Abstracts

Rationale: Post-stroke epilepsy (PSE) is a major source of preventable epilepsy burden. Several stroke-related clinical and neuroimaging features are recognized as risk factors for PSE development. However, the role of early electroencephalogram (EEG) findings in predicting PSE development remains poorly explored. The primary aim of our study was to test the hypothesis that acute epileptiform activity after stroke is associated with risk of PSE development.

Methods: After IRB approval, our prospectively maintained stroke and EEG database were used to identify adults (≥18 years) with acute ischemic stroke who underwent continuous EEG (cEEG) within 7 days of last-known well time. Patients with epilepsy prior to acute stroke were excluded at this stage. Patients with seizures after hospital discharge (PSE; cases) were matched by age at the time of acute stroke (±5 years) and duration of follow-up (±3 months) with controls (no post-hospital discharge seizure) in a 1:2 ratio. Subsequently, the acute clinical, neuroimaging, and EEG data for the matched cases and controls was extracted from the EMR. Univariate logistic regression with a random effect accounting for the matching and multivariable analysis, with variables in a conditional logistic regression model, and backwards selection was performed. To avoid multicollinearity in the multivariable analysis, cEEG with electrographic seizure and interictal epileptiform activity (IEA) were collectively classified as having acute epileptiform abnormalities (EAs).

Results: A total of 36 cases and 72 controls were included in the study. Cases and controls were well matched by the age at the time of acute stroke (66 years, p = 0.96) and duration of follow-up (33 months; p = 0.95). Table 1 compares the distribution of various clinical, neuroimaging, EEG variables between the cases and controls. On the univariate analysis, the cases at the time of acute stroke were significantly less likely to have hypertension (OR = 0.33, 95%CI = 0.12 – 0.09), smoking (OR = 0.39, 95%CI = 0.14 – 0.94), and more likely to have a hemorrhagic conversion of the stroke (OR = 3.92, 95%CI = 1.43 – 11.3). Cases were more likely to have convulsive seizure (OR = 7.28, 95%CI = 1.96 – 36.6) prior to cEEG start. The days of cEEG monitoring was longer in cases (OR = 1.59, 95%CI = 1.2 – 2.11). Electrographic seizures were only noted in cases (n = 7; 22%). IEAs on cEEG were significantly more frequently noted in cases compared to controls (Table 1). After adjusting for covariates, the multivariable analysis which included EAs, hypertension, current smoker, hemorrhagic conversion and early convulsive seizures found that only acute EAs on cEEG were at significantly higher odds in cases compared to controls (OR = 11.9, 95%CI = 1.75 – 491.6, p < 0.004).