Abstracts

Rationale: Continuous EEG (cEEG) is being increasingly utilized in detection of and monitoring for seizures in patients admitted to Intensive Care Units (ICU). There are limited data from retrospective studies that evaluate whether use of cEEG in ICUs leads to modifications of AEDs. We investigated whether cEEG impacts management of ICU patients in comparison to matched patients who did not receive cEEG.Methods: In this prospective observational study, demographic and clinical data were collected on ICU patients who received cEEG monitoring in comparison to patients who did not; patients were matched for age, gender and admission diagnosis. Patients admitted due to seizures alone were excluded. Primary outcome was any AED change (increase, decrease, start, stop). Secondary outcomes were AED changes before, during, and after cEEG monitoring. Reasons for acquiring cEEG included: encephalopathy, seizure suspected, and witnessed seizures. The term ""seizure suspected"" was defined as suspicion for seizure raised by paroxysmal vital sign changes, intracranial pressure fluctuations, and/or description of spells atypical for a seizure. Generalized linear models were used to identify factors associated with AED changes.Results: 468 patients were recruited, equally divided into the cEEG and non-cEEG groups. Baseline characteristics are shown in Table 1. cEEG Patients had significantly longer durations of hospital and ICU stay, and significantly lower admission Glasgow Coma Scale (GCS) compared to non-cEEG patients. The mean (SD) duration of cEEG monitoring was 4.2 (4.9) days. 170 (72.6%) of cEEG patients had at least one AED change compared to only 56 (24.1%) of non-cEEG patients. Mean number of AED changes (mean-nAED) for cEEG patients was 5.5 times that for non-cEEG, after adjusting for admission GCS and the number of co-morbidities (p<0.0001; Table 1). Use of cEEG resulted in a greater number of antiepileptic drugs started, discontinued, and dosages changed (Table 1). Within the cEEG group, the mean-nAED during cEEG was 3.5 times than the mean-nAED in the interval both before and after cEEG (p<0.0001) (Table 1). 123 (52.6%) of patients had AED changes made before, 102 (43.6%) during and 48 (20.5%) after cEEG monitoring. Only 20 (8.5%) patients had AED changes made in all three periods. The most common reason for acquiring cEEG was encephalopathy, and comparison in patient characteristics and mean-nAED by reason for acquiring cEEG is detailed in Table 2. Patients with witnessed seizures had the highest mean-nAED compared to those with seizure suspected, and encephalopathy.Conclusions: cEEG monitoring in ICU with any given admission diagnosis in our study resulted in a significantly greater number of AED changes than not monitoring. Information from cEEG also aids in making decisions regarding starting or discontinuing AEDs, and changing dose. Although witnessed seizures result in most AED changes, cEEG used in patients with encephalopathy or seizures suspected also yields information to guide therapy. Data are currently being analyzed to evaluate the impact of ICU cEEG on patient outcomes.