Abstracts

Rationale: The prompt management of repetitive and/or prolonged seizures in the epilepsy monitoring unit (EMU) is of paramount importance and usually requires a rapid and safe administration of an anticonvulsant. Historically, intravenous lorazepam (IVL) has been used for this indication at our institution. In January 2016, our epilepsy center transitioned from IVL to intranasal midazolam (INM) due to its ease of administration and potential improved patient comfort as it does not require intravenous access. However, a formal comparison between these two approaches (IVL versus INM) in terminating seizures in adult EMU is lacking. We sought to compare the efficacy and safety of INM to IVL as a seizure rescue medication in the EMU. Methods: A retrospective cohort evaluation was performed of consecutive adults undergoing monitoring in the EMU of a single Comprehensive Epilepsy Center who received IVL (2015) vs INM (2016) for seizure termination between Jan 2015 – Dec 2016. Patient characteristics including demographics, type of epilepsy, comorbidities, and anti-seizure drug (ASD) regimen were abstracted. Outcomes evaluated included the time from seizure onset to administration of rescue medication, time to recurrent seizure after drug administration, type and duration of seizure, the number of rescue drug therapies required for termination of seizures, and any adverse events including progression to status epilepticus, transfer to the intensive care unit, and hospital length of stay. Chi-square or Fisher’s exact tests were used for categorical variables, t-test was used for continuous variables with normal distribution, and Wilcoxon rank sum test for non-normally distributed data. Results: A total of 484 records of admitted patients to the EMU were screened. Twenty-seven patients in IVL group and 23 in INM group qualified for the final analysis.  There were no differences in the baseline demographics severity of epilepsy as measured by the number of prescribed ASDs prior to admission, type and duration of seizures, or number of rescue drug therapies between the two groups. Similar times to drug administration after initial seizure (p=0.15) [figure 1] and times to recurrent seizure (p=0.10)  [figure 2] were found between IVL and INM groups. Not surprisingly, more frequent need for IV access change was found in the IVL vs INM groups (4 vs 0.6, < 0.001), although there were no differences in documented complications related to injection, or incidence of respiratory depression, hypotension, aspiration, and other seizure-related adverse events. Conclusions: The results of our study suggest that INM is comparable to IVL for the treatment of prolonged seizures or seizures clusters in the EMU; thus, we think it is the preferred therapy for most patients. However, our study was limited by small size and did not include any quality indicators or patient comfort measures.  Future prospective randomized controlled trials comparing INM to IVL are needed to determine if INM is truly a preferred therapy in this setting. Funding: The study was supported in part by the Swebilius Foundation.