the same cortical correlate in relation to movement execution 62. Recommended rehabilitation programs support the use of MI training alone or in combination with other modalities such as GMI (Graded Motor Imagery) 23,25,63–66. MI training appears to activate areas of the brain involved in movement preparation 15,20,61,67. It is plausible that MI helps CRPS patients by activating other gray matter pathways 12,68,69, by improving body schema deficits 14,70–73, or because CRPS or chronic pain patients may not be so different from healthy patients in their MI abilities. MI training appears to be an effective treatment, but not to correct MI deficit 26.
Finally, the use of the MIQ-RS could not be a relevant questionnaire to assess the subjective abilities of MI.
Future studies should use other MI assessment questionnaires or more objective assessment (medical imagery), and should focus on a simple and quick way to help clinicians develop tailored rehabilitation programs. Our results show that patients with CLP have the same MI abilities as healthy subjects and between their healthy and painful sides for kinesthetic motor abilities. Thus, whilst MI training may be effective in rehabilitation programs for CRPS patients, it does not appear to improve a specific MI deficit, however using the movement approach to restore function and decrease pain appears to be effective 66.
Data Availability
All data produced in the present work are contained in the manuscript
ACKNOWLEDGMENTS
We thank the following people for their support and assistance: Marine Ourmet, Brigitte Laffont for regulatory documents, Willy Fagart, Dr Anaïs Pages, Julie Bourdier, Romain Dolin, Shuan Banh, Kevin Jezequel and Sarah Kabani for editing the manuscript. The authors declare no conflicts of interest. The study data are available upon reasonable request to the corresponding author.
Footnotes
the references of the appendixes have been deleted because they did not bring any additional interesting data concerning the study
References
