Comparative Analysis of Fusion and Decompression Surgery in Degenerative Cervical Myelopathy: A Multicenter Observational Study of Transcranial Motor Evoked Potentials (Tc-MEP).

Abstract

A multicenter observational study. To evaluate and compare the diagnostic performance and clinical value of transcranial motor-evoked potentials (Tc-MEPs) in predicting postoperative paralysis in degenerative cervical myelopathy (DCM), specifically on stratifying true-positive alerts in fusion versus decompression surgeries. Tc-MEP monitoring is widely used to detect intraoperative motor deficits, but its accuracy and predictive value in different surgical approaches remain unclear. A total of 3,813 patients with DCM who underwent cervical spine surgery were analyzed. Tc-MEP alerts were defined as a ≥70% reduction in amplitude. Postoperative paralysis was categorized as upper limb palsy or lower limb palsy. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of Tc-MEP alerts were calculated. Multivariate logistic regression analysis identified independent predictors of true-positive alerts. Postoperative motor paralysis occurred in 1.36% (52/3,778) of cases. The Tc-MEP sensitivity was 57.7%, specificity 93.2%, PPV 10.6%, and NPV 99.4%. Fusion surgery had a significantly higher postoperative paralysis rate (3.7% in posterior decompression with fusion [PDF] vs. 0.8% in posterior decompression alone). Fusion surgery was an independent predictor of true-positive alerts (OR = 4.62, 95% CI: 1.69-12.66), while non-attributed alerts were the primary cause of false positives (OR = 0.036, 95% CI: 0.004-0.37). Tc-MEP alerts showed higher PPV in fusion surgery, reflecting the greater risk profile in these procedures. However, the consistently high NPV across all surgical approaches supports the broad utility of Tc-MEP in ensuring intraoperative neurological safety. This study provides the first large-scale, stratified analysis of Tc-MEP alert outcomes in fusion versus decompression surgery, offering actionable insights for intraoperative neuromonitoring in complex DCM cases.