Abstracts

Rationale: Methods for the cancellation of the stimulation artifact during intracranial stimulation have as main limitation the inability to completely differentiate between artifacts and physiological responses. Tissue response to electrical stimulation is nonlinear, introducing out-of-band spectral components in the recorded signal. By further modulating the stimulation waveform, we aim at completely separating the spectral components of the artefactual and physiological components of the signal, allowing for a full recovery of the responses to stimulation. Methods: Using a clinical stimulator capable of generating arbitrary waveforms (Guideline4000LP+, FHC, Bowdoin, ME), we have designed a stimulation protocol in which the polarity of the biphasic pulses in a 50Hz train used for performing functional stimulation in patients undergoing stereo-EEG investigations, is inverted every pulse. By virtue of the modulation theorem in Fourier analysis, the frequency spectrum of such a waveform no longer has the 50 Hz component, which is split in 25 Hz and 75 Hz components and odd harmonics. Artefactual components, due to capacitive coupling and volume conduction, are in a linear relationship with the stimulation waveform, therefore they will have a similar spectral content. However, the axonal propagation of the excitation evokes pulses having the same polarity, regardless the polarity of the applied pulses. As a result, the responses will have a 50 Hz component, plus harmonics. The artefactual and physiological responses will have entirely disjunct spectral content, allowing for a full separation of the two components. We have applied the new stimulation protocol while performing functional mapping in two patients undergoing presurgical evaluation for drug-resistant epilepsy using stereo-EEG method. Results: The modulation of the pulse trains allowed a full recovery of the physiological responses during 50 Hz stimulation used for mapping functional cortex. The thresholds for evoking a clinical symptom were the same as for the standard stimulation protocol. A differential analysis based on the responses at stimulation levels that evoked a clinical symptom, compared to the responses at a sub-clinical level, allowed evidencing the recruitment of a subset of functional connections related to a particular clinical effect. Conclusions: Modulating the stimulation waveforms, in conjunction with the nonlinear response of the tissue, allows recovering the physiological responses to intracranial high-frequency stimulation. Effective brain connectivity in relationship with a particular clinical effect can be evidenced. This supports that direct electrical stimulation produce both local changes and also act as “inputs gates” into distant functional brain networks.