Background Complex Regional Pain Syndrome (CRPS) is a chronic pain condition of unknown aetiology.
Structural and functional changes within the cerebral cortex (cortical reorganisation) are implicated in some chronic pain conditions. We investigated the high density EEG (128 channels) markers suggestive of cortical reorganisation in CRPS by studying the somatosensory ERPs (Event Related Potential) elicited on painless finger stimulation.
Objectives The main objective was to determine whether the EEG markers of cortical reorganisation in CRPS correlates with finger misperception.
Methods Prospective observational cohort study of CRPS patients. Subjects with confounding neurological conditions excluded. 5 questionnaires administered: Brief Pain Inventory, Upper Extremity Functional Index, Lower Extremity Functional Index, Neglect-like Symptom Questionnaire, Hospital Anxiety and Depression Score. Painless stimuli delivered to the fingertips by custom made handboxes and EEG data collected. In experiment 1, subjects did not have to respond to stimuli whereas in experiment 2 designed to elicit P300 in addition to somatosensory ERPs, they had to tell which finger was touched. EEG data was preprocessed in EEGLAB and Global Field Power (GFP) analysed using FieldTrip implemented in MATLAB.
Results 13 CRPS patients (11 females) and 13 age-and-sex matched healthy volunteers recruited. On average, patients had significantly longer reaction times in the affected than in the unaffected hand (p=0.02). They had significantly longer reaction times compared to healthy (p=0.009). Patients had statistically significant lower accuracy in the affected than in the unaffected hand (p=0.003). Patients had statistically significant lower accuracy in the affected hand compared to the left hand in healthy (p=0.002). In experiment 1, there were no differences in GFP between the patient affected and unaffected sides or between patient and healthy. However in experiment 2, there were statistically significant differences at the group level between patient and healthy in the time segment 200 to 400 ms but not between 0-200 ms (p=0.002). There was significant variability in the patient group compared to the healthy group in terms of GFP score for both affected (Levene's test, p=0.03) and unaffected (p=0.045) sides. No significant variability in peak latency was noted. Hand laterality time was the only behavioural variable to show significant correlation (after bonferroni correction) with an EEG variable (mean GFP amplitude) of affected hand.
Conclusions CRPS patients had significantly reduced accuracy and prolonged reaction time in the affected side in the behavioural task of identifying the finger stimulated by a handbox. There was no significant difference in the GFP latency in the patient group compared to healthy subjects. There was also no difference between affected and unaffected sides of the patient group suggesting there was no impairment of somatosensory conduction from the periphery to the somatosensory cortex. However, GFP amplitude corresponding to P300 was significantly higher in the patient affected side compared to the healthy subjects suggesting cortical reorganisation.
Disclosure of Interest None declared