Recording Diffusion Responses From Contralateral Intercostal Muscles After Stimulus-Triggered Electromyography

Abstract

A new stimulus-triggered electromyography (EMG) test for detecting stimulus diffusion to contralateral intercostal muscles during thoracic pedicle screw placement was assessed in a porcine model. To determine if electromyographic thresholds in the intercostal muscles of both sides of the thorax could discriminate thoracic pedicle screw malpositions with and without neural contact at different aspects of the spinal cord and nerve roots. There is controversy about the value of triggered EMG stimulation for aiding precise insertion of thoracic pedicle screws. A universally validated threshold that confirms screw malposition has not been established. Diffusion of EMG responses to the contralateral intercostal muscles has not previously been investigated. Nine domestic pigs weighing 60 to 75 kg had 108 pedicle screws placed bilaterally in the thoracic spine from T8-T13. Before spine instrumentation, neural structures were stimulated in 4 animals under direct vision at different anatomic locations from T9-T12. Recording electrodes were placed over the right and left intercostal muscles. Increasing stimulus intensity was applied until muscle response was detected at the contralateral side (EMG diffusion phenomenon). After this first experiment, the thoracic spine was instrumented in all 9 animals. Screws were placed in the pedicle in different positions, the anatomic intrapedicular location and within the spinal canal, with and without contact with the neural elements. Response thresholds to direct nerve root stimulation were significantly lower than those obtained by stimulation of the dorsal aspect of the spinal cord (0.44 +/- 0.22 mA vs. 1.38 +/- 0.71 mA, P < 0.01). However, a 14-fold stimulation intensity (6.50 +/- 0.29 mA) was necessary to obtain diffusion of the EMG response to the opposite (left) side if the right nerve root was stimulated. A 2-fold increment (3.17 +/- 0.93 mA) elicited diffusion of the EMG responses to the contralateral side when stimulation was applied to the dorsal aspect of the spinal cord. EMG recordings of the 108 stimulated screws showed a significant decrease in the EMG response when the screw was in contact with the spinal cord (2.72 +/- 1.48 mA; P < 0.01) compared with that found when the pedicle track was intact (mean: 5.01 +/- 1.89 mA). Screws violating the medial wall of the pedicle, but not touching neural tissues, responded to slightly lower intensities than well-positioned screws, but this was not statistically significant (3.91 +/- 1.39 mA vs. 4.89 +/- 1.30 mA, P > 0.05). Stimulus-triggered EMG can identify screws that violate the medial pedicle wall if they are in contact with neural tissues. EMG thresholds could not discriminate screws that violated the medial pedicle wall without neural contact from screws with accurate intraosseous placement. However, recording EMG potentials at the contralateral intercostal muscles (stimulus diffusion phenomenon) proved to be a reliable method for identifying the neural structures at risk.