Intracranial Electrocorticography for Post-Traumatic Epilepsy Prediction After Acute Traumatic Brain Injury
Abstract number :
1.006
Submission category :
1. Basic Mechanisms / 1A. Epileptogenesis of acquired epilepsies
Year :
2025
Submission ID :
449
Source :
www.aesnet.org
Presentation date :
12/6/2025 12:00:00 AM
Published date :
Authors :
Presenting Author: Xiaojue Zhou, PhD – University of California San Francisco
Roxanne Simmons, MD – University of California San Francisco
Rajanikant Panda, PhD – University of California San Francisco
Anthony Mefford, MD – University of California San Francisco
Siddharth Narula, MS – University of California San Francisco
Kevin Bao, BA – University of California San Francisco
Adeline Hayman, BA – University of California San Francisco
Narayan Sankaran, PhD – University of San Francisco
David Caldwell, MD, PhD – University of California San Francisco
Matheus Otero, MD – Mount Sinai Health System
Britta Lindquist, MD, PhD – University of California San Francisco
Vishnu Karukonda, BS – Keck School of Medicine of the University of Southern California
Anthony Digiorgio, DO – University of California San Francisco
Phiroz Tarapore, MD – University of California San Francisco
Lawrence Chyall, RN – Zuckerberg San Francisco General Hospital
Edward Chang, MD – University of California, San Francisco
Claude Hemphill, MD, MAS – University of California San Francisco
Geoffrey Manley, MD, PhD – University of California San Francisco
Michael Huang, MD – University of California San Francisco
Edilberto Amorim, MD – Zuckerberg San Francisco General Hospital, University of California San Franscisco
Rationale: Traumatic brain injury (TBI) is one of the most common causes of acquired epilepsy. While the early post-injury period is critical for the development of post-traumatic epilepsy (PTE), identifying neurophysiological biomarkers of epileptogenesis remains a major clinical challenge. Invasive electrocorticography (ECoG) offers a novel opportunity to observe cortical dynamics in the acute phase after injury, providing new insights into epileptogenesis.
Methods: We analyzed a retrospective cohort of patients with acute moderate-to-severe TBI who underwent continuous ECoG monitoring using 4–8 electrodes. ECoG was reviewed for the presence of seizures by two epileptologists. We defined PTE as unprovoked seizures occurring more than 7 days post-injury during in-hospital stay or outpatient phone follow up. We pursued a preliminary analysis on a subset of patients (n = 9; 4 with PTE and 5 without) and evaluated ECoG spectral and connectivity-based features. Multivariable logistic regression models were adjusted for ECoG feature time from the time of TBI.
Results: Among the 162 patients, the average age was 50.7 years, and 30.2% were female. The majority sustained injuries from falls (44%) or motor vehicle accidents (32%), with 62% demonstrating bihemispheric brain injury. Strip ECoG was pursued for 69.1% of patients and 49.4% had depth electrodes. The ECoG recordings lasted an average of 36.4 hours, with monitoring initiated approximately 36 hours post-injury and ending around 137 hours post-injury. Acute seizures were observed in 56 (35%) patients and 34 (21%) developed PTE. Acute seizures observed within 7 days did not predict PTE. Preliminary analysis of the ECoG features showed that spectral kurtosis (p = 0.02), theta-band synergy (p = 0.02), and burst suppression amplitude ratio (p = 0.04) were associated with PTE.
Conclusions: Continuous intracranial ECoG monitoring in TBI patients identified acute seizures in over a third of patients. While acute seizures alone did not predict long-term PTE, preliminary analysis of quantitative ECoG features indicates that cortical dynamics may support the development of PTE prediction biomarkers.
Funding: This work was supported by the U.S. Department of Defense HT9425-23-1-0242.
Basic Mechanisms