High-definition transcranial direct current stimulation desynchronizes refractory status epilepticus

Abstract

Recently, we showed that high-definition transcranial direct current stimulation (hd-tDCS) can acutely reduce epileptic spike rates during and after stimulation in refractory status epilepticus (RSE), with a greater likelihood of patient discharge from the intensive care unit compared to historical controls. We investigate whether electroencephalographic (EEG) desynchronization during hd-tDCS can help account for observed anti-epileptic effects. Defining desynchronization as greater power in higher frequencies such as above 30 ​Hz ("gamma") and lesser power in frequency bands lower than 30 ​Hz, we analyzed 27 EEG sessions from 10 RSE patients who had received 20-minute session(s) of 2-milliamperes of transcranial direct current custom-targeted at the epileptic focus as previously determined by a clinical EEGer monitoring the EEG in real-time. During hd-tDCS, median relative power change over the EEG electrode chains in which power changes were maximal was +4.84%, -5.25%, -1.88%, -1.94%, and +4.99% for respective delta, theta, alpha, beta, and gamma frequency bands in the bipolar longitudinal montage (p ​= ​0.0001); and +4.13%, -5.44%, -1.81%, -3.23%, and +5.41% in the referential Laplacian montage (p ​= ​0.0012). After hd-tDCS, median relative power changes reversed over the EEG electrode chains in which power changes were maximal: -2.74%, +4.20%, +1.74%, +1.75%, and -4.68% for the respective delta, theta, alpha, beta, and gamma frequency bands in the bipolar longitudinal montage (p ​= ​0.0001); and +1.59%, +5.07%, +1.74%, +2.40%, and -5.12% in the referential Laplacian montage (p ​= ​0.0004). These findings are consistent with EEG desynchronization through theta-alpha-beta-gamma bands during hd-tDCS, helping account for the efficacy of hd-tDCS as an emerging novel anti-epileptic therapy against RSE.

Keywords: Desynchronization; Electroencephalography (EEG); High-definition transcranial direct current stimulation (hd-tDCS); Neuromodulation; Refractory status epilepticus (RSE); Relative power.

Fig. 1

(A) Maximal relative power changes during vs. before stimulation. (B) Maximal relative power changes after vs. during stimulation. Bars represent the median of the maximal relative power change within an electrode chain per patient session within a given frequency band. Dots represent the mean of the maximal relative power change within an electrode chain per patient session within a given frequency band. Blue bars or dots represent the bipolar longitudinal montage. Orange bars or dots represent the Laplacian montage.

Fig. 2

Relative powers derived from the ‘Pwelch’ analysis in 10 ​s increments within the bipolar longitudinal montage filtering. Every row of images represents a different patient-session while every column represents a different frequency band ranging from left to right: delta (1–4 ​Hz), theta (4–8 ​Hz), alpha (8–12 ​Hz), beta (12–30 ​Hz), and gamma (30–100 ​Hz). Within each individual image, there are 4 distinct bipolar chains ranging from left to right: left parasagittal, right parasagittal, left temporal, and right temporal. Also, within each image, there are blue, black, and orange lines which represent the relative power for the pre-stimulation, during-stimulation, and post-stimulation respectively. Sessions chosen provided a comprehensive representation of inter-individual and inter-sessional variability of the entire dataset.

Fig. 3

Relative powers derived from the ‘Pwelch’ analysis in 10 ​s increments within the Laplacian montage filtering. Every row of images represents a different patient-session while every column represents a different frequency band ranging from left to right: delta (1–4 ​Hz), theta (4–8 ​Hz), alpha (8–12 ​Hz), beta (12–30 ​Hz), and gamma (30–100 ​Hz). Within each individual image, there are 5 distinct Laplacian chains ranging from left to right: left temporal, left parasagittal, midline, right parasagittal, and right temporal. Also, within each image, there are blue, black, and orange lines which represent the relative power for the pre-stimulation, during-stimulation, and post-stimulation respectively. Sessions chosen provided a comprehensive representation of inter-individual and inter-sessional variability of the entire dataset.