Spatial localisation in Complex Regional Pain Syndrome under uni- and multi-sensory guidance

Aim of Investigation
Patients with Complex Regional Pain Syndrome (CRPS) can have problems with tasks that require spatial processing, such as localising their affected limb or attending to touch on the affected side of space. Also, some patients report that pain can be exacerbated by loud noises and ambiguous visual stimuli, suggesting changes in multisensory integration.

The aim of this study is to investigate spatial localisation and multisensory integration in patients with CRPS.

Methods
Participants were 8 CRPS patients (3 upper limb, 5 lower limb) and 17 pain-free controls.

In separate blocks participants performed a target pointing task under visual-only (V) guidance, proprioceptive/motor-only (P/M) guidance and combined visual and proprioceptive/motor (V-P/M) guidance. During each block participants rested their head in a chin rest and directed their gaze to a fixation cross on the surface of a table. Laser pointers were used to project brief (200ms) targets 10cm to the left or right of the fixation cross (10 / Side). In the V block the participants gave verbal directions to the experimenter as to where the target had appeared and the experimenter placed a mark at the location indicated. In the P/M blocks the participant’s vision of the table surface and hand was occluded immediately after the target disappeared and the participant touched the target’s location with their index finger. In the V-P/M blocks the participant touched the target’s location with full vision of the table surface and their hand.

Participants completed each P/M and V-P/M block once with their left and once with their right hand. Pointing end-points were recorded by placing ink on the participant’s finger, which was transferred to the paper when they touched the page. The paper was refreshed after every trial.


Results
For each participant, estimates of variable error (VE: the sum of the variance of the points in x and y directions) were calculated separately for each condition (V, P/M, V-P/M). To evaluate localisation ability in CRPS, the VEs of individual patients were compared to bootstrapped (1000 samples) 95% Confidence Intervals (CIs) around the controls’ means. One upper limb and one lower limb CRPS patient had VEs that were within the controls’ 95% CI for all three conditions. The remaining six patients had VEs that exceeded the upper boundary of the 95% CI for the controls’ means in at least one condition. Specifically, four had larger-than-normal VEs in the V condition, two had larger-than-normal VEs in the P/M condition, and five had larger-than-normal VEs in the V-P/M condition. This suggests that CRPS can be accompanied by deficits in localisation under both uni- and multi-modal guidance.

Preliminary inspection of VE data when separated by Hand and Target Side suggests that more patients were impaired in P/M guidance for targets that appeared on the affected side compared to the unaffected side of space, regardless of which hand was used.

Planned paired-samples t-tests with bootstrapping (1000 samples) showed that for control participants the VEs in the V-P/M condition were lower than in both the V condition (p=.002) and the P/M condition (p=.001). This is consistent with previous findings that uncertainty is reduced during multi-sensory compared to uni-sensory localisation. For the patients, VEs in the V-P/M condition were not significantly different to VEs in the V condition (p=.11) nor the P/M condition (p=.35). This is suggestive of a deficit in multisensory integration.

Conclusions
The results provide evidence for impaired spatial localisation in CRPS under conditions of visual-only, proprioceptive/motor-only, and combined visual and proprioceptive/motor guidance. Four of the five lower limb CRPS patients showed some form of localisation deficit, indicating that these changes are not limited to tasks involving the affected limb but generalise to broader spatial cognition. The results also provide preliminary evidence for deficits in multisensory integration in CRPS, although this conclusion should be treated with caution due to the small and heterogeneous sample.