Intraoperative neuromonitoring: can the results of direct stimulation of titanium-alloy pedicle screws in the thoracic spine be trusted?

Objective: Intraoperative neuromonitoring of thoracic-level pedicle screw implantation for detecting breaches in the pedicle cortex has adopted methods originally developed in the early 1990s for stainless steel (SS) alloy screws used at lumbosacral levels. In our recent attempts to monitor thoracic-level pedicle screw placement, we were surprised to find that these widely used stimulation parameters were largely ineffectual when stimulating directly through titanium alloy (Ti-alloy) pedicle screws. The objectives of this study, then, were twofold: (1) to report the number of episodes in which intraoperative neuromonitoring of thoracic screw position failed to detect a medially directed breach (or malplacement) in a previously described and limited sample set; and (2) to compare the frequency-specific impedance of a sample of Ti-alloy pedicle screws to comparably sized screws made of SS alloys. We predicted that Ti-alloy screws would demonstrate impairment in conduction properties that could help explain the difficulties we, and others, have recently experienced with neuromonitoring of thoracic pedicle screw placement. Methods: Based on threshold values for train-of-four stimulation of spinal motor pathways, we quantified the incidence of medial breaches of thoracic-level pedicles in a small cohort of subjects. We also evaluated the conductive properties of Ti-alloy pedicle screws and compared these with SS screws. Eleven pedicle screws were examined using energy-dispersive x-ray spectroscopy to identify their alloys, after which DC resistance and AC impedance for each screw was measured. Furthermore, a subset of five screws was used to investigate the current delivery under dynamic testing conditions. Results: Postoperative computed tomography of 6 subjects revealed 10 instances of significant medial screw malpositioning, out of a total of 88 screws placed. In each of these 10 instances, direct stimulation of thoracic pedicle screws at intensities considered in the literature to be clinically significant (i.e., <= 11 mA) failed to predict these medial pedicle breaches, yet each breach was reliably identified with low-intensity stimulation applied via a ball-tipped probe. For in vitro studies, most screws made of titanium alloys had higher resistance and impedance at tested frequencies compared with their SS counterparts. Moreover, there was widespread variability in conduction properties between Ti-alloy screws, whereas SS screws behaved in a more homogeneous manner. Conclusions: When compared with screws made of SS, most Ti-alloy pedicle screws behaved more like semiconductors, showing conduction properties that were highly frequency dependent. These properties likely contributed to the difficulties we encountered in interpreting thoracic screw placements based on stimulus-evoked electromyography from direct screw stimulation.