Abstracts

Rationale: Non-convulsive seizures (NCS) and other electrographic abnormalities are common in emergency department (ED) patients with altered mental status (AMS). Non-convulsive status epilepticus (NCSE) is serious, treatable acute neurological emergency given the time-dependent survival of neurons during seizure activity. Approximately 5% of ED patients with AMS have NCS (including NCSE). According to the literature, approximately half of the patients with NCSE are diagnosed more than 24 hours after arrival to the ED. Therefore, there is a clear need for early, accurate diagnosis of NCS/NCSE on electroencephalograms (EEG) and initiating the treatment as soon as possible.Early ED-based diagnosis and treatment of NCS/NCSE requires that an EEG is recorded and interpreted in a timely fashion, as soon as the high risk for NCS/NCSE is determined. Since ED physicians encounter such patients first in the ED, they should be familiar with general EEG principles as well as the EEG patterns of NCS/NCSE. Objective: To test the utility of a brief training module in enhancing and assessing ED physicians’ ability to identify seizures (NCS/NCSE) on EEG. Methods: This was a randomized controlled trial conducted in 3 academic institutions. Board certified emergency medicine physicians were recruited. Those with previous neurology or EEG training were excluded. Variables included level of experience (years from graduation), gender, and test scores. Subjects were randomized to control or intervention groups using a random number generating software. Participants allocated to the intervention group received a self-learning PowerPoint presentation (training module) and were asked to take a quiz after reviewing the PowerPoint presentation. The control group was asked to take the quiz without reviewing the training slides.EEG training module: A slide presentation describing the basic principles of EEG including EEG recording techniques, montages, and views, followed by characteristics of normal and abnormal patterns was developed with assistance of epileptologists and experts in educational research. The goal of the presentation was to familiarize the participants with EEG patterns of seizure.Test material: Participants in both groups were tested in their ability to identify abnormal from normal EEG as well as seizure by reviewing 20 test EEGs (one-page snapshots). These de-identified EEGs were previously recorded from actual patients. Each test EEG was accompanied by two questions: normal or abnormal, and seizure versus no seizure. The test scores range from 0 (all wrong answers) and 40 (all correct answers).Outcomes: Overall correct percentage of test scores were calculated upon completion of the training.Data are reported as medians and quartiles. The medians for percentages of correct answers were calculated and compared between the two groups using Mann-Whitney-U test. Results: A total of 30 emergency physicians were enrolled (10 per site, total of 15 controls and 15 interventions). Participants were 52% male with median years of practice of 9.5 years (3,14). Groups were similar in regards to years of practice and gender. The percentage of correct answers in the intervention group (65%: 63%, 75%) was significantly different (p=0.002) from that of control group (50%: 45%, 60%). Conclusions: A brief self-learning training module improved the ability of emergency physicians in identifying EEG seizure patterns. Funding: None