# Interictal epileptiform activity in the acute stroke phase: an independent predictor of poor outcome

## TL;DR

This study shows that abnormal brain wave patterns detected early in stroke patients predict worse recovery outcomes, highlighting the importance of EEG monitoring.

## Contribution

The study demonstrates that interictal epileptiform discharges are an independent predictor of poor stroke outcomes.

## Key findings

- 18.2% of stroke patients showed interictal epileptiform discharges within 72 hours of admission.
- The presence of IEDs was an independent predictor of poor functional outcomes (mRS 3-6) at 3 months.
- The logistic regression model achieved 81% accuracy in predicting functional outcomes.

## Abstract

Electroencephalography (EEG) features are emerging as valuable prognostic indicators in acute stroke. However, data on the predictive value of interictal epileptiform discharges (IEDs) remain limited. This study aimed to assess the prognostic role of IEDs in predicting functional outcomes in stroke patients without symptomatic seizures who underwent point-of-care EEG within 72 h of admission.

We retrospectively analysed the clinical, neurophysiological and neuroimaging data of acute stroke patients who underwent point-of-care EEG within 72 h of admission. Interictal epileptiform discharges were identified according to the International Federation of Clinical Neurophysiology criteria. A multivariate logistic regression model identified variables associated with modified Rankin scale (mRS) scores of 3-6 at 3 months.

Among 593 stroke patients (median age 77 years, range 22-98; median National Institutes of Health Stroke Scale [NIHSS] 5, range 0-25), 18.2% exhibited IEDs on EEG within 72 h of admission. At 3-month follow-up, 223 patients (37.6%) demonstrated poor functional outcome (mRS 3-6). The presence of IEDs on EEG (odds ratio [OR] = 1.088, P = .037), along with age (OR = 1.004, P < .001), NIHSS at admission (OR = 1.032, P < .001), premorbid disability (OR = 1.111, P < .001), hemorrhagic stroke (OR = 1.120, P < .001) and lesion extent (OR = 1.070, P < .001), was an independent predictor of poor clinical outcomes at 3 months (mRS 3-6). The logistic regression model, including these factors, achieved 81% accuracy in predicting functional outcomes.

Early IEDs on EEG within 72 h are independent predictors of poor clinical outcomes (mRS 3-6) at 3 months. These findings underscore the importance of EEG monitoring in the acute phase of stroke and suggest that IED detection may serve as an additional prognostic marker.

Graphical abstract