Abstracts

Rationale: Continuous electroencephalogram (cEEG) is frequently utilized in assessing neurologic status in critically ill patients admitted to intensive care units (ICU). Patients with brain injuries such as traumatic brain injury (TBI), stroke, or subarachnoid hemorrhage (SAH), and critically ill patients without structural brain injury, are potentially at high risk for NCSz (nonconulsive seizures) or NCSE (nonconvulsive status epilepticus). cEEG aids in detection of such findings. cEEG is also valuable for monitoring response to treatment of seizures and status, for assisting in detecting cerebral ischemia and providing prognostic information. Our study aims to evaluate elements of cEEG monitoring performed at our center, including demographic characteristics of the patient population, EEG patterns, and impact on therapeutic decision making.Methods: We performed a retrospective review of 50 cEEG monitoring studies and medical records in patients admitted to the neurologic intensive care unit (NICU) at the University of Maryland between January-April 2015. Patients admitted to the NICU undergoing cEEG, and only the first cEEG monitoring periods was included in the study. Patients less than 20 yo and repeat cEEG studies were excluded. Five variables were collected: Age, Sex, Indication for cEEG, Duration of cEEG and EEG patterns. In cases of seizures detected on cEEG, records were reviewed to determine if seizure detection resulted in a change in treatment.Results: Age groups 40-59 (38%) & 60-79 (46%) were the most common groups undergoing cEEG. Males comprised 54% and females 46% of our population. Vascular etiologies were the most common indication for monitoring (54%). 32% of patients were monitored for >48 hours, 30% for 13-24 hours, 28% for 25-48 hours, 6% for 7-12 hours and 4% for 2-6 hours. The longest duration was 258 hours and shortest duration was 4 hours. Diffuse slowing was seen in 88%, focal slowing in 42%. Electro-clinical seizures were seen in 10% and subclinical seizures in 2%. Interictal epileptiform discharges (IED) were seen in 24%. Periodic patterns including generalized periodic discharges (GPD), lateralized periodic discharges (LPD), rhythmic delta activity (RDA), and stimulus-induced rhythmic, periodic, or ictal discharges (SIRPIDS) were seen in 20%. Burst-suppression pattern was seen in 2%. Identification of seizures on cEEG resulted in a change in treatment plan in all such patients.Conclusions: cEEG monitoring is commonly used in our NICU, often for a prolonged duration if needed. Within our population, vascular etiologies were the most common indication for monitoring. Our sample included only NICU patients, therefore EEG indications and findings may likely be somewhat different in other ICU’s such as MICU, Trauma and Cardiac ICUs. Although seizures were captured in only 12% of patients, cEEG provided additional information including: guidance on risk of seizures, response to treatment, and prognostication. Identification of seizures on cEEG resulted in a change in treatment for all 6 patients in our sample.