Abstracts

Rationale: Recent studies have shown an association of 100-500 Hz high frequency oscillations (HFOs) in the local field potential (LFP) with epileptogenic regions in humans and animal models of temporal lobe epilepsy (TLE). Recordings from humans and rodents have also shown interictal EEG-band activity (IEA), such as interictal spikes and epileptiform discharges that is often localized to epileptic brain. What is unknown from these studies is the spatial relationship of HFOs to IEA across the hippocampus in TLE. Thus, we developed a novel technique to probe the 2D distribution of IEA and HFOs in the hippocampus of chronically epileptic rats using a multielectrode array in vivo. Methods: Young adult (200-275g) male Spague-Dawley rats were pretreated with atropine methylbromide (2.5mg/kg s.c.) then injected with 380mg/kg pilocarpine (380mg/kg s.c.;pilo) to induce status epilepticus (pilo-SE). After 1hr of SE, rats were injected with diazepam (10mg/kg i.p.). Four to eight weeks after SE, rats were anesthetized with urethane (1.5g/kg) ± ketamine (25mg/kg) and placed in a stereotaxic frame. After craniotomy, the cortex overlying the right CA1 and subiculum was gently aspirated away and a 96-electrode silicon array (10 rows and 10 columns of electrodes, 400μm inter-electrode spacing, 0.5-0.75 mm insertion depth; Cyberkinetics, Inc.) was placed in the pyramidal cell layer/stratum radiatum using a pneumatic inserter. Recordings (30kS/s, 1-5kHz filtering) began 1.5 to 3 hours later, after evoked responses to contralateral CA3 stimulation had stabilized. HFOs were detected using an automated algorithm (Gardner et al. 2007, Clin Neurophysiol 18:1134-43) and the mean HFO rate was determined for each electrode in the recording array. To estimate IEA, mean line-length of LFP activity between 1-32 Hz was calculated for each electrode. Both HFO rate and IEA were averaged over 15 minutes of baseline recording and the spatial relationship between the two measures was determined by linear correlation. Electrode positions were verified histologically after recording. Results: Six pilo-SE rats underwent successful implantation. All pilo-SE rats had at least one observed spontaneous seizure. Three pilo-treated rats that did not have SE as well as 4 untreated, age-matched rats served as controls. We found the mean HFO rate was significantly higher in pilo-SE rats (13.7±1.6 vs 8.8±0.9 min-1; p=0.023, two-tailed t-test). In pilo-SE animals, electrodes which showed the highest rate of HFOs were often distinct from those that had the highest IEA. The spatial distribution of HFO rate was more negatively correlated with the distribution of EEG-band activity in pilo-SE compared to controls (z=-0.21±0.15 vs 0.13±0.08;p=0.038). Conclusions: In this study, we found that the greatest amount of HFOs in epileptic hippocampus occur at sites separate from but near areas of maximal IEA. Our findings suggest that HFOs, by surrounding areas of maximal interical discharges, may reflect neuronal activity that serves to control focal epileptiform activity in cortical circuits.