Automated quantification of spike-wave activity may be used to predict the development of electrical status epilepticus in sleep (ESES) in children with perinatal stroke

Objective: Continuous spike and wave in slow-wave sleep (CSWS), an epileptic encephalopathy, occurs after perinatal stroke where it is associated with cognitive decline. CSWS features a distinct EEG pattern, electrical status epilepticus in sleep (ESES). Biomarkers for the prediction of ESES have not been identified but will facilitate earlier diagnosis and treatment. We hypothesized that spike-frequency and differences in power spectra would be predictive of subsequent ESES.

Methods: A cross-sectional study comparing EEG spike-frequency and Power before the development of ESES in patients with perinatal stroke, patients with focal epilepsy, and appropriate controls.

Results: 43 patients met the inclusion criteria; 11 stroke-ESES, 10 stroke controls, 14 epilepsy-ESES, 8 epilepsy controls. ESES patients had higher pre-diagnosis mean spike-frequency (24.0 ± 24 versus 6.6 ± 9.1 SW/min, p = 0.002) than patients that did not develop ESES; these differences present ~ 3 years before ESES diagnosis. Pre-diagnosis, normalized delta power (1-4 Hz) was higher in the stroke-ESES group (105.7 ± 58 dB/Hz) compared to stroke controls (57.4 ± 45 dB/Hz, p = 0.036).

Conclusion: Spike-frequency and delta power may represent EEG biomarkers of the risk of developing ESES in children with perinatal stroke.

Significance: EEG biomarkers may be used by clinicians to assess which patients are more at-risk for ESES. Using spike-frequency, clinicians may be able to identify patients at risk of developing ESES.