Poststroke Seizures: New Study Suggests First-line Treatment Options

Poststroke seizures are associated with increased mortality risk and worse cognitive outcomes. Approximately 8 to 10% of people who’ve had a stroke experience unprovoked late seizures, defined as seizures seven days after the stroke. The cerebrovascular disorder, however, accounts for approximately 50% of seizures in those over age 65.

Antiseizure medications (ASMs) are commonly used to prevent seizures in poststroke seizure patients, but there is significant variability in the choice of antiseizure medications. Nishant K. Mishra, MD, PhD, assistant professor of neurology at Yale School of Medicine and stroke director at the West Haven VA Medical Center, led a multi-institutional team to evaluate different ASMs prescribed to treat poststroke seizures and the associated outcomes. The network meta-analysis was recently published online in Neurology.

“We wanted to identify the most effective anti-seizure mediations for the poststroke seizure populations,” said Mishra.“While a few studies have been published that look at mortality in this population, we wanted to look at additional outcome measures that are clinically meaningful—for example, adverse events, seizure recurrence rate, and drug discontinuation rates.”

The results from this analysis could guide physicians when selecting an ASM for patients who experience a seizure after a stroke. Preventing seizures is also a primary goal in epilepsy management.

The investigation systematically evaluated the prospective and retrospective data from articles published in the scientific literature that evaluated different ASMs to manage poststroke seizures in adults with ischemic stroke (caused by a blocked blood vessel) or intracerebral hemorrhage, also known as a hemorrhagic stroke (due to bleeding inside the brain caused by a ruptured blood vessel).

The researchers identified 15 studies published between 2007 and 2023 that included data on more than 18,000 patients (13,067 who had experienced an ischemic stroke and 5,567 who had a hemorrhagic stroke). The data includes 13 ASMs (levetiracetam, carbamazepine, phenytoin, valproic acid, lamotrigine, lacosamide, phenobarbital, oxcarbazepine, zonisamide, eslicarbazepine, brivaracetam, perampanel, and topiramate). The researchers conducted a network meta-analysis to determine the effect size for each outcome measure and used the GRADE methodology to evaluate the outcome of these medications.

The research analysis showed that levetiracetam and lamotrigine were effective at preventing additional seizures with fewer drug interactions. These patients also had fewer adverse events, such as allergic reactions, and lower mortality rates.

The analysis did not link a single ASM to patient outcomes, nor did the research point to a specific dose of medication that provided positive or negative outcomes or the ability of patients to tolerate it—mainly because these data were not reported in the studies included in this network meta-analysis. Mishra also noted that newer epilepsy drugs are available but could not be investigated in this meta-analysis because relevant data haven’t been published in the poststroke population.

The results of this analysis highlight the importance of standardizing data collection and reporting to establish reliable and comparable studies. To promote this effort, Mishra founded the International Post Stroke Epilepsy Research Consortium (IPSERC). The consortium aims to increase progress in developing neurotherapeutics and interventions to prevent poststroke seizures and epilepsy.

“For patients, in general, we need to highlight that poststroke seizure is a problem that, I believe, is being underestimated because certain seizures, such as behavioral seizures that can manifest as a lapse of awareness or cognitive impairment, may go undetected,” said Mishra. “Patients, family, and general clinicians should be aware of these characteristics.”

Mishra noted that the team could not measure confounding factors, such as design and follow-up duration variation, and stratify treatment comparisons by seizure and stroke subtypes, which could have influenced the results. They also did not stratify stroke type based on seizure occurrence.