Abstracts

Point-of-Care Electroencephalography Impacts Management of Non-Convulsive Seizures in the Intensive Care Unit at a Community Hospital

Abstract number : 3.122
Submission category : 3. Neurophysiology / 3B. ICU EEG
Year : 2023
Submission ID : 1165
Source : www.aesnet.org
Presentation date : 12/4/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: Parshaw Dorriz, MD – Providence Mission Medical Center

Kapil Gururangan, MD – Department of Neurology – David Geffen School of Medicine at UCLA; Matthew Kaplan, MD – Department of Emergency Medicine – Providence Mission Medical Center; Richard Kozak, MD – Department of Emergency Medicine – Providence Mission Medical Center

Rationale: Critically ill patients are at increased risk of non-convulsive seizures (NCS), but management is hampered at many hospitals due to limited EEG accessibility. Point-of-care EEG (pocEEG) could address this barrier. We aim to describe our institutional experience of pocEEG in the management of critically ill patients with suspected NCS.

Methods: We retrospectively identified patients who underwent pocEEG monitoring in our community hospital ICU. We extracted pocEEG findings (categorized as seizure, highly epileptiform patterns [HEP], slowing, and normal activity; excluding repeat studies for the same patient), and anti-seizure medication (ASM) treatment information (pre-pocEEG, post-pocEEG) to determine whether pocEEG led to appropriate ASM escalation for patients with NCS or avoidance of inappropriate ASM escalation for patients with slow/normal activity. Statistical significance was assessed using chi-square tests (categorical data) and Kruskal-Wallis tests (continuous data).

Results: Among 92 ICU patients (mean age 62.5±20.7 years, 41% female, median Glasgow coma scale score 12.5 [IQR 3.0-15.0], 81.5% of patients were intubated during hospitalization), pocEEG revealed seizures in 6 (6.5%), HEP in 21 (22.8%), slowing in 55 (59.8%), or normal activity in 10 (10.9%). Of the six patients with seizures on pocEEG, three were found to be in non-convulsive status epilepticus. The primary indications for pocEEG monitoring were unexplained encephalopathy in 64 (69.6%), cardiac arrest in 20 (21.7%), and preceding clinical seizure in 7 (7.6%). pocEEG monitoring occurred for a median duration of 3.4 [IQR 2.0-8.5] hours, and 60.9% were performed after-hours (5p-9a). ASM treatment occurred pre-pocEEG in 63 patients (68.5%), including 100.0% of patients with a preceding clinical seizure, 85.0% of patients with cardiac arrest, and 59.4% of patients with unexplained encephalopathy, and ASM treatment occurred post-pocEEG in 24 patients (26.1%), including 40.7% of patients with seizures/HEP and 20.0% of patients with slow/normal activity. Among patients treated pre-pocEEG, 51 (81.0%) did not have subsequent ASM escalation (86.4% among patients with slow/normal activity vs. 40.0% among patients with seizures, p=0.03). Compared to those with seizures, patients with slow/normal activity were less likely to receive ASMs post-pocEEG (20.0% vs. 40.7%, p=0.04) and received fewer median ASM treatments (1.0 [IQR 1.0-2.0] vs. 3.5 [2.3-4.0], p=0.001). 

Conclusions: Critically ill patients with suspected NCS who underwent pocEEG monitoring at our community hospital experienced significant shifts in ASM treatment patterns such that patients without epileptiform activity were prevented from having escalation of ASM treatment.

Funding: Study funded by Ceribell, Inc.

Neurophysiology