Complex regional pain syndrome (CRPS) is a chronic pain condition associating sensory, motor, trophic and autonomic symptoms in one limb. Cognitive difficulties have also been reported, affecting the patients’ ability to mentally represent, perceive and use their affected limb. However, the nature of these deficits is still a matter of debate. Recent studies suggest that cognitive deficits are limited to body-related information and body perception, while not extending to external space. Here we challenge that statement, by using temporal order judgment (TOJ) tasks with tactile (i.e. body) or visual (i.e. extra-body) stimuli in patients with upper-limb CRPS. TOJ tasks allow characterizing cognitive biases to the advantage of one of the two sides of space. While the tactile TOJ tasks did not show any significant results, significant cognitive biases were observed in the visual TOJ tasks, affecting mostly the perception of visual stimuli occurring in the immediate vicinity of the affected limb. Our results clearly demonstrate the presence of visuospatial deficits in CRPS, corroborating the cortical contribution to the CRPS pathophysiology, and supporting the utility of developing rehabilitation techniques modifying visuospatial abilities to treat chronic pain.

Complex regional pain syndrome (CRPS) is thought to be characterized by cognitive deficits affecting patients' ability to represent, perceive, and use their affected limb as well as its surrounding space. This has been tested, among others, by straight-ahead tasks testing oneself's egocentric representation, but such experiments lead to inconsistent results. Because spatial cognitive abilities encompass various processes, we completed such evaluations by varying the sensory inputs used to perform the task. Complex regional pain syndrome and matched control participants were asked to assess their own body midline either visually (ie, by means of a moving visual cue) or manually (ie, by straight-ahead pointing with one of their upper limbs) and to reach and point to visual targets at different spatial locations. Although the 2 former tasks only required one single sensory input to be performed (ie, either visual or proprioceptive), the latter task was based on the ability to coordinate perception of the position of one's own limb with visuospatial perception. However, in this latter task, limb position could only be estimated by proprioception, as vision of the limb was prevented. Whereas in the 2 former tasks CRPS participants' performance was not different from that of controls, they made significantly more deviations errors during the visuospatial task, regardless of the limb used to point or the direction of pointing. Results suggest that CRPS patients are not specifically characterized by difficulties in representing their body but, more particularly, in integrating somatic information (ie, proprioception) during visually guided movements of the limb.

Complex Regional Pain Syndrome (CRPS) is characterized by pain, motor and inflammatory symptoms usually affecting one limb. Cognitive difficulties have been reported to affect patients’ ability to represent, perceive and use their affected limb. It is debated whether these difficulties result from deficits in controlling goal-directed movements in space or from a learned strategy to protect the affected limb. In order to dissociate the two hypotheses, patients with upper-limb CRPS were asked to move with their unaffected hand towards visual targets projected at different positions on a horizontal semi-reflexive mirror. By means of a robotic handle placed below the screen, they were asked to move a cursor, to reach and cross lines at their estimated midpoint. In some of the stimulation series, the affected hand was placed below the mirror so that some lines appeared projected onto that hand. Vision of the hands and the robotic handle was preserved or prevented by opening or closing a shutter below the mirror. Lines were displayed on the mirror according to which part of the body was affected (ispi- vs. contralateral) and the actual position of the affected hand (inside vs. outside the workspace). Comparatively to control participants, CRPS patients generally biased their estimation by bisecting the lines towards their left side, irrelative of which part of the body was affected and the position of the affected hand, both in ipsi- and contralateral space, with only a few exceptions. Our results are in line with previous studies having described a visuospatial deficit in CRPS patients and discard the explanation of observed symptoms in terms of learned nonuse strategies, as only the unaffected hand was used to perform the task. It is suggested that CRPS patients can display difficulties to perform tasks requesting visuo-motor coordination, reflecting the complex cortical reorganization occurring in CRPS.

Part of the multifaceted pathophysiology of Complex Regional Pain Syndrome (CRPS) is ascribed to lateralized maladaptive neuroplasticity in sensorimotor cortices, corroborated by behavioral studies indicating that patients present difficulties in mentally representing their painful limb. Such difficulties are widely measured with hand laterality judgment tasks (HLT), which are also used in the rehabilitation of CRPS to activate motor imagery and restore the cortical representation of the painful limb. The potential of these tasks to elicit motor imagery is critical to their use in therapy, yet, the influence of the body’s biomechanical constraints (BMC) on HLT reaction time, supposed to index motor imagery activation, is rarely verified. Here we investigated the influence of BMC on the perception of hand postures and movements in upper-limb CRPS. Patients were slower than controls in judging hand laterality, whether or not stimuli corresponded to their painful hand. Reaction time patterns reflecting BMC were mostly absent in CRPS and controls. A second experiment therefore directly investigated the influence of implicit knowledge of BMC on hand movement judgments. Participants judged the perceived path of movement between two depicted hand positions, with only one of two proposed paths that was biomechanically plausible. While the controls mostly chose the biomechanically plausible path, patients did not. These findings show non-lateralized body representation impairments in CRPS, possibly related to difficulties in using correct knowledge of the body’s biomechanics. Importantly, they demonstrate the challenge of reliably measuring motor imagery with the HLT, which has important implications for the rehabilitation with these tasks.

There is growing evidence in the literature suggesting that sensori-motor deficits have a detrimental impact on cognitive abilities, such as the perception and representation of space. Most of the tasks classically used to assess spatial perception abilities after brain damage seem however not adequate to reveal potentially more discrete impairments in patients that are predominantly suffering from sensori-motor symptoms. To this aim we adapted the classic line bisection task in a virtual reality environment for the use with patients with upper-limb sensory-motor deficits. Here, we report the results from two pre-clinical experiments with healthy volunteers. Lines were projected horizontally on a semi-reflective mirror. Participants were asked to bisect the lines at their estimated midpoint by means of a robotic handle. Manipulated factors were the static hand, the position of the static hand (inside or outside of workspace), the visibility of the hands and the location of the lines. Results showed that participants neglected the most lateral part of the lines when they were projected distantly from the starting point. Bisection biases were in general more important when participants could see their hands. Additionally, participants took more time to initiate movements when they had to cover a short distance to reach the lines. We conclude that our robotic adaptation of the line bisection task was able to highlight subtle perceptual asymmetries in healthy individuals and that its use to reveal discrete cognitive deficits in patients with sensory-motor impairments seems therefore promising.