Abstracts

Outpatient EEG Epileptiform Abnormalities as Predictive Biomarkers of Epileptogenesis Following Acute Brain Injury

Abstract number : 1.22
Submission category : 3. Neurophysiology / 3B. ICU EEG
Year : 2025
Submission ID : 1074
Source : www.aesnet.org
Presentation date : 12/6/2025 12:00:00 AM
Published date :

Authors :
Presenting Author: Jackson Narrett, MD – Yale New Haven Hospital

MarieElena Byrnes, DO, MS – Cleveland Clinic
Neishay Ayub, MD – Brown University
Clio Rubinos, MD, MS, FACNS – University of North Carolina, Chapel Hill, North Carolina, USA.
Nishant Mishra, MD, PhD – Yale University & West Haven VA HCS
Jennifer Kim, MD, PhD – Yale School of Medicine
Emily Gilmore, MD – Yale School of Medicine
Lawrence Hirsch, MD – Yale University School of Medicine
Sahar Zafar, MD, MBBS – Massachusetts General Hospital
Vineet Punia, MD – Cleveland Clinic, Cleveland, OH, USA
Adithya Sivaraju, MD – Yale New Haven Hospital

Rationale: Epileptiform abnormalities reflect core epileptogenic mechanisms and are a well-established predictor of seizure recurrence, doubling the risk after a first unprovoked seizure.1-3 While acute EEG abnormalities following brain injury have been linked to late seizures, the prognostic significance of post-acute EEG findings remains limited.4,5 This study evaluates whether epileptiform abnormalities on outpatient EEGs—performed as part of routine care—parallel electrophysiological processes of epileptogenesis and can serve as a biomarker for seizure recurrence.

Methods: We conducted a two-center retrospective cohort study at Yale New Haven Hospital and Cleveland Clinic, including patients with acute symptomatic seizures (clinical or electrographic) and/or epileptiform abnormalities on EEG during index hospitalization. All were discharged on anti-seizure medications (ASMs) and had at least one outpatient EEG and follow-up data. EEGs were classified per ACNS criteria. Cox proportional hazards models assessed time to late seizures, using outpatient EEG findings as the key covariate. Analyses were conducted in R (v4.3.1), with significance set at p < 0.05.
Neurophysiology