Abstracts

Progressive preictal dynamical entrainment and postictal disentrainment (resetting) has been reported in human epilepsy ([italic]IEEE TBME[/italic] 2003;50;616-27). There are also reports of anticonvulsant effects of high frequency stimulation of the hippocampus in humans. We postulate that this anticonvulsant effect is due to dynamical resetting by the stimulation. We used a spontaneously seizing animal model of chronic limbic epilepsy to test the hypothesis of dynamical resetting by electrical stimulation of the hippocampus. Adult male Sprague Dawley rats were used for the preliminary set of experiments. The animal model was developed using a modified chronic hippocampal stimulation protocol first described by Lothman and Bertram ([italic]Epi. Res.[/italic] 1990). The animals were connected to an online automated seizure warning system that monitored the T-index calculated from short-term maximum Lyapunov (STL[italic][sub]max[/sub][/italic]) exponents over time. Appropriate electrode groups for monitoring were chosen in a training session. A stimulus-response study with a wide range of parameters was conducted during the periods when there was no entrainment (interictal state). The current intensities used were set to remain well below afterdischarge (AD) threshold for each animal. Stimulation was then done on observation of a preictal entrainment, with a biphasic square-wave train of constant sub-AD threshold current intensity; freq =125 Hz; pulse width = 400 msec and duration = 10 sec using either left or right hippocampal bipolar stimulating electrodes. In this model, a convergence in STL[sub][italic]max [/italic][/sub]values (dynamical entrainment) was followed by a seizure after a mean duration of 21[plusmn]5.6 min. Hippocampal stimulation when the T-index was above the upper threshold (U[italic][sub]T [/sub][/italic]= 5) showed no discernable effects on dynamical measures. Stimulation upon entrainment, when the T-index was below a lower threshold (L[italic][sub]T [/sub][/italic]= 2.662), followed by a resetting, delayed the occurrence of a seizure by 203.7[plusmn]89.1 min. In instances where an entrainment was not reset by stimulation, a seizure occurred within 30 minutes. The mean inter-seizure interval was 2.7[plusmn]1 hrs during no stimulation and 7.2[plusmn]1.3 hrs during post-entrainment stimulation. Stimulations within 10 min. of the entrainment appeared to be more effective than longer wait periods in their ability to cause dynamical resetting. These preliminary results suggest that the mechanism underlying the anticonvulsant effect of hippocampal stimulation may involve dynamical resetting. (Supported by NIH grant R01EB002089, Children[apos]s Miracle Network, University of Florida Division of Sponsored Research and Development and Department of Veterans Affairs.)