Neurology · 3 h ago
Slow-wave sleep highlights epileptogenic-zone network features in focal epilepsy
An intracranial EEG study of 22 patients with drug-resistant focal epilepsy found that NREM stage 3 selectively increased gamma-band synchrony within the presumed epileptogenic zone. A classifier using sleep-derived network features localized that zone with a median AUC of 0.859, but requires validation in larger cohorts.
- NREM3 increased pEZ gamma synchrony versus non-involved tissue.
- Sleep-related network features classified pEZ with median AUC 0.859.
- Prospective validation is needed before clinical use.
Researchers analyzed resting-state intracranial EEG from 22 patients with drug-resistant focal epilepsy undergoing presurgical evaluation. They compared wakefulness, NREM stage 2, NREM stage 3, and REM sleep, examining spectral power, connectivity, graph measures, and the aperiodic 1/f exponent across presumed epileptogenic (pEZ), irritative, and non-involved zones. The analyzed epochs were free of interictal epileptiform discharges.
During NREM stage 3, gamma-band synchrony was higher within the pEZ than in non-involved tissue (effect size d=1.204; P=0.027). Local efficiency and betweenness centrality in the pEZ also increased in beta1 and gamma bands compared with wakefulness. A three-feature classifier combining NREM3 aperiodic exponent, beta2 connectivity, and gamma local efficiency achieved a median AUC of 0.859 for distinguishing pEZ from non-pEZ electrodes. The authors suggest these sleep-related features could complement presurgical biomarkers, but the small cohort and need for prospective validation mean it is not yet established that they improve localization, surgical planning, or outcomes.
In practice
In patients with drug-resistant focal epilepsy undergoing presurgical evaluation, NREM3-derived EEG network measures may offer additional features for identifying the presumed epileptogenic zone, including when ictal recordings are not used. The evidence comes from 22 patients and a classifier assessed with leave-one-subject-out cross-validation; the reported median AUC was 0.859. These findings are exploratory and do not establish improved surgical planning or outcomes, so they do not support changing current localization practice without prospective validation.
AI-generated clinical interpretation · Not editor-reviewed
Is this summary clinically accurate?
Help fellow clinicians: your rating sends inaccurate summaries straight to our editors.
Sign in to rate this summary →Source
SLEEP: Slow-wave sleep selectively amplifies network signatures of the presumed epileptogenic zone in focal epilepsy ↗This is an automated AI-condensed summary that has not yet been reviewed by an editor. Always consult the full item at the original source.