Differences in sensitivity between magnetic motor-evoked potentials and somatosensory-evoked potentials in experimental spinal cord lesions

Spine (Phila Pa 1976). 1996 Oct 1;21(19):2190-6. doi: 10.1097/00007632-199610010-00003.


Study design: Graded posterior spinal cord compression and partial sectioning of the spinal cord were performed, and magnetically induced descending spinal cord potentials were recorded.

Objectives: To compare the sensitivity of transcranial magnetic motor-evoked potentials in the spinal cord and other spinal cord evoked potentials to spinal cord lesions.

Summary of background data: Somatosensory-evoked potentials have been the standard technique for monitoring spinal cord function during spinal surgery. These potentials, however, do not necessarily reflect descending motor tract function. Transcranial electric or magnetic stimulation for motor-evoked potentials is a more direct measure of motor tract function. However, more research on magnetic motor-evoked potentials is needed.

Methods: Fifteen adult cats were used. Graded posterior spinal cord compression was performed at L2 in 10 cats, and partial sectioning of the spinal cord was performed at L2 in five cats. The location sequence of lesioning was dorsal column section, dorsal 1/2 section, dorsal 2/3 section, and total spinal cord section. Magnetic motor-evoked potentials were recorded by epidural catheter electrodes placed above at, and below the lesion. Electric motor evoked potentials and spinal and cortical somatosensory-evoked potentials were recorded serially for a comparison of their sensitivity to spinal cord dysfunction.

Results: In posterior spinal cord compression, N1 amplitude of magnetic motor-evoked potentials at and below the lesion decreased after 1 minute of compression with a 70 g weight, and N1 and N2 amplitude disappeared after 1 minute of compression with a 100 to 120-g weight. Electric motor-evoked potentials changed at amplitudes comparable with those shown by magnetic motor-evoked potentials. Spinal somatosensory-evoked potentials showed the most sensitive changes to spinal cord posterior compression and disappeared after 1 minute of compression with 80 g Cortical somatosensory-evoked potentials in five cats were not sensitive enough for spinal cord posterior compression injury and did not disappear even after 1 minute of compression with 120 g. In magnetic motor-evoked potentials, after dorsal hemisectioning of the spinal cord only N3 disappeared; N1 and N2 disappeared after ventral spinal cord sectioning in spinal somatosensory-evoked potentials, amplitudes decreased after dorsal column sectioning, and all negative peaks disappeared after dorsal hemisectioning of the spinal cord.

Conclusions: Posterior compression injuries are diagnosed more easily with spinal somatosensory-evoked potentials. Motor-evoked potentials were slightly less sensitive, but they were significantly more useful in diagnosing posterior compression injuries than were cortical somatosensory-evoked potentials.

Publication types

  • Comparative Study

MeSH terms

  • Animals
  • Cats
  • Electric Stimulation / methods
  • Evoked Potentials, Motor / physiology*
  • Evoked Potentials, Somatosensory / physiology*
  • Sensitivity and Specificity
  • Spinal Cord Compression / physiopathology
  • Spinal Cord Injuries / physiopathology*