Authors :
Presenting Author: Vaclav Kremen, PhD – Mayo Clinic
Vladimir Sladky, MSc – Mayo Clinic
Filip Mivalt, PhD – Mayo Clinic Rochester
Nicholas Gregg, MD – Mayo clinic, Rochester, Minnesota
Brian Lundstrom, MD PhD – Mayo clinic, Rochester, Minnesota
Benjamin Brinkmann, PhD – Mayo Clinic
Erik St Louis, MD – Mayo Clinic
Jamie J Van Gompel, MD – Mayo Clinic, Rochester MN, USA.
Kai Miller, MD, PhD – Mayo Clinic
Gregory Worrell, MD,PhD – Mayo Clinics
Rationale:
Epilepsy is characterized by sudden, debilitating seizures arising from abrupt state transitions in brain dynamics. Understanding and predicting these critical transitions remains a significant clinical challenge. While passive electrophysiological monitoring has identified some pre-ictal indicators, actively probing brain network states may offer a more robust and sensitive approach to assess underlying neural excitability and seizure susceptibility.
Methods:
We investigated the effects of electrical stimulation of the anterior nucleus of the thalamus (ANT) on hippocampal excitability. We delivered therapeutic electrical pulses (4-6 mA, 100 µs pulse width, 2 Hz) to the ANT. This stimulation was designed to transiently perturb neural activity without inducing seizures. Data were acquired using a chronically implanted neurotechnology platform that allowed for continuous seizure tracking and intracranial EEG (iEEG) recording enabling long-term, seamless data streaming to mobile devices and a cloud-based environment. We evaluated hippocampal evoked potentials in response to the ANT stimulation pulses. The magnitude and recovery dynamics of these responses were quantified using several features. To assess for differences in brain states, we employed generalized linear mixed-effects models.
Results:
We analyzed months (6 ± 3 months) of continuous iEEG from five human participants with temporal lobe epilepsy in their naturalistic, ambulatory environments. Seizure events and interictal spikes were quantified. Hippocampal evoked responses to ANT stimulation were reconstructed across diverse behavioral states, including sleep, wakefulness, and interictal and pre-ictal periods. Seizures predominantly originated during wakefulness. Critically, we identified significant alterations in hippocampal response characteristics across these behavioral (RMS 30.1+/-4.6uV in awake, 38.1+/-3.6uV in NREM sleep, Latency of N1 peak 43.8+/-2.1 msec in awake, 47.1+/-2.3 msec in NREM sleep) and pre-ictal states. Analysis with a linear mixed-effects model revealed a significant (p< 0.001) increase in Peak-to-Peak amplitude of 9.00 µV peaking at the 15-minute pre-ictal window compared to the inter-ictal baseline. These changes in evoked response dynamics served as early warning signals of impending critical transitions to seizure.