Neural Correlates of Pain-Autonomic Coupling in Patients With Complex Regional Pain Syndrome Treated by Repetitive Transcranial Magnetic Stimulation of the Motor Cortex.

OBJECTIVES: Complex regional pain syndrome (CRPS) is a chronic pain condition involving autonomic dysregulation. In this study, we report the results of an ancillary study to a larger clinical trial investigating the treatment of CRPS by neuromodulation. This ancillary study, based on functional magnetic resonance imaging (fMRI), evaluated the neural correlates of pain in patients with CRPS in relation to the sympathetic nervous system and for its potential relief after repetitive transcranial magnetic stimulation of the motor cortex. MATERIALS AND METHODS: Eleven patients with CRPS at one limb (six women, five men, aged 52.0 ± 9.6 years) were assessed before and one month after the end of a five-month repetitive transcranial magnetic stimulation (rTMS) therapy targeting the motor cortex contralateral to the painful limb, by means of electrochemical skin conductance (ESC) measurement, daily pain intensity scores on a visual numerical scale (VNS), and fMRI with motor tasks (alternation of finger movements and rest). The fMRI scans were analyzed voxelwise using ESC and VNS pain score as regressors to derive their neural correlates. The criterion of response to rTMS therapy was defined as ≥30% reduction in VNS pain score one month after treatment compared with baseline. RESULTS: At baseline, ESC values were reduced in the affected limb vs the nonaffected limb. There was a covariance of VNS with brain activation in a small region of the primary somatosensory cortex (S1) contralateral to the painful side on fMRI investigation. After rTMS therapy on motor cortex related to the painful limb, the VNS pain scores significantly decreased by 22% on average. The criterion of response was met in six of 11 patients (55%). In these responders, at one month after treatment, ESC value increased and returned to normal in the CRPS-affected limb, and overall, the increase in ESC correlated with the decrease in VNS after motor cortex rTMS therapy. At one month after treatment, there also was a covariance of both variables (ESC and VNS) with fMRI activation of the S1 region previously mentioned. The fMRI activation of other brain regions (middle frontal gyrus and temporo-parietal junction) showed correlation with ESC values before and after treatment. Finally, we found a positive correlation at one month after treatment (not at baseline) between VNS pain score and fMRI activation in the temporo-parietal junction contralateral to painful side. CONCLUSIONS: This study first shows a functional pain-autonomic coupling in patients with CRPS, which could involve a specific S1 region. However, the modulation of sympathetic sudomotor activities expressed by ESC changes was rather correlated with functional changes in other brain regions. Finally, the pain relief observed at one month after rTMS treatment was associated with a reduced activation of the temporo-parietal junction on the side in which rTMS was performed. These findings open perspectives to define new targets or biomarkers for using rTMS to treat CRPS-associated pain. CLINICAL TRIAL REGISTRATION: The Clinicaltrials.gov registration number for the study is NCT02817880.