Background & Aims
Phantom-limb pain (PLP) is commonly classified as neuropathic pain following amputation, brachial plexus avulsion injury, and is often resistant to pharmacotherapy and traditional rehabilitation techniques. Some previous studies have reported abnormal resting-state brain networks in amputees. For example, a functional magnetic resonance imaging (fMRI) study showed increased connectivity strength in the parietal cortex in amputees [1]. Another Electroencephalography (EEG) study also reported increased connectivity strength in the sensorimotor area in amputees [2]. However, there were few reports analyzing resting-state brain networks in patients suffering from PLP. The present preliminary study conducted resting-EEG recording and analysis with PLP patients to reveal abnormal resting-state brain networks in PLP. Our results would contribute to neuro-rehabilitation techniques for PLP, such as brain-machine interface and non-invasive brain stimulation.
Methods
Four patients with PLP and fourteen healthy controls took part in the present study. Three patients with PLP lost their arm due to accidents or treatment of osteosarcoma. One patient with PLP lost his sensorimotor function of the arm due to brachial plexus injury. The experimental protocols followed the Declaration of Helsinki. This study was approved by the Ethics Committees of Kio University. During EEG recording, each subject was seated comfortably in a dim room and was asked to stay in a resting state with eyes closed. The EEG signals were recorded continuously at 1024 Hz with a 32-channel EEG signal amplifier (Active Two; BioSemi, Amsterdam, The Netherlands) for at least three minutes. After preprocessing EEG data (i.e., re-sampling, re-referencing, noise removal), EEG network analysis was conducted using the Brain Connectivity Toolbox [3].
Results
We computed the debiased weighted phase-lag index (dwPLI) of all pairwise channels for each epoch and constructed a 32 × 32 connectivity matrix covering all electrodes. Based on the matrix, the degree of connectivity strength at each node was calculated. In the results, the connectivity strength of the alpha band at the midline area in PLP patients was larger than those in healthy controls. Additionally, the connectivity strength of the beta band at the ipsilateral sensorimotor in PLP patients was larger than those in healthy controls.
Conclusions
The excessive connectivity strength of the beta band at the sensorimotor area in PLP patients indicated the disturbance of inter-hemispheric inhibition between the bilateral sensorimotor cortices. The excessive connectivity strength of the beta band at the sensorimotor area in PLP patients might indicate the abnormal default mode network. Therefore, neuro-rehabilitation techniques should target the sensorimotor and default mode network to alleviate PLP.
References
[1] Bao B, Sun Y, Lin J, Gao T, Shen J, Hu W, Zhu H, Zhu T, Li J, Wang Z, Wei H, Zheng X. Altered cortical thickness and structural covariance networks in upper limb amputees: A graph theoretical analysis. CNS Neurosci Ther. 2023 Oct;29(10):2901-2911.
[2] Yuanyuan Lyu, Xiaoli Guo, Zhuo Wang, Shanbao Tong. Resting-state EEG network change in alpha and beta bands after upper limb amputation. Annu Int Conf IEEE Eng Med Biol Soc. 2016 Aug;2016:49-52.
[3] Rubinov M, Sporns O. Complex network measures of brain connectivity: uses and interpretations. Neuroimage. 2010 Sep;52(3):1059-69.