49. EMG to detect early motor tract injury during resection of intramedullary spinal cord lesions; How reliable is this form of monitoring?

Abstract

Inadvertent spinal cord impact generates trains of descending pulses which can depolarize motor neurons caudally. Similar to intended electrical spinal cord stimulation, these pulses may traverse along the corticospinal tract … or they may descend via non-corticospinal tracts. In porcine models, hindlimb EMG “injury discharges” may be recorded by direct intramedullary mechanical irritation or rapidly actuated extradural spinal cord impact. The key element is the high end velocity of the stimulus. With repeated/prolonged impact (generating “injury discharges”), MEP recording fails. Spinal cord heating (electrocautery/laser) results in similar EMG/MEP findings. EMG activity, predictive of MEP loss, can be recorded in multiple clinical situations: tumor resection near the motor strip, head/neck positioning, and decompression of the spinal cord at multiple sites. We have rarely seen central EMG activity with deformity correction. This is probably related to slow distraction or compression which is less likely to generate intraspinal depolarization. The external anal sphincter enjoys rich innervation by sacral motor neurons which seem to fire repetitively when excited. Therefore, sphincter recordings add a heightened level of EMG surveillance. We have labeled central EMG injury discharges: “suprasegmentally-generated EMG discharges” (SEDs). Central EMG monitoring adds a neuronavigation component to spinal cord IOM.