Abstracts

Rationale: Management of epileptic encephalopathy with continuous spike and waves during sleep (CSWS) is based largely on expert opinion and retrospective case series. We aimed to evaluate how CSWS is currently being treated at pediatric neurology centers in the U.S., and to assess clinical and EEG response to these treatments. Methods: This was a multicenter retrospective database study, with 8 participating sites. Patients were children age 2-18 years old, who carried a diagnosis of CSWS and were among the last 10 patients treated at a participating institution between 2014 and 2016. Primary outcomes were initial treatment effectiveness, and post-treatment spike wave index (SWI) improvement of more than 50%, as judged by the evaluating neurologist. Fisher’s exact test was used to evaluate categorical values. Results: Data were collected for 54 patients. Of these, 30 (56%) were male. Median age of epilepsy onset was 3 (0-10) years, median age of CSWS diagnosis was 6 (2-11) years, and median age at CSWS treatment start was 7 (2-11) years. The median pre-treatment SWI was 90% (33-100%). Clinical characteristics included a history of seizures in 45 (83%) and a history of developmental regression in 31 (57%). The most common clinical reasons for initiation of CSWS treatment included: EEG abnormality (83%), cognitive concerns (57%), clinical seizures (44%), expressive language concerns (41%) and social/behavioral concerns (39%). Only 17 patients (32%) had a baseline neuropsychology evaluation prior to treatment. Thirteen patients (24%) had received prior treatment for CSWS at an outside institution and 43 (80%) were being treated with an antiepileptic drug (AED) at the time of CSWS diagnosis. The most common initial treatment choice was high-dose diazepam in 25 patients (46%), followed by clobazam (12 (22%)), a non-benzodiazepine AED (7 (13%)), steroids (6 (11%)), and other (4 (7%)). Outcomes after first treatment are displayed in Table 1. Twenty-five patients (46%) went on to receive a second treatment trial, which was judged to effective in 13 (52%) and reduced SWI by greater than 50% in 10 (40%). Six of these patients (11%) received a third treatment trial, which was subjectively effective in 4 (67%) and significantly reduced SWI in 3 (50%). In our small sample, neither etiology of CSWS nor choice of initial treatment for CSWS had a statistically significant association with SWI improvement (see Table 2) or subjective treatment effectiveness. However, near statistical significance (p=0.05) was reached when comparing etiology of CSWS with SWI improvement after initial treatment. When looking at structural versus non-structural etiology, the non-structural group was associated with a higher rate of SWI improvement (p=0.02). Conclusions: High-dose diazepam was the most common initial treatment for CSWS among participating U.S. pediatric neurology centers. While not statistically significant in initial analysis, higher rates of both clinical effectiveness and SWI improvement following initial treatment were achieved in patients treated with high-dose diazepam compared to other treatments. This may be limited by small numbers of patients in certain treatment groups. A non-structural etiology of CSWS was associated with favorable EEG response to initial treatment for CSWS. Funding: No funding was received in support of this abstract.