Object: Motor evoked potentials are widely used for intraoperative spinal cord monitoring. However, there are problems with anesthetic constraints and high trial-by-trial variability of compound muscle action potential amplitude in muscle motor evoked potential monitoring. It is difficult to determine when to warn the surgeon of an occurrence of spinal cord risk. A method of estimation for motor function in the spinal cord has not been established. To monitor spinal cord function with reliable evoked potentials, including the upper cervical spinal cord and the ventral spinal cord, the authors developed a nasopharyngeal tube electrode that can be placed in front of the upper and ventral cervical spinal cord. The purpose of this study was to investigate the origins and pathways of descending or ascending spinal cord evoked potentials (SCEPs) elicited with this electrode, and the usefulness and limitations of this method.
Methods: A nasopharyngeal tube electrode was inserted into the nostril. A catheter electrode was placed in the epidural or subarachnoid space at the thoracic spine. Ventral SCEP was recorded from the thoracic spinal cord after transpharyngeal stimulation, and dorsal SCEP was recorded with the nasopharyngeal electrode after thoracic spinal cord stimulation. There was no restriction of anesthetic technique in recording. When the amplitude of either of the SCEPs declined to 80% of the baseline, a warning was provided to the surgeon during the observed operative procedure. At the end of surgery, less than 50% or more than 30% of the baseline amplitude was considered a significant change in both SCEPs. The sensitivity and specificity for both SCEPs to detect neurological deterioration were calculated.
Results: The electrode provided noninvasive access to the ventral cervicomedullary junction. The SCEPs showed stable responses. A response change was only observed in situations involving a risky procedure for the spinal cord. Ventral SCEPs showed high sensitivity (73.1%) for identifying patients with new neurological deficits or an exacerbation of preexisting neurological deficits after surgery, but dorsal SCEPs showed lower sensitivity (46.1%) in the total number of cases. Both SCEPs showed high specificities. The sensitivities of ventral SCEP, dorsal SCEP, and either SCEP were 100.0%, 50.0%, and 100.0% for the upper cervical spinal cord, 33.3%, 0%, and 55.6% for the lower cervical spinal cord, and 77.8%, 64.7%, and 88.2% for the thoracic spinal cord.
Conclusions: Combined recording of both SCEPs estimated the ventral and dorsal white matter function in the spinal cord. Measuring the SCEPs with the nasopharyngeal electrode can be another useful approach for upper cervical and thoracic spinal cord monitoring. Ventral SCEP was more reliable for monitoring postoperative spinal cord function than dorsal SCEP. Ventral SCEP does not estimate the gray matter and spinal root functions in the lower cervical spinal cord.