Predictive Value of Transcranial Evoked Potential Monitoring for Intramedullary Spinal Cord Tumors.

Background Intraoperative neurophysiologic monitoring (IONM) has increased patient safety and extent of resection in patients with eloquent brain tumors. Despite its comprehensive capability for the resection of intramedullary spinal cord tumors (ISCTs), the application during the resection of these tumors is controversial. Methods We retrospectively analyzed the resection of ISCTs in 83 consecutive cases. IONM was performed in all cases. Each patient's motor status and the McCormick scale was determined preoperatively, directly after surgery, at the day of discharge, and at long-term follow-up. Results IONM was feasible in 71 cases (85.5%). Gross total resection was performed in 75 cases (90.4%). Postoperatively, patients showed new transient deficits in 12 cases (14.5%) and new permanent deficits in 12 cases (14.5%). The mean McCormick variance between baseline and long-term follow-up was − 0.08 ± 0.54. IONM's sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for the patient's motor status at the day of discharge was 75.0%, 64.7%, 45.5%, and 86.8%. It was 88.9%, 59.7%, 24.2%, and 97.4% for the motor outcome at long-term follow-up. Patients experienced postoperative complications in 15 cases (18.1%). Conclusion IONM, as performed in the present study, shows a high sensitivity and NPV but low specificity and PPV, particularly for the patient's motor status at the long-term follow-up. As far as practicable by a retrospective study on IONM, our results confirm IONM's usefulness for its application during the resection of ISCTs. However, these results must be approved by a prospective study.