Abstracts

Rationale: Inpatient video EEG monitoring is a critical tool for characterizing paroxysmal events as epileptic or non-epileptic and for localizing seizures in order to guide therapeutic decisions. It is often necessary to capture events through discontinuation or reduction of antiepileptic drugs (AEDs), but there are no guidelines as to how quickly or to what degree they can be reduced to limit length of stay without challenging patient safety. This study describes our experience tapering AEDs beginning with the morning dose prior to admission to the epilepsy monitoring unit (EMU). Methods: We conducted a retrospective chart review of patients admitted to our EMU during 2009. Background information collected included age, sex, and history of status epilepticus or convulsion. According to our standard practice, patients were contacted by phone the week prior to admission to update seizure history. Patients who were without daily seizures, who had no history of status epilepticus, and for whom admission prior to 2pm was feasible were instructed to hold their morning dose the day of admission and given a reduced AED dosage once in the EMU. We determined length of stay, amount of AED reduction, and time from admission to dose nadir and first typical event. Complications were noted and included seizure prior to admission or to EEG hook-up and status epilepticus. Yield of seizure recording was also determined. Results: There were 125 admissions to our EMU in 2009. Among them, 115 were for recording of clinical events. Eighty-two of the 115 were women (71%). Mean age was 40 years (range 17-76). In 39 patients (34%), AEDs were stopped with the 8am dose and, in 40 patients (35%), the AED dose was reduced with the 8am dose. Twenty patients (17%) were not taking an AED. In 8 patients, AEDs were tapered after admission. The remaining 8 had no dose reduction. Eighty-five patients (74%) had a typical event recorded. The mean length of stay was 74.3 hours (range 25.4 to 525.8). The mean time from admission to dose nadir was 9.1 hours (range 0 to 102.5). The mean time from admission to first typical event was 15.9 hours (range 0.6 to 68). Fifty patients (43%) were discharged with a diagnosis of epilepsy and 38 (33%) of non-epileptic events. Four patients with epilepsy were found to also have non-epileptic events. In 27 patients (24%), a definitive diagnosis could not be made. One patient with a history of daily seizures had a non-convulsive seizure in the hours prior to admission. Another had a non-convulsive seizure between admission and EEG hook-up which was only captured on video. None of the patients were found to be in status epilepticus upon or during the admission. Conclusions: In an EMU with scheduled admissions before 2PM and updated information regarding seizure type and frequency, AEDs can safely be held or reduced the morning of admission. In our practice, this has led to a mean length of stay of 3.1 days (74.3 hours) with a 74% efficiency of recording typical events.