Abstracts

Rationale: Continuous EEG (cEEG) monitoring involves prolonged recording of EEG and is typically performed at the bedside in critically ill, hospitalized patients. It is increasingly being used to evaluate patients for unexplained change in mental status, detection of subclinical (nonconvulsive) seizures and management of status epilepticus (SE). The diagnostic yield of the test has not been fully characterized. There are no definite recommendations regarding how long it should be continued before nonconvulsive seizures or status epilepticus are excluded, and when to stop the study after controlling seizures. Its influence on the clinical outcome of SE has not been established. Methods: All adult patients who underwent cEEG monitoring between 01/01/2007 and 03/15/2010 at Penn State University Hershey Medical Center were included in this retrospective study. Video was simultaneously recorded with cEEG at the bedside in all patients. Patients admitted electively to the epilepsy monitoring unit for diagnostic or presurgical evaluation were excluded. Results: 132 patients were identified. Of these, 60 were males and 74 were females. The mean age of the patients at the time of the cEEG was 59.5 years (range: 22-94 years). The mean duration of cEEG monitoring was 58.2 hours (range: 4-360 hours). 55 (41.6%) patients had status epilepticus. Of these, 19 (35%) had clinical status epilepticus (CSE) and 36 (65%) had only nonconvulsive status epilepticus (NCSE). Among all patients with status epilepticus (CSE and NCSE), 15 (27%) patients had recurrence of nonconvulsive seizures after initial resolution of the status. In 3 (5%) patients, SE was not noted during the initial portion of the cEEG recording but was identified within 24 hours, and in 2 (3%), SE was detected only after 24 hours of cEEG monitoring. In addition, in 2 (3%) patients, SE was not identified in an initial routine 30-minute EEG performed before cEEG was started. 36 (27%) of all patients who underwent cEEG died. Among patients with SE, 22 (40%) died, 7 (13%) were discharged to a long term acute care hospital, 10 (18%) were discharged to an acute rehabilitation facility, 3 (5%) were discharged to other facilities, and 13 (24%) recovered fully and were discharged home. Conclusions: cEEG monitoring is helpful in the accurate diagnosis and optimal management of SE. It enables the identification of NCSE in a substantial proportion of patients as the sole manifestation of SE or following resolution of CSE. Most seizures are detected within the first 24 hours but longer monitoring may be required in some patients. Routine EEGs or initial portions of cEEG monitoring may fail to record NCSE. Status epilepticus is associated with high mortality and morbidity, but early detection and treatment of nonconvulsive seizures and status using cEEG may improve the clinical outcome.