Intraoperative somatosensory evoked potential monitoring in scoliosis.

During surgical correction of scoliosis, 63 patients had somatosensory evoked potential (SEP) monitoring of the spinal cord. Tibial nerves were unilaterally stimulated, and the potentials were recorded from the midcervical spine with surface and epidural needle electrodes. Over 85% had no significant change in their SEP and no postoperative neurologic deficits. Eleven percent had a significant change in their potential (amplitude decrease of greater than 60% and/or latency increase of greater than 2.5 msec) with no neurologic complications. One patient had a significant potential change and temporary postoperative sensory deficits did occur. One additional patient experienced postoperative neurologic complications but had no SEP change. This single false negative case, however, was clinically significant only for motor dysfunction, which is not monitored by the SEP. When changes in patient core temperature were compared to changes in SEP amplitude and latency, an intraoperative decrease in core temperature increased SEP latency and decreased amplitude, which may explain in part the false positive rate of the procedure.