>electroencephalographic Biomarkers of Post-stroke Seizures: A Systematic Review and Meta-analysis
Abstract number :
1.313
Submission category :
4. Clinical Epilepsy / 4B. Clinical Diagnosis
Year :
2024
Submission ID :
1114
Source :
www.aesnet.org
Presentation date :
12/7/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Vaibhav Goswami, MD – Drexel University College of Medicine & Tower Health
Shubham Misra, PhD – Yale University
Kapil Gururangan, MD – Northwestern University
Anuradha Singh, MD – Icahn School of Medicine at Mount Sinai
L. Brian Hickman, MD, MSc – University of California, Los Angeles
Alexandra Short, MSLS, AHIP – Tower Health
Scott Woolf, DO – Garnet Health
Haneef Zulfi, M.B.B.S, MD – Baylor College of Medicine
Karen Medin, DO – Veteran Affairs Medical Center
Hamada Altalib, DO, MPH – Yale University School of Medicine
Ece Eldem, BS – Yale University
John-Paul Nicolo, MBBS (Hons), BMedSci, FRACP – Monash University
Carla Bentes, MD PhD – Department of Neurosciences and Mental Health (Neurology), Hospital de Santa Maria-CHULN. Centro de Estudos Egas Moniz, Faculdade de Medicina, Universidade de Lisboa, Lisboa, Portugal
Marissa Kellogg, MD – Oregon Health & Science University (OHSU)
Laure-Peter Derex, MD, PhD – Hospices Civils de Lyon (Centre Hospitalier Universitaire de Lyon),
Felix Benninger, MD PhD – Tel Aviv University
Jorge Burneo, MD, MSPH, FAAN, FAES, FRCPC – Western University, London Ontario Canada
David S Liebeskind, MD – UCLA
Paul Vespa, MD – UCLA
Lawrence Hirsch, MD – Yale University School of Medicine
Nishant Mishra, MD, PhD – Yale University & West Haven VA HCS
Rationale: Post-stroke seizure (PSS), defined as early plus late seizures, is a well-known complication of stroke and is associated with greater mortality risk, poor functional outcomes, and cognitive decline. These patients have seizure recurrence despite the use of antiseizure medications, thus adversely impacting their quality of life. Identifying patients at risk is critical, enabling earlier and more targeted interventions. We systematically evaluated the evidence on the predictive value of electroencephalogram (EEG) patterns associated with PSS risk and poststroke epilepsy (PSE) development.
Methods: We conducted a comprehensive literature search of electronic databases until 31st December 2023. We focused on studies that analyzed EEG patterns in stroke patients, both ischemic and hemorrhagic, aged 18 and older, evaluating their association with PSS or PSE risk. One reviewer (KG) reviewed terminology for coded EEG patterns and categorized prior terminology to current, standardized nomenclature; only standardized nomenclature was used for the analyses. Our outcomes were the pooled proportion of first EEG patterns and the association of EEG patterns with seizure recurrence, PSE, and PSS risk. We used the Quality in Prognostic Studies tool to assess the risk of bias. We tested the association of EEG features with PSE/PSS risk and report odds ratio (OR) and 95% confidence interval (CI). We used R for statistical analyses.
Results: We included 38 studies comprising 6522 ischemic stroke and 574 hemorrhagic stroke patients. 2,652 patients had PSS, and 1,443 developed PSE. Risk of bias was low in three studies (8%), moderate in 12 (32%), and high in 23 studies (61%). Ten studies reported continuous (c) EEG monitoring for 4 to 48 hours. The most frequent patterns observed were focal or lateralized slowing (52%), generalized or diffuse slowing (22%), and sporadic epileptiform discharge (27%). Sporadic epileptiform discharges significantly predicted PSS (OR 3.3; CI 2.4-8.0) and PSE (OR 7.5; CI 3 -18). In stroke patients undergoing cEEG, epileptiform abnormalities (OR 6.2; CI 2.5-15.5), seizure (OR 9.1; CI 2.1-39.6), and regional attenuation without delta (OR 7.5; CI 1.8-31.1) were significantly associated with PSE. Lateralized periodic discharges were associated with early seizures in PSS patients (OR 27; CI: 6 -115), but it failed to reach statistical significance in the PSE population, i.e, stroke survivors with late seizures (OR 4.7; CI 0.5 -42.7).
Conclusions: Sporadic epileptiform discharges are associated with a significantly higher risk of PSS and PSE. EEG patterns are viable biomarkers for predicting PSE. A pragmatic randomized controlled trial evaluating outcomes in patients assigned to early cEEG compared to standard care is currently warranted. Using standardized nomenclature, a unified semantic description of the EEGs allows for systematically conducting such centrally adjudicated multicentric study.
Funding: None to disclose
Clinical Epilepsy