Predictors of one-year language and seizure outcomes in children with epileptic encephalopathy with continuous spike-and-wave during sleep (CSWS).

PURPOSE: Epileptic encephalopathy with continuous spike-and-wave during sleep (CSWS), with its associated impact on language, is an important cause of morbidity with epilepsy in children. The effects of various treatment-approaches and the aetiological/electrophysiological factors affecting therapeutic response are not fully understood.

METHOD: A retrospective study of patients admitted to the institute and diagnosed to have CSWS pattern on EEG was conducted. Spike and Wave Frequency/100 s(SWF) was calculated from sleep-EEG records. Language development and seizure outcomes were assessed at baseline and 1 year.

RESULTS: Fifty-two children were included (idiopathic CSWS, N = 19; symptomatic CSWS N = 33).The 2 groups differed in terms of younger age at seizure onset in symptomatic CSWS (p = 0.006), early age at language regression (p = 0.046), history of neonatal seizures (p = 0.038) and slowing of background activity on EEG (p = 0.024). Language regression was noted in 63.5 % of the cohort. Twenty-five (48%) patients received steroids with improvement in seizures (p < 0.001). Twenty-one (40.3 %) received steroids and intravenous immunoglobulin (IVIG) with improvement in seizure score (p = 0.002) at 1 year. Both immune-modulation arms irrespective of etiological subgroups had comparable proportions of children with expressive and receptive language gains. On 1 year follow-up, seizure remission was noted in 13(25 %) patients, with improved seizure score in 32/39 (61.5 %) patients and language improvement in 32 children (60.8 %). Patients with normal background on baseline EEG, generalised spikes, absent frontal-negative spikes and SWF < 170 were found to have improved language estimates on follow-up.

CONCLUSIONS: Despite clinical and therapeutic outcome differences between idiopathic and symptomatic CSWS, immune-modulation appears effective irrespective of aetiology. Analysis of EEG variables enables prediction of language outcomes at 1 year follow-up.