Abstract Purpose of study: Somatosensory evoked potential (SSEP) monitoring has enjoyed widespread application for assessing global spinal cord function during corrective thoracic spine surgery for scoliosis and trauma. Because the SSEP is dorsal column mediated, however, it can be insufficiently sensitive to identifying inadequate perfusion or impending ischemia to the anterior spinal cord. Transcranial electric motor-evoked potentials (TCEMEPs), on the other hand, permit real-time monitoring of the corticospinal tracts and, as such, provide instantaneous assessment of changes in spinal cord function secondary either to surgical insult or inadequate spinal cord blood flow from hypotension. The purpose of this study was to determine the sensitivity of TCEMEP monitoring as an early indicator of impending spinal cord injury during corrective thoracic spine surgery. Methods used: A retrospective chart review was performed for all adult patients who underwent thoracic or thoracolumbar spine surgery between January 2000 and December 2001 at a single institution. Perioperative neurophysiological monitoring included recording of posterior tibial nerve SSEPs and TCEMEPs (Digitimer, Ltd., Garden City, UK) recorded from hand, leg and foot muscles. Monitoring commenced after intubation and continued throughout the entire operative procedure. All patients who experienced a significant (greater than 75%) intraoperative TCEMEP or SSEP amplitude loss were identified. Hospital and office charts were also reviewed to obtain diagnosis, preoperative, immediate postoperative and follow-up neurologic data. of findings: Of 125 cases reviewed, significant intraoperative TCEMEP amplitude decreases that prompted surgical alert were noted in seven (5.6%). A concomitant significant SSEP amplitude loss was noted in only one of these patients and lagged behind the loss of TCEMEPs by 20 minutes. Of the remaining six patients, TCEMEP amplitude diminished by greater than 75% during placement of corrective hardware in four, whereas that for the two patients was secondary to acute hypotension (MAP < 50 mm Hg). SSEP amplitudes never changed across all six of these patients. When TCEMEP amplitude was significantly depressed, intraoperative intervention included reversal of corrective forces or complete hardware removal, increasing mean arterial blood pressure to al least 90 torr and administering high-dose methylprednisolone (NASCIS-2 protocol). Improvement in spinal cord motor function occurred in seven of the eight patients after surgical and/or medical intervention. None of these presented with postoperative neurologic deficit. The one patient who lost transcranial motor-evoked potentials after placement of thoracic pedicle screws, followed 20 minutes later by complete SSEP loss, never regained neurophysiological function and awoke with dense paraplegia as predicted. Relationship between findings and existing knowledge: These data suggest that TCEMEPs are a more sensitive and earlier indicator of impending spinal cord ischemic injury then somatosensory evoked potentials. Overall significance of findings: This study highlights the feared limitation of SSEPs for spinal cord monitoring; namely, the potential for false-negative results in the presence of impending spinal cord ischemic injury. As a result, TCEMEPs are recommended as the primary monitoring modality with SSEPs serving adjunctively. Disclosures: Device or drug: Digitimer D-185 Multi-pulse Motor Evoked Potential Stimulator. Status: investigational. Conflict of interest: No conflicts.