Peripheral Electrical Stimulation Modulates Cortical Beta-Band Activity

Abstract

Low-frequency peripheral electrical stimulation using a matrix electrode (PEMS) modulates spinal nociceptive pathways. However, the effects of this intervention on cortical oscillatory activity have not been assessed yet. The aim of this study was to investigate the effects of low-frequency PEMS (4 Hz) on cortical oscillatory activity in different brain states in healthy pain-free participants. In experiment 1, PEMS was compared to sham stimulation. In experiment 2, motor imagery (MI) was used to modulate the sensorimotor brain state. PEMS was applied either during MI-induced oscillatory desynchronization (concurrent PEMS) or after MI (delayed PEMS) in a cross-over design. For both experiments, PEMS was applied on the left forearm and resting-state electroencephalography (EEG) was recording before and after each stimulation condition. Experiment 1 showed a significant decrease of global resting-state beta power after PEMS compared to sham (p = 0.016), with a median change from baseline of -16% for PEMS and -0.54% for sham. A cluster-based permutation test showed a significant difference in resting-state beta power comparing pre- and post-PEMS (p = 0.018) that was most pronounced over bilateral central and left frontal sensors. Experiment 2 did not identify a significant difference in the change from baseline of global EEG power for concurrent PEMS compared to delayed PEMS. Two cluster-based permutation tests suggested that frontal beta power may be increased following both concurrent and delayed PEMS. This study provides novel evidence for supraspinal effects of low-frequency PEMS and an initial indication that the presence of a cognitive task such as MI may influence the effects of PEMS on beta activity. Chronic pain has been associated with changes in beta activity, in particular an increase of beta power in frontal regions. Thus, brain state-dependent PEMS may offer a novel approach to the treatment of chronic pain. However, further studies are warranted to investigate optimal stimulation conditions to achieve a reduction of pain.

Keywords: electroencephalography; nociception; peripheral electrical stimulation; sensorimotor rhythm; state-dependent stimulation.

FIGURE 1

Overview of the study procedure. Experiment 1: Two stimulation conditions (verum and sham) were delivered during two separate stimulation sessions (session 1 = verum, session 2 = sham). The same overall procedure was followed for each session, starting with 5-min recording of resting-state EEG, followed by a total of 20 min of PEMS (verum) or sham stimulation, and finally another 5-min recording of resting-state EEG. Experiment 2: Two stimulation conditions (concurrent and delayed PEMS) were delivered during a single session, in two separate blocks. Order of stimulation conditions was randomized. In the concurrent PEMS condition, PEMS was applied during the MI phase of the trial. In the delayed PEMS condition, PEMS was applied during the Relax phase of the trial.

FIGURE 2

PEMS set up. The following equipment was used to deliver the PEMS for both experiments: (A) The PEMS stimulator (The Small Fiber Activator; Bomedus GmbH); (B) A circular matrix electrode (Bomedus GmbH). (C) Shows the placement of the matrix electrode on the left forearm. A bandage was wrapped around the electrode to ensure optimal contact between the electrode and the skin.

FIGURE 3

(A) Boxplots of the change from baseline scores ((EEG power post – EEG power pre)/EEG power pre) for the verum and sham PEMS condition, and for the theta, alpha, and beta band, separately. The ° and the * in the figure indicate outliers. (B) A topography plot of the average change from baseline of beta power for the PEMS condition (verum stimulation).

FIGURE 4

Topoplot for the cluster-based permutation test that compared resting-state EEG power (4–30 Hz) before and after verum stimulation. The plot shows the negative cluster (indicated by the “x” markers) that demonstrated a significant difference between EEG power before and after PEMS, plotted on top of the t-statistic of the difference as calculated for every frequency-electrode pair (repeated-measures t-test). The cluster showed a significant difference most pronounced over bilateral central and left frontal regions and in the frequency range of 23–26 Hz.

FIGURE 5

Topoplots for the cluster-based permutation test that compared resting-state EEG power (4–30 Hz) before and after concurrent and delayed stimulation. Each plot shows the largest positive cluster (indicated by the “x” markers) that demonstrated a difference between EEG power before and after PEMS, plotted on top of the t-statistic of the difference as calculated for every frequency-electrode pair (repeated-measures t-test). The cluster showed a difference most pronounced over frontal regions and in the frequency range of 17–21 Hz for concurrent PEMS and 16–25 Hz for delayed PEMS.