Abstracts

Rationale: Traumatic brain injury (TBI) is a heterogeneous injury, often involving both focal and diffuse components. The high incidence of post-traumatic epilepsy (PTE) is well-established. However, the mechanisms underlying epileptogenesis are unknown, especially the contributions of the different components of TBI. In order to address these problems, a large animal model of PTE that accurately compares the diffuse and focal components of injury is required, as is a methodology for video-EEG in these animals. Methods: Male Yucatan swine (age 5 months) underwent coronal rotational acceleration, a model of diffuse brain injury that induces little or no loss of consciousness and minimal subdural hemorrhage but significant axonal pathology.  Acute in vivo electrophysiology performed 7 days post-injury allowed for examination of the changes in hippocampal function in sham (n=3) versus injured animals (n=6). Baseline oscillatory activity in the hippocampus was recorded using high-density (32-channel) laminar depth probes. Hippocampal activity was also analyzed in response to stimulation of the Schaffer collaterals and entorhinal cortex.  A well-characterized controlled cortical impact (CCI) methodology was applied to another group of animals (n=3) in order to examine the neuropathological response to this contusion injury. In addition, a chronic implantation methodology was developed for this cohort using both a 32-channel depth electrode in the hippocampus and a 24 channel subdural grid adjacent to the contusion. A custom enclosure system consisting of a 64-channel headstage, batteries, and a wireless transmitter was developed to enable wireless 30kHz data acquisition. This system was integrated with an infrared camera to allow for continuous video-EEG monitoring of implanted animals. Results: In the acute rotational animals, dorsal hippocampal traces recorded in response to pulse stimulations had significantly altered waveforms in CA1, and theta burst stimulation provoked epileptiform activity. These findings were not observed in sham animals, suggesting post­synaptic hyperexcitability or a shift in the excitation-inhibition balance of local circuitry after injury.  At 48 hours post-CCI with an intact dura (n=1), there was very minimal hemorrhage at the cortical surface and an associated foci of pyknotic neurons. Minimal APP positive axonal pathology was observed in the white matter immediately below the impact site. In contrast, at the same time point following CCI, performed through an open dura and using the same impact parameters (n=2), there was a more extensive focal hemorrhagic contusion extending through the full cortical thickness and into the digitate white matter. Extensive neuronal pyknosis and axonal pathology was observed throughout both the affected and adjacent gyri. In chronically implanted animals, video-EEG was successfully performed for both the depth and subdural arrays using the wireless enclosure, with one animal implanted thus far for 3 months with no complications. Conclusions: The rotational injury data suggest that diffuse brain injury may induce hippocampal axonal and synaptic dysfunction and changes in hippocampal cellular excitability.  The CCI model leads to substantial cortical and axonal pathology that has the potential for inducing seizure activity. Comparison and superimposition of these injuries combined with long-term video-EEG in this large animal model could clarify the contribution of each injury type to post-traumatic epileptogenesis. Funding: DoD CDMRP W81XWH-16-1-0675