Abstracts

Rationale: Early and accurate seizure detection is an important and challenging problem among most medical centers in the United States. Nonconvulsive seizures (NCS), or subtle seizures, often have very subtle signs and are unrecognized. Emergent or stat EEGs (sEEG) are essential in the evaluation of patients with acute or unexplained mental status changes to rule out either subtle or subclinical seizure activity. A prior study found that the delay from clinical deficit until EEG diagnosis of NCS or nonconvulsive status epilepticus (NCSE) was as long as 48 hours for patients who had clinically observable seizures. For those without any overt signs of seizure activity, the delay was even longer at 72 hours. Several studies have clearly shown that seizure duration and delay to diagnosis is associated with increased morbidity and mortality. This study will examine the various contributing factors in the delays of detection and treatment of nonconvulsive seizures in a tertiary care center, in the hopes of reducing the time to diagnosis and treatment and improving clinical outcome. Methods: We collected clinical and EEG data from 81 consecutive patients from 9/2009 to 12/2013 with electrographic seizures detected on EEG at Beth Israel Deaconess Medical Center (BIDMC), a tertiary care center. The following time points were collected: 1. Time of seizure concern, 2. Time of EEG request, 3.Time of EEG recording, 4. Time of first antiepileptic drug (AED) administration. Results: The time delays are illustrated in Figure 1a. The average time from seizure concern to AED order is 48 hours and 29 minutes, which is shorter than the previously reported 48 -72 hours. The average delay from seizure concern to EEG request is 6 hours and 50 minutes. This delay may be attributed to lack of clinical knowledge or concern of NCS in patients with altered mental status by the treating physician. The average time from EEG order to start of EEG recording is 10 hours and 38 minutes, likely attributable to a number of factors: shortage of EEG technologists during the day, inadequate night coverage, insufficient equipment availability, and the time-consuming EEG set-up process. Most importantly, even after an electrographic seizure is clearly identified, the average time to first AED order is 10 hours and 10 minutes. This delay is mostly related to the lack of immediate EEG interpretation by a board-certified epileptologist and hence lack of treatment recommendations. Conclusions: In summary, NCS and subtle seizures remain an under-diagnosed and sub-optimally managed entity, despite the current literature available regarding the morbidity and mortality associated with delayed diagnosis and treatment. Timely management relies upon a high index of suspicion by the clinician and immediate access to sEEGs. These findings underscore the multiple factors that contribute to the significant gap that currently exists in the diagnosis and treatment of NCS and NCSE. It is imperative that we develop an effective clinical protocol and diagnostic tool to identify NCS and NCSE as early as possible to potentially improve the clinical outcome of these patients.