Intraoperative Use of Neuromonitoring in Multilevel Thoracolumbar Spine Instrumentation and the Effects on Postoperative Neurological Injuries.

STUDY DESIGN: Retrospective cohort analysis of a national database between 2005 and 2011.

OBJECTIVE: To investigate the current usage of neuromonitoring in patients undergoing multilevel thoracolumbar spine surgery. We hypothesize that the use of neuromonitoring would be associated with a reduced incidence of postoperative neurological injuries.

SUMMARY OF BACKGROUND DATA: Intraoperative neuromonitoring is a common technique utilized in spine surgery to improve safety and reduce neurological injuries. However, the literature remains unclear in defining the populations that benefit from the use of neuromonitoring.

METHODS: The PearlDiver Medicare database was queried to identify patients undergoing multilevel thoracolumbar spine instrumentation (defined as >3 thoracolumbar levels) from 2005 to 2011. The use of neuromonitoring was identified by Current Procedural Terminology codes. Neurological injuries were identified by codes from the International Classification of Diseases, Ninth Revision.

RESULTS: Within 15,032 patients, the postoperative rate of neurological injury diagnosis was higher when neuromonitoring was used at both 1 week (1.3% vs. 1.0%, P=0.06) and 6 months (5.9% vs. 4.6%, P=0.0005). However, a lower incidence of neurological injury was associated with neuromonitoring in patients undergoing specifically anterior fusion of 4-7 levels, posterior fusion of 7-12 levels, and in adults below 65 years old (P=0.0266, 0.0458, 0.032).

CONCLUSION: Within the total Medicare cohort, the use of neuromonitoring was not associated with a decreased rate of neurological injury in multilevel thoracolumbar instrumentation procedures. This is likely due to the possible selection and detection bias of utilizing neuromonitoring when there is an increased risk of neurological injury based on patient-specific pathology and/or surgical procedure. However, despite the overall potential bias, it was appreciated that in subgroups: age below 65 years old, anterior fusion of 4-7 segments, and posterior fusion of 7-12 segments, there was a statistically significant reduction in the incidence of neurological injuries with neuromonitoring.

LEVEL OF EVIDENCE: Level III.