The reliability of motor evoked potentials to predict dorsiflexion injuries during lumbosacral deformity surgery: importance of multiple myotomal monitoring

Abstract

STUDY DESIGNCase-control analysis of transcranial motor evoked potential (MEP) responses and clinical outcome. OBJECTIVETo determine the sensitivity and specificity of MEPs to predict isolated nerve root injury causing dorsiflexion weakness in selected patients having complex lumbar spine surgery. SUMMARY OF BACKGROUND DATAThe surgical correction of distal lumbar spine deformity involves significant risk for damage to neural structures that control muscles of ankle and toe dorsiflexion. Procedures often include vertebral translation, interbody fusion, and posterior-based osteotomies. The benefit of using MEP monitoring to predict dorsiflexion weakness has not been well-established. The purpose of this paper is to describe the relationship between neural complications from lumbar surgery and intraoperative MEP changes. METHODSIncluded were 542 neurologically intact patients who underwent posterior spinal fusion for the correction of distal lumbar deformity. Two myotomes, including tibialis anterior (TA) and extensor hallucis longus (EHL), were monitored. MEP and free-running electromyography data were assessed in each patient. Cases of new dorsiflexion weakness noted postoperatively were identified. Data in case and control patients were compared. There was no direct funding for this work. The Department of Anesthesiology and Perioperative Care provides salary support for authors one and six. Authors two and three report employment in the field of intraoperative neurophysiological monitoring as a study-specific conflict of interest. RESULTSTwenty-five patients (cases) developed dorsiflexion weakness. MEP amplitude decreased in the injured myotomes by an average of 65 ± 21% (TA) and 60±26% (EHL), which was significantly greater than the contralateral uninjured side or for control subjects. (p < .01) Receiver operator characteristic (ROC) curves showed high sensitivity, specificity, and predictive value for changes in MEP amplitude using either the TA or EHL. Analysis of MEP changes to either TA or EHL yielded a superior ROC curve. Net reclassification improvement analysis showed assessing MEP changes to both TA and EHL improved the predictability of injury. CONCLUSIONSThe use of MEP amplitude change is highly sensitive and specific to predict a new postoperative dorsiflexion injury. Monitoring two myotomes (both TA and EHL) is superior to relying on MEP changes from a single myotome. Electromyography activity was less accurate but compliments MEP use. Additional studies are needed to define optimal intraoperative MEP warning thresholds.