Correlation of Clinical Outcomes to Intraoperative Transcranial Motor Evoked Potential during L4-L5 Lateral Lumbar Interbody Fusion

Introduction: Transpsoas lateral lumbar interbody fusion (LLIF) at the L4-5 level has historically been associated with the risk of femoral nerve injury. Spontaneous and triggered electromyography are routinely employed, but these neuromonitoring modalities do not provide data about the health of the femoral nerve during retraction of the lumbar plexus. This study aimed to assess the utility of intraoperative transcranial motor evoked potentials (tcMEP) in detecting postoperative femoral nerve injury following L4-L5 LLIF.

Methods: This study is a consecutive series of patients undergoing L4-L5 LLIF with tcMEP. A decrease in amplitude by 50% was defined as significant, and surgeons were alerted. Information about patient demographic variables, neuromonitoring data, and postoperative neurologic examinations were collected. The sensitivity, specificity, and area under the curve (AUC) for the receiver operating characteristics (ROC) curve were calculated. A simple logistic regression analyzed the relationship between a significant reduction in tcMEP and the presence of postoperative femoral nerve injury, defined as concurrent knee extension and hip flexion weakness.

Results: Sixty-two patients (69% female) were included with a mean age and follow-up of 62.8 ± 10.0 years and 13.7 ± 14.6 months, respectively. Seven patients had motor weakness at the time of the last follow-up. In an analysis of tcMEP, 24 of 62 patients (38.7%) had a significant decrease in tcMEP amplitude, as defined above. The sensitivity of tcMEP to detect persistent neurologic deficit was 29% (95% CI 3.7% to 71%), and the specificity was 60% (95% CI 46% to 73%). The AUC for the ROC curve of tcMEP amplitude reduction was 0.54. Logistic regression demonstrated no association with postoperative neurologic deficit with an odds ratio of 1.0 (95% CI 0.98 to 1.02).

Conclusion : Reduction in tcMEP amplitude by 50% during L4-L5 LLIF has poor accuracy in detecting persistent neurologic deficits.