Electromyographic thresholds after thoracic screw stimulation depend on the distance of the screw from the spinal cord and not on pedicle cortex integrity

Abstract

Present studies concerning the safety and reliability of neurophysiological monitoring during thoracic pedicle screw placement remain inconclusive, and therefore, universally validated threshold levels that confirm osseous breakage of the instrumented pedicles have not been properly established. The objective of this work was to analyze whether electromyographic (EMG) thresholds, after stimulation of the thoracic pedicle screw, depend on the distance between the neural structures and the screws. The modifier effect of different interposed tissues between a breached pedicle and neural structures was also investigated. This experimental study uses a domestic pig model. Electromyographic thresholds were recorded after the stimulation of 18 thoracic pedicle screws that had been inserted into five experimental animals using varying distances between each screw and the spinal cord (8 and 2 mm). Electromyographic thresholds were also registered after the medial pedicle cortex was broken and after different biological tissues were interposed (blood, muscle, fat, and bone) between the screw and the spinal cord. Mean EMG thresholds increased to 14.1±5.5 mA for screws with pedicle cortex integrity that were placed 8 mm away from the dural sac. After the medial pedicle cortex was broken and without varying the distance of the screw to the dural sac, the mean EMG thresholds were not appreciably changed (13.6±6.3 mA). After repositioning the screw at a distance of 2 mm from the spinal cord and after medial cortical breach of the pedicle, the mean threshold significantly slowed to 7.4±3.4 mA (p<.001). When the screw was placed in contact with the spinal dural sac, even lower EMG thresholds were registered (4.9±1.9, p<.001). Medial pedicle cortex rupture and the interposition of different biological tissues in experimental animals did not alter the stimulation thresholds of the thoracic pedicle screws. In the experimental animals, the observed electrical impedance depended on the distance of screws from the neural structures and not on the integrity of the pedicle cortex. The screw-triggered EMG technique did not reliably discriminate the presence or absence of bone integrity after pedicle screw placement. The response intensity was not related to the type of interposed tissue.