Okresowe uogólnione wyładowania niezapadowe podczas snu głęboskiego przed napadami nocnymi w epilepsji skroniowej i pozaskroniowej: badanie eksploracyjne z analizą ilościową EEG
Periodic generalized nonepileptiform discharges (PGNDs) during NREM sleep before nocturnal seizures in temporal and extratemporal lobe epilepsy: An exploratory quantitative EEG study
W skrócie
Badacze analizowali zapisy EEG pacjentów z epilepsją lekooporną, u których występowały napady nocne. Okazało się, że pewne characteristic wzory aktywności mózgu pojawiające się tuż przed napadem różnią się między dwoma typami epilepsji: skroniową i pozaskroniową, szczególnie w zakresie wolnych fal mózgowych. Jednak różnice te są nieznaczne i wymagają potwierdzenia w kolejnych badaniach, zanim będą mogły być stosowane w praktyce klinicznej.
Oryginalny abstract (angielski)
STUDY OBJECTIVES: To explore whether the spectral power of periodic generalized nonepileptiform discharges (PGNDs), a newly operationalized subset of CAP A-like NREM slow-wave events, differs between temporal lobe epilepsy (TLE) and extratemporal lobe epilepsy (ETLE), and to identify demographic, clinical, and etiological factors associated with delta power. METHODS: We retrospectively analyzed drug-resistant focal epilepsy patients undergoing long-term video-EEG monitoring (2018-2024) with at least two PGNDs in the preictal period of nocturnal seizures. PGNDs were visually scored and subtyped, and matched interictal NREM controls were selected. Absolute and relative (amplitude-normalized) spectral power was calculated for five frequency bands. Analyses were performed at the patient level, with false discovery rate (FDR) correction, cross-validated ROC analysis, confounder-adjusted models, and a mixed-effects model for determinants of delta power. RESULTS: Sixty-eight patients (35 TLE, 33 ETLE; 124 seizures, 711 PGNDs) met inclusion criteria. Delta (0.5-2.4 Hz) power of the PGND just before seizure onset (PGND1) was higher in ETLE than TLE, both in absolute terms (p = 0.016) and relative to total power (51.6% vs. 45.1%, p = 0.006, FDR q = 0.035), and the difference persisted after adjustment for carbamazepine use. Discrimination was modest (AUC=0.669; leave-one-out sensitivity 61%, specificity 74%). Absolute electrode-level differences did not survive FDR correction, whereas relative PGND1 delta power remained higher in ETLE at 16 of 20 electrodes after correction, indicating a widespread rather than regional difference. No within-patient preictal increase in PGND power was observed. Only febrile seizure history was independently associated with delta power. CONCLUSIONS: PGND delta power differs between ETLE and TLE independently of overall EEG amplitude, suggesting localization-related differences in NREM slow-wave events rather than a preictal biomarker. Given the modest discrimination and exploratory design, validation in independent cohorts is required before any clinical application.