ID 202 – Characterising seizures in anti-NMDA-receptor encephalitis with Dynamic Causal Modelling

Abstract

Objective To characterise the pathophysiology of seizures in terms of slow fluctuations in synaptic efficacy in patients with anti-N-methyl- d -aspartate Receptor (NMDA-R) encephalitis. Method Non-invasive clinical EEG recordings were obtained from two patients with anti-NMDA-R encephalitis with recurrent partial seizures (both females of age 19 and 33) using nine scalp electrodes positioned according to the 10–20 system. Electrographic seizure activity was localised using a Beamformer inversion. A virtual electrode at this position was used to estimate the underlying dynamics of the cortex using a Dynamic Causal Model (DCM). Results The DCM generated activity with a close fit to measured seizure activity which allowed us to confidently track changes in both excitatory and inhibitory connectivity. Seizure initiation was induced with a disruption in both excitatory and inhibitory connectivity resulting in excitatory–inhibitory imbalance, with an increasing weight on excitation as the seizure progressed. Conclusion Our findings suggest that seizures in anti-NMDA-R encephalitis are due to disruption in both inhibitory, GABA dependent, and excitatory, partly NMDA dependent, connectivity. Key message This study illustrates the inversion of dynamic causal models of seizure activity – recorded from the scalp in a clinical setting – to identify the underlying pathophysiological mechanisms.