Abstracts

Rationale: Post-Stroke Epilepsy (PSE) is the most common cause of new-onset seizures in adults. PSE is associated with increased mortality and worsened functional outcomes in the civilian population. The prevalence and outcomes of PSE in the U.S. Veteran population have not previously been published and may differ from the civilian population due to unique exposures and risk factors. The aim of this study is to deeply characterize U.S. Veterans with PSE who are admitted to the VA Epilepsy Monitoring Units (EMUs) to 1) assess survival after EMU discharge in patients with epilepsy compared to PSE, and 2) compare demographics that may influence outcomes in these groups.

Methods: This is a 2-center retrospective longitudinal cohort study. The inclusion criterion is admission to the Portland or Houston VA EMU for video EEG monitoring between 01/01/2012 and 12/31/2014. Data was collected from the qualifying admission until 5/1/2024. Key variables collected included stroke date, EMU discharge diagnosis and comorbidities, anti-seizure medications (ASMs) prescribed at the time of EMU admission, and date of death or last VA visit. Survival was compared by Kaplan-Meier curve and a Cox proportional hazards model, adjusting for sex, age at admission, and the Elixhauser Comorbidity Index (ECI) calculated at the time of admission [Figure 1].

Results: Of 564 qualifying veterans, 61 (10.8%) had a stroke prior to their EMU evaluation. The risk of death was 66% higher if the patient had a stroke prior to EMU admission, after adjusting for age and ECI (OR 1.66, p=0.010, 95% CI: 1.13-2.44). In comparing the EMU diagnoses of the patients with history of prior stroke vs those without stroke, epilepsy was diagnosed in 20 (32.8%) versus 88 (17.5%), PNES in 9 (14.8%) versus 181 (36%), mixed diagnoses in 1 (1.6%) versus 4 (0.8%), other diagnoses in 9 (14.8%) versus 54 (10.7%), and non-diagnostic admissions for 22 (36%) versus 176 (35%), respectively. Patients with PSE had a higher mean ASM count at the time of EMU admission (1.5 for PSE vs 0.9 for all others), however this was not statistically significant [Figure 1]. When restricting the analysis to patients diagnosed with epilepsy, the risk of death was 2.8 times higher in patients with PSE than non-PSE epilepsy (p=0.007, 95% CI: 1.3 - 6.0) [Figure 2].

Conclusions: In this study, Veterans with PSE experienced significantly higher rates of mortality than veterans with non-stroke related epilepsy, showing a nearly 3-fold increase in death rate even after adjusting for sex, age, and comorbidity index. This difference in mortality was less pronounced in the overall EMU population, where a history of stroke only increased the adjusted mortality rate by 66%. There was no statistically significant association between PSE and the number of ASMs prescribed, and while PSE was associated with higher ECI scores, this lost statistical significance when adjusted for age.

Funding: n/a