Article Commentary: “Association of the Timing and Type of Acute Symptomatic Seizures With Poststroke Epilepsy and Mortality”

 Massive incompetence shown here!

Let's see how long you've known of the problem and HAVEN'T SOLVED IT!

We've known of this problem a long time. Provide solutions!

• 10% seizures post stroke (19 posts to April 2017)

• 5% epileptic seizures after stroke (10 posts to April 2021)

• epileptic seizures (6 posts to December 2015)

• post-stroke epilepsy (17 posts to December 2016)

Article Commentary: “Association of the Timing and Type of Acute Symptomatic Seizures With Poststroke Epilepsy and Mortality”

Schubert KM, Zieglgänsberger D, Bicciato G, Abraira L, Santamarina E, Álvarez-Sabín J, Ferreira-Atuesta C, Katan M, Sinka L, Terziev R, et al. Association of the Timing and Type of Acute Symptomatic Seizures With Poststroke Epilepsy and Mortality. Stroke. 2025;56:1748–1757.

Poststroke epilepsy (PSE) represents the most common cause of new-onset epilepsy in older adults.1 Recent research has shown it to be associated with increased morbidity and mortality in patients with stroke, prompting interest in the identification of risk factors and the development of prediction tools for PSE.1,2

(WOW! Prediction; NOT PREVENTION!  I'd have you all fired for cause!)

Acute symptomatic seizures (ASyS), or seizures that occur within 7 days of acute stroke, have been identified as a prominent risk factor for PSE and are included in the SeLECT score, which estimates risk of seizures in the 12 months following acute ischemic stroke.3 In their recently published cohort study, Schubert et al. take the next step by asking whether specific features of ASyS more strongly predict PSE and mortality.

The study population included patients with imaging-confirmed acute ischemic stroke at nine different international centers and was derived from the registry used for the creation of the SeLECT score (n=4552). Investigators focused this study on those patients with ASyS, which represented 5% of the total study population (n=233). The primary outcome measures were the development of PSE, defined by the occurrence of remote symptomatic seizures (RSyS), and mortality.

To answer their research question, Schubert et al. characterized ASyS by seizure type and time from stroke onset and performed statistical analyses a multivariable Cox proportional hazard regression to ascertain hazard ratios for development of PSE and mortality, adjusting for covariates such as demographic and stroke features.  

Timing of ASyS in relation to stroke was defined as days between stroke onset and seizure; as 55% (n=127) were found to have seizures on day 0, timing was dichotomized into day 0 and day 1-7. They found that patients with ASyS on day 0 have an adjusted hazard ratio of 2.3 (95% CI 1.3-4.0, p=0.003) for PSE as compared to those with ASyS on days 1-7.

Types of ASyS were subdivided first by presence of status epilepticus. Those without status epilepticus were characterized as having “short ASyS” and further subdivided by initial seizure into focal aware, focal unaware, and focal to bilateral tonic-clonic (FBTCS). Of these four types, status epilepticus and FBTCS were significantly associated with development of PSE with higher risk for status epilepticus (aHR of 9.6, p<0.001) than FBTCS (aHR of 3.4, p<0.001). Presence of status epilepticus or occurrence of FBTCS on day 0 were also associated with increase in mortality.

In addition to seizure type and timing, Schubert et al. also looked at the role of a handful of other variables associated with the development of PSE within the ASyS population, including age, male sex, stroke severity, location, and cause. Cortical involvement and large artery atherosclerosis as cause of stroke were each statistically associated with increased risk of PSE with hazard ratios of 2.2 and 1.6, respectively.

Finally, the investigators adjusted the current SeLECT score to incorporate these findings, creating the SeLECT-ASyS score to predict risk of recurrent seizure, or RSyS, which would qualify as diagnosis of PSE. Validation of the score was performed using the same three cohorts used for the initial SeLECT score, and results demonstrated its superior discrimination as compared to the existing SeLECT model in predicting PSE. The updated SeLECT-ASyS score has already been integrated into the SeLECT mobile application.

To summarize, this study by Schubert et al. demonstrates the importance of ASyS timing and type to the development of PSE and mortality. In addition, it offers a convenient tool to assist in real-time clinical assessment for this patient population. While assessing risk for PSE is desirable for clinicians, many questions remain about what to do with this knowledge of risk. Should those patients with ASyS at higher risk for PSE be treated empirically with antiseizure medications? If so, how long should they be continued? These are the natural next questions as our knowledge of PSE continues to expand.