Direct brainstem stimulation for intraoperative somatosensory evoked potential monitoring: A prospective study
Related Experiment Video
Updated: Nov 2, 2026
Intra-Operative Behavioral Tasks in Awake Humans Undergoing Deep Brain Stimulation Surgery
Published on: January 6, 2011
Kristina Lapteva1, Andrey Gavrjushin2, Aleksei Veselkov3
1Laboratory of Clinical Neurophysiology, Burdenko Neurosurgery Center, 4th Tverskaya-Yamskaya Street 16, Moscow 125047, Russia.
Objective:
Intraoperative somatosensory evoked potential (SSEP) monitoring during brainstem and cervicomedullary surgery is limited by poor reproducibility and delayed feedback. We evaluated brainstem-stimulus-induced SSEPs (bsiSSEPs) and compared their diagnostic performance with conventional peripheral SSEPs (pSSEPs).
Methods:
Twenty-three patients undergoing brainstem or cervicomedullary lesion resection were enrolled in this prospective single-center study. Both pSSEPs and bsiSSEPs were recorded intraoperatively. The primary outcome was postoperative worsening of deep sensation. Diagnostic performance was assessed using ROC analysis, McNemar's exact test, Cohen's κ, and DeLong's test.
Results:
Conventional pSSEPs were recordable in 18/23 patients (78%), whereas bsiSSEPs were obtained in all patients with 15-30 averages. Using the exploratory ≥33% amplitude-reduction threshold, estimated sensitivity was 100% (95% CI 47.8-100.0%), specificity 94.4% (95% CI 72.7-99.9%), and AUC 0.978 (95% CI 0.927-1.000). In the paired subgroup, sensitivity and specificity were 100% and 92.9% for bsiSSEPs versus 75.0% and 50.0% for pSSEPs. Exploratory ROC analysis yielded AUCs of 0.964 and 0.625, respectively; however, predictor definitions differed.
Conclusion:
In this exploratory cohort, bsiSSEPs were feasible and consistently recordable, including in patients without recordable conventional pSSEPs. Although diagnostic performance estimates were favorable, differences in predictor definition and in-sample derivation of the bsiSSEP threshold preclude definitive conclusions regarding superiority over conventional pSSEPs. Further prospective validation using harmonized comparison criteria is required.
Significance:
Direct brainstem or upper cervical spinal cord stimulation addresses a major limitation of conventional sensory monitoring by providing rapid scalp-recorded evoked responses without relying on peripheral nerve or the sublesional portion of the spinal somatosensory pathway.
