Time Is Brain: Using Point-of-Care Electroencephalography for Rapid Diagnosis of Non-Convulsive Seizures Presenting as Stroke Mimic at a Community Hospital
Abstract number :
2.017
Submission category :
3. Neurophysiology / 3B. ICU EEG
Year :
2023
Submission ID :
928
Source :
www.aesnet.org
Presentation date :
12/3/2023 12:00:00 AM
Published date :
Authors :
Presenting Author: Richard Kozak, MD – Providence Mission Medical Center
Kapil Gururangan, MD – Epilepsy Fellow, Department of Neurology, David Geffen School of Medicine at UCLA; Matthew Kaplan, MD – Medical Director US Acute Care Solutions, Department of Emergency Medicine, Providence Mission Medical Center; Parshaw Dorriz, MD – Adjunct Assistant Professor of Clinical Neurology, Department of Neurology, Providence Mission Medical Center
Rationale: Epileptic seizures may present as a mimic of acute stroke (e.g., focal subclinical status epilepticus causing expressive aphasia), however differentiating these entities acutely during a stroke code can be difficult. Prior research (Lucas et al. Am J Neuroradiol 2021;42:49 and Van Cauwenberge et al. Neurology 2018;91:e1918) has demonstrated low sensitivity of perfusion imaging findings of hyperperfusion crossing vascular territories for epileptic seizures. While EEG would be the gold standard test for detecting seizures with high sensitivity, rapid EEG acquisition with conventional EEG systems is not realistic at most hospitals. Novel point-of-care EEG (pocEEG) devices could expedite the acquisition of EEG data and serve as adjuncts to acute stroke imaging for ruling in or ruling out seizures in patients with acute focal neurological deficits.
Methods: Patients who underwent pocEEG monitoring in the wake of a stroke code in our community hospital were retrospectively identified within a one-year period. Both stroke code activation and pocEEG monitoring were conducted according to standard of care to evaluate clinical suspicions of acute stroke or epileptic seizure, respectively. We classified pocEEG findings as seizure or status epilepticus, highly epileptiform patterns (HEP), slowing, or normal activity to describe the prevalence of EEG findings in this context.
Results: Seventy patients underwent both pocEEG monitoring and a stroke code. Acute stroke was confirmed in 34 cases (ischemic in 28, hemorrhagic in 6), among which pocEEG detected status epilepticus in 2, HEP in 4, slowing in 22, and normal activity in 6. Stroke mimics were diagnosed in 36 cases (final diagnosis based on clinical presentation or diagnostic testing were seizure in 15, toxic-metabolic encephalopathy in 9, transient ischemic attack in 4, hypertensive encephalopathy in 4, subdural hematoma in 1, metastatic cancer in 1, transient global amnesia in 1, and shock in 1), among which pocEEG detected electrographic seizures in 2, HEP in 9, slowing in 16, and normal activity in 9. Of the 15 patients diagnosed with seizures as a stroke mimic (most often due to a preceding or subsequent clinical seizure), the two patients with confirmed electrographic seizures presented with persistent expressive aphasia, and the remainder had pocEEG findings of HEP in 3, slowing in 9, and normal activity in 1.
Conclusions: The differential diagnosis for patients undergoing acute stroke evaluation often includes epileptic seizures. In our community hospital, pocEEG devices enabled rapid diagnosis of non-convulsive seizures as either stroke mimics or complications of acute stroke, as well as rapid exclusion of ongoing seizures in the majority of cases. Such devices open the possibility of EEG as a valuable adjunctive diagnostic tool during acute stroke evaluations.
Funding: Study funded by Ceribell, Inc.
Neurophysiology