Back

Slow Oscillations Gate Interictal Spikes Across the Human Thalamocortical-Epileptogenic Network

Alipour, M.; Drongelen, W. v.; Nordli, D. R.; Voss, J.; Lee, K.; Satzer, D.

2026-06-03 neuroscience
10.64898/2026.05.30.728961 bioRxiv
Show abstract

BackgroundSlow oscillations (SOs; 0.5-1.5 Hz), a hallmark of non-rapid eye movement (NREM) sleep, are associated with a marked amplification of interictal epileptiform spike (IIS) activity in focal epilepsy. However, the network-level organization of this effect across the thalamocortical-epileptogenic system, and whether IIS-permissive SOs can be predicted from pre-onset brain states, remain unclear. MethodsWe analyzed simultaneous scalp EEG and stereo-EEG (SEEG) recordings from 6 patients with drug-resistant focal epilepsy across 24 full-day recording days, sampling prefrontal cortex (PFC), thalamus, and seizure onset zone (SOZ). SO-IIS coupling was characterized across vigilance states using peri-event and phase-based analyses, with a gamma-based validation step to reduce contamination by IIS-related slow potentials. Pre-onset phase-amplitude coupling (PAC) was compared between IIS-permissive and non-permissive SOs. ResultsSO-IIS coupling was observed across all regions, with the strongest and most temporally precise pre-trough IIS clustering in the SOZ (peak 4.4% in NREM), exceeding PFC (1.7%) and thalamic coupling. Thalamic coupling was preserved across wakefulness and NREM and was significant in 7/11 nuclei, with nucleus-specific phase preferences. SO morphological features, particularly up-slope and peak-to-peak amplitude at PFC contacts, predicted IIS occurrence in the SOZ. Pre-onset PAC differed significantly between permissive and non-permissive SOs across regions. ConclusionsSO-IIS coupling is a distributed, network-level phenomenon with region- and state-specific characteristics, and pre-onset PAC provides a predictive signature of IIS-permissive brain states. These findings support the feasibility of developing personalized, closed-loop neuromodulatory strategies targeting SO-gated IIS suppression in focal epilepsy.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.