Paul VanNess, MD                                          
Department of Neurology

Continuous video-EEG monitoring in traumatic brain injury
Traumatic brain injury (TBI) is a major cause of mortality and morbidity, particularly among persons below the age of 45. Outcome after TBI is only partly predictable by factors such as age and premorbid status of the patient and severity of initial injury, and it is clear that secondary injury resulting from altered glucose or energy metabolism, changes in cerebral blood flow, lipid peroxidation, release of excitatory amino acids, and other ongoing neurochemical changes significantly influence outcome. One possible cause of such secondary perturbations are epileptic seizures. Convulsive seizures occur during the acute period in 4-10% of patients with severe TBI, and it is likely that these behaviorally evident seizures represent only a small fraction of all epileptic phenomena. Nonconvulsive seizures, which have either no or very subtle behavioral manifestations and can be reliably diagnosed only by EEG monitoring, are substantially more common and occur in 20-35% of patients with severe TBI. The occurrence of nonconvulsive seizures is a poor prognostic sign, but it is unclear whether they a simply a reflection of an extensive insult or whether they contribute to ongoing secondary injury. Further, it is unknown whether treatment with high dose antiepileptic drugs is effective in controlling nonconvulsive seizures or in improving the neurologic outcome. Finally, such treatments, which usually include high dose benzodiazepines, phenobarbital, and occasionally even anesthetic doses of pentobarbital, carry a small but appreciable risk.
The goal of this project is to determine the feasibility efficacy of continuous video-EEG (CVEEG) monitoring and aggressive treatment of nonconvulsive seizures in patients with severe TBI. Eligible patients will be those with TBI, GCS < 12, or any unexplained deterioration in sensorium within 72 hours of admission. Enrolled patients will be monitored continuously with video and digital EEG (CVEEG monitoring), that will be continuously reviewed by trained EEG technologists with the help of computerized seizure-detection algorithms and backup from a board certified neurophysiologist. We anticipate that approximately 1 patient per week will be monitored, for a duration of 48 hours. If seizures are detected, a treatment algorithm will be activated, initially with lorazepam, followed by additional phenytoin, and, if seizures persist, pentobarbital at anesthetic doses, titrated to obtain a burst suppression electrographic pattern. The study will have one primary and three secondary endpoints:
Primary endpoint. Electrographic control of seizures by the treatment protocol. Does treatment with high dose antiepileptic drugs result in control of the seizures electrographically?
Secondary endpoint 1. Association of nonconvulsive seizures with poor outcome. Within the group randomized to invasive monitoring, is the occurrence of nonconvulsive seizures a poor prognostic sign, independent of factors such as age, severity of initial injury, or occurrence of other recognized secondary complications? Is outcome worse in patients whose nonconvulsive seizures were refractory to high dose antiepileptic drugs, compared to those whose seizures were successfully controlled?
Secondary endpoint 2. Feasibility of real-time diagnosis of nonconvulsive seizures. Prior studies using continuous EEG monitoring have identified nonconvulsive seizures through review of EEG records done hours before, and typically there is a delay of many hours before the diagnosis is established and treatment is instituted. In this study measures will be taken to attain diagnosis and treatment of nonconvulsive seizures within 30 minutes of onset, which has not been achieved in standard clinical practice. Determination of the feasibility and efficacy of these measures is an important goal of this study.
Secondary endpoint 3. Toxicity/side effects of high dose antiepileptic drugs. What are the toxicities, if any, associated with aggressive treatment of nonconvulsive seizures with high dose benzodiazepines, phenytoin, and pentobarbital anesthesia?
Role of the Doris Duke Fellow:
The fellow will be primarily involved in identifying eligible patients, arranging for video-EEG monitoring to be performed, and with the assistance of a clinical neurophysiologist (Drs. Van Ness, Agostini, or Diaz-Arrastia), monitoring the EEG for nonconvulsive seizures. Duties will also include prospectively collecting clinical information relating to the severity of the initial traumatic injury, following the patients daily during their acute hospitalization, and obtaining outcome data at 6 months, using a structured interview. This will be accomplished by having the fellow attend daily morning rounds with the neurosurgical service at Parkland Memorial Hospital, identify patients who are meet eligibility criteria for the study, and coordinate with the clinical neurophysiology department to perform video-EEG monitoring. Afternoons will be spent obtaining outcome information via telephone interviews. Opportunities to learn about EEG interpretation of patients in coma will also be available.