Abstracts

Rationale: Electrical Status Epilepticus of Sleep (ESES) is an electrographic pattern characterized by nearly continuous spike-wave discharges comprising > 85% of non-REM sleep. It is associated with variety of cognitive and behavioral problems. An overnight EEG is considered the gold standard for diagnosing ESES and compressed spectral array (CSA) is reported to be a useful to examine the changes in EEG power. This study has two aims: 1) determine if CSA is informative to calculate the spike index during non-REM sleep; 2) compare spike index of the initial first 30 minute and remaining non-REM sleep EEG to establish the diagnosis of ESES.Methods: Six prolonged video EEG recording were retrieved in four children, age 5 to 10 years, with the diagnosis of ESES. Two patients had two independent monitoring, on two separate occasions. All patients had various degrees of cognitive and behavioral delay, were diagnosed with symptomatic generalized or multifocal epilepsy and presented to EMU due to increased frequency of clinical seizures. CSA analysis (using MagicMarker, Persyst, Inc, Prescott) was performed on each EEG data at the time of recording. Spike index was calculated in the first 30 minutes and remaining of non-REM sleep recording for comparison. Three EEG segments, a-100 second long, were identified randomly within initial 30 minutes (ISI-100) and remaining non-REM recording (RSI-100) based on CSA findings. Then, these selected EEG segments were reviewed visually to confirm non-REM sleep and to determine spike index. Sleep was determined based on at least 1 minute after loss of muscle artifact, loss of anterior to posterior gradient and loss of posterior reactive rhythm on EEG. Results: All EEG recordings (100%, 6/6) fulfilled the diagnostic criteria for ESES. Focal ESES was identified in two EEG files on the same patient. When ISI-100 and RSI-100 were compared, initial 30 minutes EEG recording showed the same or increased percentage of non-REM sleep occupied by generalized or focal spike-wave discharges. In six EEG files, ISI-100 were 98%, 94%, 90%, 85%, 100% and 100% (mean 94.5%) whereas RSI-100 were 96%, 96%, 90%, 85%, 90% and 100% (mean 92.8%). Conclusions: CSA remains a useful tool to analyze EEG power in non-REM sleep recording in children with ESES. Furthermore, CSA-guided spike index analysis within the first 30 minutes can be considered as a reliable and practical method of analysis to predict the diagnosis of ESES.