Rationale: According to the ILAE definition of epileptic syndromes from 2022, DEE or EE – SWAS (spike wave activation in sleep) is characterized by cognitive, language, behavioral and motor regression associated with an EEG pattern of activation of discharges during Stage II sleep. Measuring spike wave index (SWI) is used to quantify the activation of discharges during sleep, and historically, varying cut-offs between 50-85% were used for a SWAS diagnosis. Studies suggest that standardized scores like the “BASED” score used in hypsarrhythmia can increase interrater agreement and reporting consistency in EEG. Here we aim to develop a quick, easy to understand, and reliable grading scale for SWAS.
Methods: In this retrospective study, we analyzed EEGs of 100 patients who were requested for measuring spike and wave activation during sleep from January 2023-May 2024.
Three licensed neurophysiologists randomly and blinded reviewed all recordings. Each EEG was evaluated twice using the traditional SWAS evaluation (presence or absence of SWAS), and the newly developed standardized EEG grading scale (SWAS-I, figure 1).
The SWAS-I grading scale includes the following parameters: the presence or absence of activation of spikes during sleep, the percentage of spike wave index and the location of spikes (focal, unilateral, or bilateral). An awake and stage II sleep recording were assessed to verify activation during sleep. Five minutes of Stage 2 sleep was used to calculate the SWI.
Results: Preliminary results from thirty EEGs are reported here. Using the traditional SWAS evaluation, Cohen’s kappa on the presence or absence of SWAS was 0.73, spread of SWAS (either generalized or non-generalized) was 0.63
.In the SWAS-I score, Cohen’s kappa was 0.83 (figure 2). There was high interrater reliability in categorizing patients in the ‘low likelihood of SWAS’ and ‘high disease burden SWAS’. Differing results were seen in the middle of the SWAS spectrum. Interrater reliability was also calculated for the spread location (no spread, focal, unilateral, or bilateral spread), with a kappa score of 0.39.
Conclusions: Our preliminary analysis suggest that a high interrater agreement can be achieved using the standardized SWAS-I score. For the remaining EEGs we will adjust the use to provide more detailed definitions especially in the score of “spread” which showed the largest variability. High interrater agreement is a first step towards developing a standardized reporting of SWAS. We hope that SWAS-I will not only allow to diagnose SWAS but also provide information about improvements and deterioration over time and aim to correlate the score with clinical changes.
Funding: none