F175. Connectivity changes in selected EEG-networks during partial seizures with different types of alteration in consciousness: Ictal and interictal findings in 10 cases

Abstract

Introduction Consciousness is a central concept in epilepsy: a transient loss characterizes generalized seizures and different types of consciousness impairment may occur during partial ones. Jackson described the “dreamy state” and suggested a consciousness dissociation caused by electrical disturbances at higher levels in hierarchical neural networks. The following decades accumulated evidence of ictal and interictal experiential phenomena and several physiopathogenic hypotheses were raised. Yet, even after the EEG breakthrough, the mechanisms of impaired consciousness and its relation to partial seizures remains elusive. Focal discharges cannot explain the flexibility and variability of experiential phenomena. Current models of brain function imply multiple processes in hierarchical modular assemblies with dynamical non-linear interactions. From this perspective, a focal discharge can modify network connectivity (NC) resulting in more varied and longer lasting changes. OBJECTIVES: (1) to identify changes in NC accompanying ictal and peri-ictal discharges in 10 patients that presented seizures with different psychic contents and (2) to correlate NC changes with the type of consciousness impairment during the seizure. Methods Seizures with altered consciousness, no motor component, were documented in 10 psychiatric patients. The Ictal Consciousness Inventory scored level (ICI-L) and content (ICI-C) of consciousness; psychotic features obtained from the Positive and Negative Symptom Scale (PANSS-P). Two minutes of EEG seizure activity, artifact free, eyes-closed, were analyzed with Neuroguide software to generate Low Resolution Electromagnetic Topography (LORETA) scores for each of the Broadman Areas (BA). Pearson correlations of spectral power and phase differences for each of the BA pairs in selected networks: Anxiety (AN), Dorsal Attention (DAN), Ventral Attention (VAN), Default-Mode (DMN), Affective (AN), Executive (EN), Memory (MN) and Language (LN). Connectivity Z-scores were calculated with Neuroguide normative database. Multiple correlation analysis tested correlations between affected networks and clinical scales. Results Seizures defined by EEG ictal activity were accompanied by psychic changes in all subjects. Psychic changes and NC patterns differed among patients. The DMN was altered in all subjects, differing in direction and magnitude in correlation with ICI-l. AN and VAN correlated with the PANSS scales. Changes in other networks varied across subjects, uncorrelated with clinical scales. Conclusion Our findings suggest that: (1) EEG-networks can be used to study different types of ictal consciousness impairment, (2) level, content and reality testing may vary independently, (3) variations correlate with patterns of altered NC, (4) DMN structures modulate ICI-L, (5) AN and VAN are involved in reality testing.