Use of motor evoked potentials during lateral lumbar interbody fusion reduces postoperative deficits.

BACKGROUND CONTEXT

Intraoperative neurophysiological monitoring (IONM) has gained rather widespread acceptance as a method to mitigate risk to the lumbar plexus during lateral lumbar interbody fusion (LLIF) surgery. The most common approach to IONM involves using only electromyography (EMG) monitoring, and the rate of postoperative deficit remains unacceptably high. Other test modalities, such as transcranial electric motor-evoked potentials (tcMEPs) and somatosensory-evoked potentials, may be more suitable for monitoring neural integrity, but they have not been widely adopted during LLIF. Recent studies have begun to examine their utility in monitoring LLIF surgery with favorable results.

PURPOSE

This study aimed to evaluate the efficacy of different IONM paradigms in the prevention of iatrogenic neurologic sequelae during LLIF and to specifically evaluate the utility of including tcMEPs in an IONM strategy for LLIF surgery.

STUDY DESIGN/SETTING: A non-randomized, retrospective analysis of 479 LLIF procedures at a single institution over a 4-year period was conducted. During the study epoch, three different IONM strategies were used for LLIF procedures: (1) surgeon-directed T-EMG monitoring ("SD-EMG"), (2) neurophysiologist-controlled T-EMG monitoring ("NC-EMG"), and (3) neurophysiologist-controlled T-EMG monitoring supplemented with MEP monitoring ("NC-MEP").

PATIENT SAMPLE

The patient population comprised 254 men (53.5%) and 221 women (46.5%). Patient age ranged from a minimum of 21 years to a maximum of 89 years, with a mean of 56.6 years.

OUTCOME MEASURES

Physician-documented physiological measures included manual muscle test grading of hip-flexion, hip-adduction, or knee-extension, as well as hypo- or hyperesthesia of the groin or anterolateral thigh on the surgical side. Self-reported measures included numbness or tingling in the groin or anterolateral thigh on the surgical side.

METHODS

Patient progress notes were reviewed from the postoperative period up to 12 months after surgery. The rates of postoperative sensory-motor deficit consistent with lumbar plexopathy or peripheral nerve palsy on the surgical side were compared between the three cohorts.

RESULTS

Using the dependent measure of neurologic deficit, whether motor or sensory, patients with NC-MEP monitoring had the lowest rate of immediate postoperative deficit (22.3%) compared with NC-EMG monitoring (37.1%) and SD-EMG monitoring (40.4%). This result extended to sensory deficits consistent with lumbar plexopathy (pure motor deficits being excluded); patients with NC-MEP monitoring had the lowest rate (20.5%) compared with NC-EMG monitoring (34.3%) and SD-EMG monitoring (36.9%). Additionally, evaluation of postoperative motor deficits consistent with peripheral nerve palsy (pure sensory deficits being excluded) revealed that the NC-MEP group had the lowest rate (5.7%) of motor deficit compared with the SD-EMG (17.0%) and NC-EMG (17.1%) cohorts. Finally, when assessing only those patients whose last follow-up was greater than or equal to 12 months (n=251), the rate of unresolved motor deficits was significantly lower in the NC-MEP group (0.9%) compared with NC-EMG (6.9%) and SD-EMG (11.0%). A comparison of the NC-MEP versus NC-EMG and SD-EMG groups, both independently and combined, was statistically significant (>95% confidence level) for all analyses.

CONCLUSIONS

The results of the present study indicate that preservation of tcMEPs from the adductor longus, quadriceps, and tibialis anterior muscles are of paramount importance for limiting iatrogenic sensory and motor injuries during LLIF surgery. In this regard, the inclusion of tcMEPs serves to compliment EMG and allows for the periodic, functional assessment of at-risk nerves during these procedures. Thus, tcMEPs appear to be the most effective modality for the prevention of both transient and permanent neurologic injury during LLIF surgery. We propose that the standard paradigm for protecting the nervous system during LLIF be adapted to include tcMEPs.