Abstracts

Rationale: Hypoxic-ischemic (HI) brain injury is the most common cause of acute seizures in children during the neonatal period. The susceptibility to stroke is highest during the earliest and latest stages of life, making the perinatal period second to late adulthood in vulnerability to this condition. EEG recordings from human neonates that have suffered an HI insult have shown that seizure activity was present for 48 to 72 h after birth in most cases that subsequently developed epilepsy (Toet et al., 2005, Pediatr. Neurol. 32:241). Neonatal stroke following HI can be modeled by exposing rat pups at postnatal day 7 to unilateral carotid artery ligation followed by 2 h of hypoxia with 8% oxygen (Levine, 1960, Arch. Pathol. 69:544; Rice el al., 1981, Ann. Neurol. 9:131; Kadam and Dudek, 2007, J. Comp Neurol. 505:716). In this study, we aimed to analyze the properties of acute seizures that occur during and after the HI event in this animal model of perinatal stroke using a custom-built, miniature telemetry device as well as traditional wired recording of EEG. Methods: EEG electrodes on the cortical surface, connected to either a miniature telemetry device or a tethered system, were implanted in rat pups at postnatal day 7 or 8. Animals were allowed to recover for 6-24 h, and were then administered the HI treatment. To induce HI, the left common carotid artery was cut with a cauterizer. The pups were allowed to recover for 2 h with the dam, and were treated for 2 h with 8% oxygen and 92% nitrogen (i.e., hypoxia) at 37°C. EEG was monitored before, during and after the treatment. The animals were then allowed to recover, and monitored for 2 h during the next 2 days. Results: After carotid occlusion but before introduction of the hypoxic-gas mixture, no abnormal EEG activity was observed. During the hypoxic period, seizures were recorded from all ischemic animals with satisfactory EEG signals (n=10), both ipsilateral (n=8) and contralateral (n=2) to the injured hemisphere. Average duration of seizures during hypoxic treatment was 51 +/- 36 sec (n=71), and time between seizures ranged from 7 sec to 8 min. Periods with low-frequency spiking between the seizures could be observed in some animals. After re-introduction of normal air into the chamber, 2 animals continued to exhibit abnormal activity on the EEG. Additionally, spontaneous seizures were recorded 24 h after HI (7.3 min, n=1) and 48 h after HI (6 min, n=1). Data recorded using the miniature telemetry device had higher signal-to-noise ratio and less movement artifact. Conclusions: These preliminary data suggest that this rat model of neonatal HI is a viable platform for studying the acute seizures associated with perinatal stroke. This approach allows a quantitative analysis of the severity of the electrographic seizures and possible therapeutic strategies. Additionally, the miniature telemetry device is able to record surface EEG activity with better signal-to-noise ratio, and it allows longer monitoring of the animals than a conventional tethered approach. This project was supported by the American Epilepsy Society and NS045144.