Abstracts

Rationale: Hypoxic-ischemic (HI) brain injury during the neonatal period is a common cause of poor neurological outcomes such as cerebral palsy, intellectual disability and epilepsy. Clinical data suggest that the best predictor of negative neurologic outcomes after an injury is abnormal background EEG activity (Tharp et al., Electroenceph. Clin. Neurophysiol,1981,51:219; Monod et al. Electroenceph. Clin Neurophysiol,1972,32:529). In animal models, however, acute behavioral seizures are often used as an outcome measure, rather than background EEG. The usefulness of epileptiform events as a predictor of negative neurological outcome has been questioned in the clinical literature (Holmes and Lombroso, J. Clin. Neurophysiol,1993,10:323). We compared background EEG patterns from rat pups using three treatment groups hypoxia-ischemia (acute seizures with brain damage), hypoxia (acute seizures with no brain damage), and untreated controls (no seizures, no brain damage). We hypothesized that the animals with brain damage would have the most severe abnormalities in the background EEG after treatment, regardless of any epileptiform events. Methods: A miniature EEG telemetry device was implanted at postnatal day 6. Following 24 h recovery, animals were treated with HI (n=9), hypoxia alone (n=11), or normal air (control, n=12). In the HI group, the left common carotid artery was cauterized and animals were exposed to 8% oxygen for 2 h. EEG was recorded and analyzed using fast Fourier transforms. Results: At 15 min after HI treatment, EEG recordings showed extensive signal suppression (9/9) in all frequency bands. Acute seizures (>1 min) were present in 2/9 animals at 15 min. Neither hypoxia-treated animals nor controls had background suppression or acute seizures. At 6 h after treatment, HI animals exhibited gradual recovery of activity in delta, theta and alpha frequency bands, but not in higher frequencies (beta, gamma). This pattern persisted in recordings at 24 h, 48 h, 72 h and 96 h after treatment. Only one HI-treated animal (1/9) had seizure-like spike-wave discharges 96 h after administration of HI for prolonged periods of time, but all HI animals had suppressed beta and gamma activity. Animals were euthanized at 96 h to verify the injury. All HI-treated animals had evidence of brain damage either as overt macroscopic brain lesions (8/9) or as less severe, diffuse injury with neuronal degeneration in cortex, hippocampus and thalamus. All animals with an injury, but only those with an injury had suppressed background EEG. Conclusions: Suppressed beta and gamma patterns in background EEG may be a biomarker for neuronal death that can be detected as early as 15 min and as late as 96 h after treatment. Recent evidence (Kadam et al., 2010 J Neurosci 30:404) suggests that HI-treated neonatal rats with brain damage go on to develop epilepsy. These results suggest that suppressed beta and gamma activity in background EEG may be an early biomarker for early epileptogenesis.