Abstracts

Rationale: An estimated 42,000 epileptic patients die from status epilepticus (SE) every year in the United States alone. Evaluation of anti-epileptic drugs and protocols for SE treatment, in terms of the dynamics of concurrently monitored electroencephalogram (EEG), may lead to the design of new, more effective treatment paradigms for successfully controlling SE. Such monitoring techniques may have a profound effect in the treatment of SE in the Emergency Department (ED) and Intensive Care Unit (ICU), where AEDs are given in rapid succession in the hope of patient recovery. In the past, techniques from nonlinear dynamics and chaos theory provided a first evidence that successful treatment with anti-epileptic drugs (AEDs) results in dynamical disentrainment (desynchronization) of entrained brain sites in SE, a phenomenon we have called dynamical resetting. We herein utilize this nonlinear dynamical analysis of EEG from two patients admitted to the ED and ICU and successfully treated with AEDs to further support our hypothesis of dynamical resetting by AEDs. Methods: Scalp EEG recordings were analyzed to estimate the short-term largest Lyapunov exponent (STLmax) profiles (measures of stability) over time. The profiles of STLmax, at the brain sites entrained in STLmax in the 10 minute period before the administration of the first AED (i.e. period where the patient was in full SE), were further monitored over time as AEDs were administered up to patient’s recovery. The convergence (entrainment) of the STLmax values across the selected sites was quantified using a paired t-test by the average T-index across pairs of these sites (low T-index values denote entrainment, high T-index values denote disentrainment). Results: During SE and prior to the administration of the first AED, the average T-index was lower than the statistical threshold of entrainment, suggesting that the corresponding critical brain sites remain entrained. Within minutes of the first AED treatment (Ativan in the first patient and Fosphenytoin in the second patient), signs of disentrainment were observed in both patients. While the T-index appeared to marginally rise above the statistical threshold of entrainment, within 10 to 20 minutes after the administration of the first AED, the brain became entrained again. The administration of subsequent AEDs (Fosphenytoin in the first patient, Fentanyl and Propofol in the second patient) dynamically disentrained the brain (statistically high T-index values) and led the patients to a slow recovery and progressively higher values of T-index. Once the patients progressed out of SE, the average T-index values remained high, and hence the respective brain sites disentrained. Conclusions: The above results indicate that the proposed measures/methodology may assist in an objective evaluation of the efficiency of current and future AEDs for the treatment of SE. Furthermore, this methodology could be clinically valuable as an independent online and real-time monitoring of the state of the brain and evaluation of the efficacy of the administered AED to a patient in SE.