Intraoperative Somatosensory and Motor Evoked Potentials in Predicting Postoperative Neurological Deficits in Neuromuscular Scoliosis: A Study Level Meta-analysis
Priti Ezhuvathra1, Abdullah Al Qudah2, Vamsi Krishna Gorijala3, Sulekhya Inturu4, Katherine Anetakis1, Jeffery Balzer1, Parthasarathy Thirumala1
1Department of Neurological Surgery, 2UPMC Stroke Institute, University of Pittsburgh Medical Center, 3University of Tennessee Health Science Center, 4Mamata Medical College
Objective:
This study aimed to evaluate the role of intraoperative neurophysiological monitoring (IONM), utilizing somatosensory evoked potentials (SSEP) and motor evoked potentials (MEP), as a real-time diagnostic tool to predict iatrogenic neurological injury and prevent postoperative neurological deficits (PPND) during spine correction surgery in patients with neuromuscular scoliosis (NMS).
Background:
Approximately 3% of NMS cases experience neurological injury during spinal correction surgery, significantly impacting outcomes. IONM, utilizing SSEP and MEP, is a common practice during surgical repair aiming to detect and reduce the incidence of iatrogenic injury.
Design/Methods:
A systematic review included studies from 1995 to 2023 incorporating IONM utilizing both SSEP and MEP during NMS surgeries. The primary outcome was PPND. The accuracy of IONM as a real-time diagnostic tool was assessed through sensitivity, specificity, and pooled diagnostic odds ratios (DOR) with 95% Confidence Interval (CI).
Results:
Eleven studies involving 514 patients undergoing NMS correction identified 20 patients with PPND, resulting in a prevalence of 3.89%. The DOR for IONM predicting PPND was 27.3 (95% CI: 7.5 – 99.2, p <0.001). IONM demonstrated a specificity of 93% (95% CI: 87% – 96%). However, sensitivity was 67% (95% CI: 44% – 84%), attributed to many false positives (changes in IONM without PPND). The analysis showed an area under the ROC curve (AUC) of 0.85, a post-test probability of 81.82%, and a positive likelihood ratio of 4.5.
Conclusions:
Given IONM's high specificity, it could serve as a real-time diagnostic tool for predicting PPND in NMS. Although its sensitivity was lower compared to specificity, this can be explained by the nature of IONM as a real-time monitoring tool, providing timely communication of changes to the surgical team, allowing for adjustments. This increases the false positive rateand thereby reduces sensitivity, but ultimately contributes to a decreased incidence of neurological injury.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.