Abstracts

Status epilepticus (SE) is defined as recurrent epileptic seizures (szs) without full recovery between the szs, or as continuous sz activity for a fixed period of time (5, 10, or 30 minutes). Operationally, sz activity is identified clinically, or on the basis of EEG changes thought to be ictal. However, there is no independent measure of [quot]ictalness[quot] with which to validate a clinical or EEG diagnosis of SE. We used nonlinear dynamical analysis of the EEG to attempt to identify an objective marker of SE. SE was induced in 15 week old male Sprague-Dawley rats 5 days after implantation of Teflon coated tungsten wire electrodes (175 [mu]m diameter); 4 in the cortex (A-P +2.0 [plusmn] 3.0 mm, and A-P -4.0 mm [plusmn] 3.0 mm lateral to midline to depth 1.0 mm) and 2 ventral hippocampal electrodes (A-P -5.6 mm [plusmn] 4.5 mm lateral to midline to depth of 5.0 mm). EEG was recorded continuously during the entire experiment with a 2 hr baseline EEG prior to SC injection of 10 mg/kg kainic acid (KA) to induce SE, and continued until cessation of epileptiform discharges. EEG SE Stages I (discrete seizures), III (continuous ictal activity), and V (periodic epileptiform discharges) (Treiman et al., Epilepsy Res. :49-60, 1990) were identified by visual inspection. The EEG data were analyzed using nonlinear dynamical techniques that use the convergence of the largest short-term Lyapunov exponent over time at each electrode site to statistically quantify the brain[apos]s dynamical entrainment (Iasemidis et al., IEEE TBME, :616-627, 2003). Mean time from KA injection to onset of Stage I was 7.3 ([plusmn] 2.1) min, Stage III 177.3 ([plusmn] 32.1) min, and Stage V 670.3 ([plusmn] 235.9) min. At onset of Stage I entrainment was observed (drop in T-index). However, this was not statistically significant with [alpha]=0.01 and did not reset (return to baseline after each discrete seizure, as occurs with isolated seizures). Entrainment was observed to progress towards the statistical threshold and by the onset of Stage II was statistically significant. Entrainment fluctuated around the threshold throughout the remaining SE stages until the PED frequency began to slow as the rat showed spontaneous recovery (up to 36 hrs after injection; PED frequency 1 per 4-5 sec) and gradually returned to baseline. These results suggest that nonlinear dynamical analysis of the EEG can provide an independent marker of SE and that electrophysiologically, SE may consist of two phases. During EEG Stage I, SE is characterized by a moderate drop in the T-index with failure to reset, in contrast to what occurs with single seizures. During Stages II-V there is a greater drop in the T index which is statistically significant for entrainment and a continuing failure to reset. This observation needs to be studied in other experimental models and in human SE, but may provide another parameter for the diagnosis of SE. (Supported by Barrow Neurological Foundation.)