Abstracts

Deep brain stimulation (DBS) is a critical therapeutic regimen for drug-resistant epilepsy. Currently, DBS is the focal point of international clinical trials for the treatment of intractable epilepsy. A key target of epilepsy DBS in both this rodent model study and in the concurrent clinical trials is the anterior thalamic nuclei (AN). The work from our group isolated AN as a key thalamic site mediating the expression of experimental seizures incited by pentylenetetrazol (PTZ). Our proposed method of assessing the neuromodulatory effects of DBS on seizure threshold and post-ictal activity is based on the principle that seizure build-up is always preceded by constantly changing EEG and field potential bursting levels.
Our results were obtained using a slow infusion (5.5 mg/kg/min) anesthetized PTZ rat model in which we recorded EEG from four cortical and three thalamic (right and left anterior and right posterior thalamic nuclei) areas. After electrode implantation and recovery, EEG recording began. During the 40 min baseline period prior to PTZ infusion, bilateral AN DBS (150 [mu]A, 100 Hz, 0.1 ms pulse duration) was applied to the DBS rats (N=5), but not to the control rats (N=4). Following baseline, PTZ was administered and infusion was stopped after onset of the second seizure. We used a novel measure, residual subband wavelet entropy (RSWE), to directly estimate the entropy of bursts, which is otherwise obscured by the ongoing EEG background activity. RSWE allows us to capture changes in activity level regardless of morphological characteristics.
Mean time to first seizure was 3412.75 [plusmn] (977 SD) s for the control group and 5057.8 [plusmn] (449 SD) s for the DBS group. A paired-sample T-test showed a significant delay (p[le]0.01) in seizure onset, and therefore an increase in seizure threshold, due to DBS. Using univariate and multivariate repeated measures ANOVA, we observed a significant increase in pre-ictal cortical bursting due to DBS, as represented by increased RSWE, in the beta (p[le]0.03), alpha (p[le]0.04), theta (p[le]0.01), and delta (p[le]0.006) frequency bands. In addition, we found no significant effects of DBS during the 0-20 min period immediately following the first seizure event, but did find heightened bursting and excitability (p[le]0.02-p[le]0.07) due to DBS in five of the seven channels during the 20-40 min post-ictal period. The fact that no statistically significant DBS effects were detected post-ictally in RAN leads us to believe that RAN shows little change of activity, i.e. that there was little lesioning effect due to DBS. This apparent lack of injury supports the claim that DBS is reversible.
Increases in EEG volatility as measured by RSWE point to increased neural excitation. Our results confirm the desynchronization or [quot]jamming[quot] mechanism through which DBS acts as a source of neural excitation on the oscillatory activity underlying this disease state.
[Supported by: NIH(NS35528) to D.L.S.]