Abstracts

Rationale: Comatose critically-ill children may experience seizures, many of which are subclinical and cannot be detected without continuous EEG monitoring (cEEG). A high seizure burden during critical illness has been associated with worse neurodevelopmental outcome, motivating greater seizure surveillance and intervention.  In this study, we aimed to measure the prevalence and burden of seizures among comatose critically ill children of diverse causes. Methods: A prospective multicenter cohort study of critically-ill children with coma of all causes was conducted at The Hospital for Sick Children in Toronto, Duke University Children’s Hospital, The Children’s Hospital of Philadelphia, and the UCSF Children's Hospital.  Eligible patients included infants and children (term neonates – age 18y) admitted to a pediatric intensive care unit with a Glasgow Coma Scale (GCS) score ≤8 for at least 1 hour. With informed consent, participants underwent at least 48h of cEEG with video, which was initiated on a research basis if not already ordered as part of clinical care. cEEG recordings underwent contemporaneous clinical interpretation with results reported to the clinical team, who were free to initiate anti-seizure therapies.  Subsequent research cEEG interpretation was performed by two independent investigators unaware of any clinical data, who annotated the duration, spatial extent and clinical correlate of every electrographic seizure. A consensus process was used to resolve inter-rater discrepancies. Results: Screening of 920 critically ill children identified 404 eligible subjects, of whom 279 were enrolled and 13 withdrew, resulting in a cohort of 266 subjects. The median (IQR) age was 4 (1-10) years and 53% were male (Figure and Table). Median GCS at coma onset was 6 (3-7). Coma was attributed to multiple causes in most children, including sedative medications (68%), systemic illness (60%), brain injury (60%) and seizures/status epilepticus (55%). cEEG was initiated on a research basis in 78 subjects (29%) and on a clinical basis in 187 (71%). Median cEEG duration was 49 (34-73) hours. Seizures were identified in 73 subjects (27%), of whom 67 (92%) experienced some subclinical seizures and 42 (58%) experienced exclusively subclinical seizures. Subjects experienced a mean (SD) of 80.2 (116.2) seizures, with a mean total seizure burden of 159.4 (180.8) minutes, and mean maximum hourly seizure burden (MHSB) of 23.5 (18.0) minutes. Status epilepticus (defined as a MHSB ≥ 30 min) occurred in 28 subjects (11%). The median time from cEEG onset to first seizure was 113 (11-639) minutes. Conclusions: Seizures were common (27%) in this cohort of comatose critically ill children, and when they occurred were often subclinical (92%).  Furthermore, the seizure burden was high: 64% of those with seizures experienced a MHSB ≥ 12 minutes and 11% had status epilepticus, both of which have been associated with adverse neurodevelopmental outcome (Crit Care Med 2013;41:215-23, Brain 2014;137:1429-38, Neurology 2014;82:396-404). The contribution of seizure burden to neurodevelopmental outcome will be assessed in subsequent analyses. Funding: This work was supported by research grants from the Canadian Institutes of Health Research, the SickKids Foundation, the PSI Foundation and the National Institutes of Health (K23NS076550).