Non-invasive recording of high-frequency signals from the human spinal cord

• Published in: NeuroImage (Orlando, Fla.) Vol. 253; p. 119050
• Main Authors: Chander, Bankim Subhash, Deliano, Matthias, Azañón, Elena, Büntjen, Lars, Stenner, Max-Philipp
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.06.2022; Elsevier Limited; Elsevier
• Subjects: EEG; Electric Stimulation; Electrodes; Electroencephalography; Epidural; Evoked Potentials, Somatosensory - physiology; Firing pattern; High-frequency oscillations; Humans; Information processing; Latency; Median nerve; Median Nerve - physiology; Median nerve stimulation; Nervous system; Neuralgia; Neurophysiology; Pain; Pathology; Population; Recording sessions; Sensory integration; Somatosensory system; Spike-like potentials; Spinal cord; Spinal Cord - physiology; Spike-like potentials; Spinal cord; SCS; ERP; Median nerve stimulation; DSS; EEG; Electroencephalography; Somatosensory system; MEG; HFO; DBS; LFP; High-frequency oscillations; SEP
• ISSN: 1053-8119; 1095-9572
• DOI: 10.1016/j.neuroimage.2022.119050

Throughout the somatosensory system, neuronal ensembles generate high-frequency signals in the range of several hundred Hertz in response to sensory input. High-frequency signals have been related to neuronal spiking, and could thus help clarify the functional architecture of sensory processing. Recording high-frequency signals from subcortical regions, however, has been limited to clinical pathology whose treatment allows for invasive recordings. Here, we demonstrate the feasibility to record 200–1200 Hz signals from the human spinal cord non-invasively, and in healthy individuals. Using standard electroencephalography equipment in a cervical electrode montage, we observed high-frequency signals between 200 and 1200 Hz in a time window between 8 and 16 ms after electric median nerve stimulation (n = 15). These signals overlapped in latency, and, partly, in frequency, with signals obtained via invasive, epidural recordings from the spinal cord in a patient with neuropathic pain. Importantly, the observed high-frequency signals were dissociable from classic spinal evoked responses. A spatial filter that optimized the signal-to-noise ratio of high-frequency signals led to submaximal amplitudes of the evoked response, and vice versa, ruling out the possibility that high-frequency signals are merely a spectral representation of the evoked response. Furthermore, we observed spontaneous fluctuations in the amplitude of high-frequency signals over time, in the absence of any concurrent, systematic change to the evoked response. High-frequency, “spike-like” signals from the human spinal cord thus carry information that is complementary to the evoked response. The possibility to assess these signals non-invasively provides a novel window onto the neurophysiology of the human spinal cord, both in a context of top-down control over perception, as well as in pathology.