Presenter Status
Fellow
Abstract Type
Clinical Research
Primary Mentor or Principal Investigator
Jessica Wallisch
Presentation Type
Oral Presentation
Start Date
14-5-2026 12:15 PM
End Date
14-5-2026 12:30 PM
Abstract Text
Background:
Accurate prognostication after pediatric severe traumatic brain injury (sTBI) remains challenging. We examined the utility of continuous EEG (cEEG) and Rotterdam head CT score in predicting short-term favorable outcome.
Objectives/Goal:
To determine whether specific characteristic features or patterns on head CT and continuous EEG (cEEG) monitoring are associated with favorable short-term outcome following severe TBI in children.
Methods/Design:
7-year (01/2018-12/2024) retrospective study of children aged 1.5-18 years admitted to our PICU with sTBI and receiving cEEG monitoring per institutional protocol. Early favorable outcome was defined as Pediatric Cerebral Performance Category scale 1-3 at PICU discharge. cEEG features (background, sleep architecture, reactivity, and seizure) during the first 12-hours of monitoring and initial Rotterdam head CT scores were recorded. Demographics, GCS score, pupillary reactivity, injury severity score (ISS), clinical seizure, cardiac arrest, and neurosurgical intervention data were collected. Significant associations with favorable outcome were analyzed by Chi-square or Fischer Exact tests for categorial data or bivariate logistic regression for continuous data. A stepwise multivariate logistic regression analysis was performed to determine significant independent associations with favorable outcome.
Results:
54 patients were studied. Median [IQR] age: 8.9 [4.5-14.8] yrs; initial GCS score: 3 [3-6]; and Rotterdam head CT score: 3 [2-5]; 33 (61%) were male; 12 (22%) had clinical seizures; 28 (52%) received neurosurgical intervention; and 41 (76%) survived. In univariate analysis, favorable outcome was significantly associated (OR[95%CI]) with GCS score (1.43 [1.04-1.97], p=0.029); pupillary reactivity (9.0 [1.77-45.71], p=0.005); Rotterdam CT score< 3 (7.14 [1.72-29.68], p=0.006); and presence of reactivity (20.0 [4.95-80.89], p< 0.001) and absence of seizure (undefined [~12.4], p=0.024) on cEEG; and inversely associated with ISS (0.87 [0.81-0.94], p< 0.001) and neurosurgical intervention (0.188 [0.06-0.57], p=0.004). Stepwise logistic regression revealed significant independent associations with favorable outcome for presence of reactivity on cEEG (9.93 [1.83-53.82], p=0.008) and Rotterdam CT score< 3 (12.14 [1.02-145.10], p=0.049), and an inverse association for ISS (0.88 [0.80-0.97], p=0.009).
Conclusions:
Reactivity on cEEG, Rotterdam head CT score < 3, and lower ISS independently predict early favorable outcome and may enhance future development of better neuroprognostication models in pediatric sTBI.
EEG Reactivity and Rotterdam Head CT Score Predict Early Favorable Outcome in Pediatric Severe Traumatic Brain Injury
Background:
Accurate prognostication after pediatric severe traumatic brain injury (sTBI) remains challenging. We examined the utility of continuous EEG (cEEG) and Rotterdam head CT score in predicting short-term favorable outcome.
Objectives/Goal:
To determine whether specific characteristic features or patterns on head CT and continuous EEG (cEEG) monitoring are associated with favorable short-term outcome following severe TBI in children.
Methods/Design:
7-year (01/2018-12/2024) retrospective study of children aged 1.5-18 years admitted to our PICU with sTBI and receiving cEEG monitoring per institutional protocol. Early favorable outcome was defined as Pediatric Cerebral Performance Category scale 1-3 at PICU discharge. cEEG features (background, sleep architecture, reactivity, and seizure) during the first 12-hours of monitoring and initial Rotterdam head CT scores were recorded. Demographics, GCS score, pupillary reactivity, injury severity score (ISS), clinical seizure, cardiac arrest, and neurosurgical intervention data were collected. Significant associations with favorable outcome were analyzed by Chi-square or Fischer Exact tests for categorial data or bivariate logistic regression for continuous data. A stepwise multivariate logistic regression analysis was performed to determine significant independent associations with favorable outcome.
Results:
54 patients were studied. Median [IQR] age: 8.9 [4.5-14.8] yrs; initial GCS score: 3 [3-6]; and Rotterdam head CT score: 3 [2-5]; 33 (61%) were male; 12 (22%) had clinical seizures; 28 (52%) received neurosurgical intervention; and 41 (76%) survived. In univariate analysis, favorable outcome was significantly associated (OR[95%CI]) with GCS score (1.43 [1.04-1.97], p=0.029); pupillary reactivity (9.0 [1.77-45.71], p=0.005); Rotterdam CT score< 3 (7.14 [1.72-29.68], p=0.006); and presence of reactivity (20.0 [4.95-80.89], p< 0.001) and absence of seizure (undefined [~12.4], p=0.024) on cEEG; and inversely associated with ISS (0.87 [0.81-0.94], p< 0.001) and neurosurgical intervention (0.188 [0.06-0.57], p=0.004). Stepwise logistic regression revealed significant independent associations with favorable outcome for presence of reactivity on cEEG (9.93 [1.83-53.82], p=0.008) and Rotterdam CT score< 3 (12.14 [1.02-145.10], p=0.049), and an inverse association for ISS (0.88 [0.80-0.97], p=0.009).
Conclusions:
Reactivity on cEEG, Rotterdam head CT score < 3, and lower ISS independently predict early favorable outcome and may enhance future development of better neuroprognostication models in pediatric sTBI.
