"Threshold-level" multipulse transcranial electrical stimulation of motor cortex for intraoperative monitoring of spinal motor tracts: description of method and comparison to somatosensory evoked potential monitoring

Numerous methods have been pursued to evaluate function in central motor pathways during surgery in the anesthetized patient. At this time, no standard has emerged, possibly because each of the methods described to date requires some degree of compromise and/or lacks sensitivity. OBJECT: The goal of this study was to develop and evaluate a protocol for intraoperative monitoring of spinal motor conduction that: 1) is safe; 2) is sensitive and specific to motor pathways; 3) provides immediate feedback; 4) is compatible with anesthesia requirements; 5) allows monitoring of spontaneous and/or nerve root stimulus-evoked electromyography; 6) requires little or no involvement of the surgical team; and 7) requires limited equipment beyond that routinely used for somatosensory evoked potential (SSEP) monitoring. Using a multipulse electrical stimulator designed for transcranial applications, the authors have developed a protocol that they term "threshold-level" multipulse transcranial electrical stimulation (TES). METHODS: Patients considered at high risk for postoperative deficit were studied. After anesthesia had been induced and the patient positioned, but prior to incision, "baseline" measures of SSEPs were obtained as well as the minimum (that is, threshold-level) TES voltage needed to evoke a motor response from each of the muscles being monitored. A brief, high-frequency pulse train (three pulses; 2-msec interpulse interval) was used for TES in all cases. Data (latency and amplitude for SSEP; threshold voltage for TES) were collected at different times throughout the surgical procedure. Postoperative neurological status, as judged by evaluation of sensory and motor status, was compared with intraoperative SSEP and TES findings for determination of the sensitivity and specificity of each electrophysiological monitoring technique. Of the 34 patients enrolled, 32 demonstrated TES-evoked responses in muscles innervated at levels caudal to the lesion when examined after anesthesia induction and positioning but prior to incision (that is, baseline). In contrast, baseline SSEPs could be resolved in only 25 of the 34 patients. During surgery, significant changes in SSEP waveforms were noted in 12 of these 25 patients, and 10 patients demonstrated changes in TES thresholds. Fifteen patients experienced varying degrees and durations of postoperative neurological deficit. Intraoperative changes in TES thresholds accurately predicted each instance of postoperative motor weakness without error, but failed to predict four instances of postoperative sensory deficit. Intraoperative SSEP monitoring was not 100% accurate in predicting postoperative sensory status and failed to predict five instances of postoperative motor deficit. As a result of intraoperative TES findings, the surgical plan was altered or otherwise influenced in six patients (roughly 15% of the sample population), possibly limiting the extent of postoperative motor deficit experienced by these patients. CONCLUSIONS: This novel method for intraoperative monitoring of spinal motor conduction appears to meet all of the goals outlined above. Although the risk of postoperative motor deficit is relatively low for the majority of spine surgeries (for example, a simple disc), high-risk procedures, such as tumor resection, correction of vascular abnormalities, and correction of major deformities, should benefit from the virtually immediate and accurate knowledge of spinal motor conduction provided by this new monitoring approach.