Abstract
Synopsis
Non-invasive electrical stimulation using Temporal Interferential (TI) stimulation can be leveraged in neuromodulation of deep brain regions where the two high-frequency (kHz) electric fields intersect, and form Amplitude-Modulated waveform with low beating frequency (tens of Hz). Superficial regions are exposed to higher-magnitude sine waveform with a steady amplitude. The cellular mechanism of TI stimulation is thus predicated on special sensitivity to Amplitude-Modulated waveforms. We analyzed the changes in gamma oscillations induced by carbachol in hippocampal slices during application of Amplitude-Modulated waveforms compared to high and low frequency sine electric fields, the latter associated with conventional tACS.
Background
It has been suggested that non-invasive electrical stimulation using Temporal Interference (TI) stimulation can be leveraged in neuromodulation of deep brain regions where the two high-frequency (kHz) electric fields intersect, and form Amplitude-Modulated (AM) waveform with low beating frequency. Superficial regions are stimulated with higher-magnitude sine waveform whereas deep brain structures are exposed to AM electric field. Explaining the cellular mechanism of Amplitude-Modulated high-frequency stimulation is thus of interest in developing rational and optimized stimulation strategies.
Methods
The effect of Amplitude-Modulated high-frequency electric fields on ongoing network activity was evaluated in carbachol-induced gamma oscillation in CA3a region of rat hippocampal slices. Extracellular local field potentials were analyzed before, during and after 2 s of stimulation. We tested unmodulated sine waveforms with low frequency (i.e. 5 Hz), kHz frequency (i.e. 1 and 2 kHz) and 5 Hz Amplitude-Modulated waveforms with 0.1, 1, and 2 kHz carrier frequencies. The hippocampal brain slice model of gamma oscillation provided a rapid experimental system to quantify the degree of modulation produced by amplitude-modulated high-frequency stimulation as well as control waveforms of unmodulated kHz stimulation and low-frequency sinusoidal stimulation. A validated MRI-derived head model predicted both peak electric fields in the cortex and depth of Amplitude-Modulation at deep regions during IF. We further developed a computational model of excitatory/inhibitory neuronal network with conductance-based synaptic interactions.
Results
Consistent with previous studies, stimulation with 5 Hz sine waveform induced modulation of ongoing gamma activity with frequency of stimulation using field intensities > 1 V/m. kHz sine waveforms produced a steady increase in gamma activity using field intensities ≥ 10 V/m for 100 Hz, 60 V/m for 1 kHz and 80 V/m for 2 kHz. Stimulation at 5 Hz AM waveform with different carrier frequencies (i.e. 100, 1 kHz, 2 kHz) induced modulation of ongoing oscillation with frequency of 5 Hz (i.e. at the beating frequency) using field intensities significantly higher compared to unmodulated 5 Hz sine waveform (≥ 5 V/m for AM-100 Hz, 40 V/m for AM-1 kHz, 60 V/m for AM-2 kHz). Results show a tradeoff between carrier frequency in AM-waveform and field intensity to produce the same level of modulation of oscillations explained by membrane low-pass characteristics. The strength of modulation by any waveform increased with increasing filed magnitude for field intensities above respective thresholds.
Discussion
Based on experimentally-constrained current flow and biophysics, we predict selective deep neuromodulation by TI stimulation depends on the amplitude modulation of ongoing oscillations, as opposed to steady increase in superficial regions, and TI stimulation requires applied current significantly above low-frequency tACS.
Disclosure
The City University of New York has inventions on tES with MB and LCP as inventor. MB and LCP have equity in Soterix Medical. MB serves on the scientific advisory boards of Boston Scientific and GlaxoSmithKline. Other authors reported no conflict of interest.
Share Poster