Submitted:
12 October 2026
Posted:
14 October 2026
You are already at the latest version
Abstract
Background/Objectives: Acute EEG responses to spinal cord stimulation (SCS) may reflect network responsiveness in disorders of consciousness (DoC). Their association with subsequent recovery remains uncertain. Methods: Thirty patients underwent EEG assessment before SCS (W1) and 30 min after stimulation cessation (W5). Relative connectivity changes across five frequency bands and 10 regional pairs were compared between 15 improved and 15 non-improved patients at 3 months. Analyses included false discovery rate (FDR) correction, bootstrap resampling, leave-one-patient-out stability assessment, duration-adjusted regression, and exploratory nested cross-validation of machine-learning models. Results: Theta central-occipital connectivity (T-CO) showed the strongest unadjusted difference (rank-biserial effect size −0.644; P = 0.0028; bootstrap 95% CI −0.920 to −0.298), with a consistent direction across all leave-one-patient-out analyses. No EEG feature survived FDR correction. Repeated nested 5-fold validation showed variable discrimination; the highest mean combined-model AUC was 0.618. Descriptive XGBoost SHAP analysis ranked T-CO and theta parietal-occipital connectivity highest. Conclusions: Acute connectivity responses contain candidate signals associated with subsequent clinical improvement. These exploratory findings support further investigation of theta posterior-network responses, but do not establish a reliable individual prognostic classifier.
Keywords:
disorders of consciousness
; spinal cord stimulation
; electroencephalography
; functional connectivity
; neuromodulation
; biomarker
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.