Abstracts

Rationale: A classification of seizures with focal onset based on dynamics, in particular onset and offset patterns, has been proposed (Jirsa et al., 2014). Among the 16 possible types, the saddle node/homoclinic (SN/SH), which accompanies the ictal baseline jump, so called ictal Direct Current shift (ictal DC shift) (Ikeda et al., 1996, 1999), seems to be predominant across species (from humans to zebra fish). Based on a general theory of bursting dynamics (Saggio et al., 2017), we predicted that individuals with epilepsy may navigate the map of brain dynamics and express different types of seizures and that the variable types of seizure may affect the propagation pattern during the ictal period. To this end, we tested these hypotheses in an experimental model of temporal lobe epilepsy. Methods: We enrolled 10 male Sprague Dawley rats treated with pilocarpine (380 mg/kg, i.p.), which developed spontaneous epileptic seizures (Toyoda et al., 2015). Electroencephaogram (EEG) was obtained with a Microdrive (NLX 9-drive, Neuralynx) with bandpass filter (0.1-1800 Hz) and sampling rate 2 kHz. In each seizure, the seizure class was established based upon the identification of the type onset (SN, SNIC, SupH, or SubH) and offset (SH, SNIC SupH, or FLC). We evaluated i) the seizure duration, ii) the amplitude of the ictal DC shift that characterizes SN onset, iii) the number of brain regions involved during seizure propagation, iv) the number of brain regions characterized by high fast ripple (FR: 300-600 Hz) power, and v) the propagation time to the thalamus. Results: In line with our hypotheses, we confirmed the existence of several types of seizures in each individual animal. The duration of seizures with SN onset tended to be shorter than that starting with another type (in 5 out of 8 rats). For SN onset seizures, the abnormal activity easily propagated to surrounding area (in 4 out of 8 rats). However, there was no relationship between the type of seizure and the number of brain regions displaying high FR power (in 7 out of 8 rats). Furthermore, the number of involved electrodes correlate more to seizure duration than to the propagation time to the thalamus (p<0.01 in 7 out of 8 rats, p<0.03 in 1 out of 8 rats, multiple regression analysis). Finally, seizures with a large ictal DC shift at onset were characterized by a smaller number of electrodes with high FR power (in 5 out of 8 rats). In these rats model, all animals display several types of seizures, which may contribute to their drug-resistance. Conclusions: The most frequent form of seizures starting with ictal DC shift (SN onset) is characterized by faster propagation to surrounding regions possibly involving the thalamus. Funding: The research reported herein was supported by the funding from the European Union’s Horizon 2020 framework program for research and innovation under grant agreement No. 765549.