Hyperperfusion on CT Perfusion Scan Correlates with Lateralizing EEG Findings in Patients Presenting with Seizures
Dalia Koujah1, James Kelbert1, Annalise Bracher1, Rachel Fisher1, Ganesh Murthy2
1University of Arizona College of Medicine- Phoenix, 2Banner University Medical Center- Phoenix
Objective:

This study aims to determine if findings of hyperperfusion seen on CT perfusion scan lateralize to findings on EEG.

Background:
Though seizures are a clinical diagnosis, ancillary tests such as an EEG and more recently CT perfusion (CTP) may help in confirming clinical suspicion. Evaluation of hyperperfusion or hypoperfusion on CTP may be a crucial asset in discriminating between stroke and seizure, especially in patients with limited history on presentation, as 13% of stroke mimics are seizures.
Design/Methods:

This study is a retrospective chart review at Banner University Medical Center-Phoenix. All stroke patients from 2019-2022 were systematically recorded and evaluated for a discharge diagnosis code of "epilepsy", "seizure(s)”. Patients were included if they showed hyperperfusion on CTP and had an EEG during the same hospital stay. Data collected includes demographics, comorbidities, presenting symptoms, time from symptom onset until CTP, comorbidities, CT head results, CTP results, focal EEG findings, time from symptoms to EEG, and time between EEG and CTP.

Results:

Of the 51 patients analyzed, 13 presented with hyperperfusion on CTP. 46% of patients with hyperperfusion had lateralized EEG findings (n=5 diffuse slowing, n=3 epileptiform discharges, n=2 electroclinical seizures) that reflected the side of CT-detected hyperperfusion. The most common presenting symptoms were aphasia, unilateral weakness, facial droop, and altered mental status. The average time before CT scan was 1.60 hours. The average time before EEG was 4.98 hours. The average interval of time between CTP and EEG was 3.38 hours.

Conclusions:

CTP may be helpful in supporting clinical suspicion of seizure in patients presenting as stroke codes. Significant delays in obtaining EEGs may underrepresent the expected concordance in this study. Future prospective studies with EEGs done with less delay from patient's point of care may elicit a clearer association in these diagnostic tests in identifying seizures.

10.1212/WNL.0000000000206543