PhD defence B. (Bart) Witjes

Brief summary: 

In this thesis, we aim to further elucidate the cortical effects of chronic pain and the mechanisms of action of spinal cord stimulation (SCS), a treatment for chronic pain, using magnetoencephalography (MEG). The research also examines the potential of MEG features for classifying pain and SCS treatment effect through machine learning.

Key findings include the identification of increased slow alpha-band brain activity in patients with chronic pain, highlighting significant differences in various cortical and subcortical regions, also in areas not typically associated with pain processing.

A review of existing studies on SCS and resting-state MEG or electroencephalography (EEG) revealed heterogeneous study designs, stimulation paradigms, and outcome measures. Common findings were modulation of different pain pathways for different stimulation paradigms and evidence pointing to modulation of thalamocortical pathways.

Resting-state MEG in patients with SCS showed no statistically significant differences in the spectral features of interest among three different SCS paradigms or between responders and nonresponders. This suggests personalized treatment plans are necessary. Machine learning models applied to the acquired MEG data showed promising results in classifying chronic pain, but struggled classifying SCS treatment effect.

Additionally, we investigated the effects of conditioned pain modulation (CPM) on neural oscillations in subjects with and without chronic pain. CPM induced a reduction in pain ratings and desynchronization of the beta band (15-35 Hz) sensorimotor cortex responses to electrical pain.

The thesis concludes with recommendations for future research, emphasizing the need for larger sample sizes and taking into account confounders of pain in patient selection.