108 Human Electrocorticography-Based Stimulation

Abstract

INTRODUCTION: Numerous experiments using non-invasive cortical stimulation have demonstrated transient increases in motor cortical plasticity and improved functional rehabilitation in humans. We examined the feasibility and the effects of performing repetitive human electrocorticography (ECoG) based stimulation in an activity-dependent manner. METHODS: Four epileptic patients undergoing subdural ECoG grid implants were enrolled in this study. Electrical stimuli were triggered by spontaneous movement leading to upper extremity electromyographic (EMG) activity, or increases in ECoG high gamma (HG) power overlying motor cortex, exclusively. An FDA-approved EMG stimulator was used to provide 5000 stimuli over 30 minutes; afterwards, ECoG recordings were taken while the subject performed cued motor tasks. No subjects experienced any adverse events. RESULTS: A consistent decrease in event-related high-gamma activity after conditioning was seen. One subject, who had received 2 sessions of 5000 stimuli, also demonstrated a broadband increase in HG power and a decrease in low frequency band (LFB) power during the resting state. These spectral changes were more apparent at sites closer to the stimulated electrodes, and lasted for at least 30 minutes. No significant changes in resting motor thresholds or correlation values between stimulation sites and the rest of the grid were noted. Latencies calculated for EMG-trigger to stimulation pulse delivery were normally distributed with a mean of 31-32 ms. CONCLUSION: These surprising results suggest that, with this stimulation paradigm, repetitive electrical stimulation caused blunting of the normal HG response seen during motor movement. In the subject with the sustained broadband resting state changes, the prolonged effects perhaps suggest physiological phenomena such as changes in neuronal resistivities or local neurotransmitter levels, or potentially reflect local charge build-up on the ECoG grid. This experiment demonstrates that invasive activity-driven human cortical stimulation is safe, feasible to accomplish within the window for spike timing dependent plasticity, and underlines the need for further investigation.