High-Resolution, Spinal Cord Stimulation for Non-Opioid Treatment of Neuropathic Pain (U44NS115111)

This data was collected over a two-year period using a high-resolution spinal cord stimulation paddle (60-64 electrodes) for 21 chronic pain patients in two phases. This study was limited to a 25-minute intraoperative neuromonitoring session under general anesthesia. The electrodes were stimulated using a tri-polar configuration with a 60Hz frequency and 300us (microsecond) pulse width at a 0-10mA current from the T6 to T11 thoracic spinal level. Amplitudes were gradually increased by 0.5mA steps until significant EMG activity was observed on IONM, or until a maximum stimulation amplitude of 10mA was reached. Data from 18 muscle groups (left and right) were collected as follows: Upper and lower rectus abdominis (upper abdominals, UAB; lower abdominals, LAB), quadriceps (QUAD), adductor magnus (ADD), tibialis anterior (TA), adductor hallucis (AH), gluteus maximus (GLUT), bicep femoris (BF), and medial gastrocnemius (MG). All signals were processed offline. Fluoroscopic imaging was used to determine the anatomic midline position for the paddles, and more details are available in the Excel data files. The patient demographic details are available in patient level data file. The EMG signals were denoised using a previously developed algorithm by setting a 10% threshold and removing values exceeding it. The root mean square (RMS) of the signals was computed at each amplitude (stimulation-ON) for all patients and normalized with respect to the baseline (stimulation-OFF). Normalized RMS (% change) = [(RMS_(Stim-ON) - RMS_(Stim-OFF))/RMS_(Stim-OFF)]*100