Abstract, originally published in Epilepsia Open
Study objectives: Traumatic brain injury (TBI) results in sequelae that include post-traumatic epilepsy (PTE) and sleep-wake disturbances. Here we sought to determine whether sleep characteristics could predict development of PTE in a model of severe TBI.
Methods: Following controlled cortical impact (CCI) or sham injury (craniotomy only) CD-1 mice were implanted with epidural electroencephalography (EEG) and nuchal electromyography (EMG) electrodes. Acute (1st week) and chronic (months 1, 2 or 3) 1-week long video EEG recordings were performed after the injury to examine epileptiform activity. High amplitude interictal events were extracted from EEG using an automated method. After scoring sleep-wake patterns, sleep spindles and EEG delta power were derived from non-rapid eye movement (NREM) sleep epochs. Brain CTs (computerized tomography) were performed in sham and CCI cohorts to quantify the brain lesions. We then employed a no craniotomy (NC) control to perform 1-week long EEG recordings at week 1 and month 1 after surgery.
Results: Posttraumatic seizures were seen in CCI group only, whereas interictal epileptiform activity was seen in CCI or sham. Sleep-wake disruptions consisted of shorter wake or NREM bout lengths and shorter duration or lower power for spindles in CCI and sham. NREM EEG delta power increased in CCI and sham groups compared to NC though CCI group with post-traumatic seizures had lower power at a chronic time point compared to those without. Follow up brain CTs showed a small lesion in the sham injury group suggesting a milder form of TBI that may account for their interictal activity and sleep changes.
Significance: In our TBI model, tracking changes in NREM delta power distinguishes CCI acutely and animals that will eventually develop PTE, but further work is necessary to identify sleep biomarkers of PTE. Employing NC controls together with sham controls should be considered in future TBI studies.