Retrospective Waveform Analysis of Transcranial Motor Evoked Potentials (MEP) to Identify Early Predictors of Impending Motor Deficits in Spinal Surgeries.

PURPOSE: Although there are guidelines analyzing transcranial motor evoked potentials (MEP) waveform criteria, they vary widely and are not applied universally during intraoperative neurophysiologic monitoring (IONM). The objective is to generate hypotheses to identify early and reliable MEP waveform characteristics prior to complete loss of MEP to predict impending motor spinal cord injuries during spinal surgeries. The ultimate goal is to enhance real-time feedback to prevent injury or detect reversible spinal cord damage.

METHODS: Fifteen true positive cases of persistent intraoperative MEP loss and new postoperative motor deficits were retrospectively identified from 2011 to 2013. Waveform characteristics of latency, amplitude, duration, phases, and area-under-the-curve (AUC) were measured, and an intraoperative spinal cord index (ISCI) was calculated for 5 traces prior to complete MEP loss. ISCI = [amplitude x duration x (phases+1) x AUC]/latency.

RESULTS: Out of 22 muscles in 15 cases, latency increased in 2, duration decreased in 12, amplitude decreased in 13, AUC decreased in 13, and ISCI decreased in 14. In 11 out of 15 cases (73%), ISCI dropped > 40% in at least one muscle before MEP were completely lost. Thirteen cases had concurrent somatosensory evoked potentials (SSEP) changes, 9 out of 13 had > 50% decrease in SSEP: 2 out of 9 changed before MEP, 5 out of 9 simultaneously, and 2 out of 9 after.

CONCLUSIONS: In these cases of motor injury, smaller and simpler MEP waveforms preceded complete loss of signal. An ISCI 40% drop could be tested as a warning threshold for impending motor compromise in future prospective studies and lead to eventual standardization to predict irreversible postoperative deficits.