Rationale:
Post-traumatic epilepsy (PTE) represents the most common cause of acquired epilepsy, occurring in up to 20% of children following severe traumatic brain injury (TBI). There is a lack of understanding of the mechanisms that follow trauma leading to epileptogenesis. Our goal was to investigate predictive biomarkers from acute neurophysiologic data in children with severe TBI who later developed PTE.
Method:
Data from the first 7 days of invasive multimodality neurologic monitoring were retrospectively analyzed from children with TBI at Phoenix Children’s Hospital from 2014 to 2019. Initial injury severity was classified using the Glasgow Coma Scale (GCS). Continuous electroencephalographic (EEG) data were evaluated for the presence of seizures within the first 24 hours of injury, seizures within 2-7 days after initial injury, interictal epileptiform discharges (EDs), and sleep spindle asymmetry. Model-based indices of cerebral autoregulation were calculated including the pressure-reactivity index (PRx) and wavelet pressure-reactivity index (wPRx). Measures of autonomic function included heart rate standard deviation (HRsd) to assess heart rate variability as well as baroreflex sensitivity (BRs). We also investigated intracranial pressure (ICP), cerebral perfusion pressure (CPP), arterial blood pressure (ABP) and heart rate (HR) as physiologic predictor variables. Univariate and multivariate logistic regression were used to model the odds of PTE and estimate the odds ratio (OR) of reported variables. PTE was categorized through diagnosis by an epileptologist at 1 year post-injury.
Results:
There were 62 children of which 11 developed PTE (17.7%). Using univariate logistic regression, the presence of seizures between days 2-7 (OR=16.13, p=0.0004, C=0.77), sleep spindle asymmetry (OR=13.12, p=0.0007, C=0.76), EDs (OR=7.52, p=0.0056, C=0.70), higher median ICP (OR=1.24, p=0.0232, C=0.73), PRx (OR=23.22, p=0.0441, C=0.64), wPRx (OR=28.95, p=0.0443, C=0.65) and lower median HRsd (OR=0.51, p=0.0061, C=0.81) were associated with PTE, whereas gender (OR=0.41, p=0.2821, C=0.59), initial GCS (OR=0.80, p=0.1302, C=0.65), CPP (OR 0.99, p=0.7386, C=0.50), ABP (OR=1.05, p=0.2850, C=0.59), HR (OR=0.90, p=0.5508, C=0.56), BRs (OR=0.85, p=0.1118, C=0.68) and seizures within the initial 24 hours post-injury (OR=1.86, p=0.5784, c=0.53)) were not. Using multivariate logistic regression models where each variable was entered individually after adjusting for GCS, seizures between days 2-7 (OR=16.13, p=0.0006, C=0.81), EDs (OR=7.94, p=0.0063, C=0.74), sleep spindle asymmetry (OR=11.34, p=0.0017, C=0.79), higher ICP (OR=1.26, p-0.0206, C=0.78) and lower HRsd (OR=0.53, p=0.0107, C=0.80) were associated with PTE, whereas median PRx (OR=20.47, p=0.0510, C=0.75) and wPRx (OR=23.71, p=0.0565, C=0.74) approached but did not meet significance.
Conclusion:
After pediatric TBI, higher ICP, lower heart rate variability, and the presence of epileptiform discharges, sleep spindle asymmetry and seizures between days 2-7 post-injury are predictive of PTE when adjusting for initial injury severity. Larger studies are needed to investigate whether poor cerebral autoregulation is predictive of PTE when adjusting for initial injury severity.
Funding:
:This work was supported by funding from the United States Department of Defense Congressionally Directed Medical Research Program - Epilepsy Research Program (W81XWH-19-1-0514).
FIGURES
Figure 1