Abstracts

Rationale: Refractory status epilepticus (RSE) is a substantial population of status epilepticus (SE) unresponsive to standard 1st and 2nd line medications. Criteria for RSE depend on duration of SE or medication failure, both of which delay immediate and aggressive treatments. A marker of RSE at the presentation of the patient would help optimize immediate abortive therapy. A potential marker is the electroencephalographic (EEG) pattern seen during SE. Five progressive EEG stages have been shown in several SE models (figure 1). From prior experiments, we have found EEG stage 3 (a pattern of continuous spike discharges) to be highly refractory to standard SE treatment. Therefore, in the lithium-pilocarpine experimental SE rodent model, we tested the hypothesis that the stage 3 pattern was a better predictor of treatment refractoriness than other EEG stages (i.e. stage 1 and 5) or the duration of SE. Methods: 87 male adult Sprague-Dawley were each implanted with 4 epidural recording electrodes. One week postsurgery, EEG monitoring was started. SE was induced with lithium (3 mmol/kg, intraperitoneal IP) and, 24 hours later, pilocarpine (30 mg/kg, subcutaneous) injections. In the control group, no abortive treatment was given. In 3 other groups, treatment occurred at stage 1, 3, or 5 with diazepam (10mg/kg) and phenobarbital (25mg/kg) injections IP. EEG monitoring continued for an additional 48 hours. Results: Of 16 control animals, all progressed through the 5 EEG stages of SE without spontaneous abortion. Treatment in the stage 1 group was given after the second electrographic seizure; in the 17 animals of this group, there was 100% abortion rate. Treatment in stage 3 involved 34 animals; use of the same abortive medications resulted in only a 3% abortive rate. Treatment in stage 5 with 20 animals resulted in an intermediate 21% abortive rate. Therefore the stage 3 EEG pattern was highly associated with a refractory SE state. Time-to-treatment (as measured from pilocarpine injection) was used as a measure of SE duration for each animal. Comparison of the stage 1 and 3 treatment groups showed an overlap in the range of 30 to 60 minutes from pilocarpine injection (figure 2). Therefore, in this time range, treatment response was better predicted by EEG stage than by SE duration.Conclusions: The results of this study show that the EEG is a potential tool for guiding SE treatment. Confirmation of this finding in other SE models is necessary and would help elucidate the mechanisms that underlie the changes in EEG pattern and responsiveness to treatment. Further examination of the excluded EEG stages of 2 and 4 would also be helpful in understanding of the transitions between responsive SE and RSE. The increased response rate from stage 3 to stage 5 is an unexpected finding. Further studies are needed to understand the mechanism of this paradoxical observation. (Source of funding: The Barrow Neurological Foundation)