Antiseizure Medications in Post-stroke Seizures: A Systematic Review and Network Meta-analysis
Shubham Misra1, Jesse Dawson2, Patrick Kwan3, Scott E. Kasner4, James C. Grotta5, David Larsson6, Tomotaka Tanaka7, Hitten Zaveri1, Selena Wang1, Erum I. Khan8, Melissa Funaro1, Sravan Perla1, Priya Dev9, Taimoor Hussain1, David S. Liebeskind10, Clarissa Yasuda11, Amr Elshahat1, Gazala Hitawala1, Ethan Wang1, Rachel Kitagawa1, Abhishek Pathak9, Yize Zhao1, Fabien Scalzo12, Masafumi Ihara7, Johan Zelano6, Katharina Sunnerhagen13, Matthew Walters2, Nathalie Jette14, Terry Quinn2, Nishant Mishra1
1Yale University, 2University of Glasgow, 3Monash University, 4The University of Pennsylvania, 5UTHealth Houston, 6Sahlgrenska University Hospital, 7National Cerebral and Cardiovascular Center, 8The University of Alabama, 9Banaras Hindu University, 10University of California Los Angeles, 11University of Campinas, 12Pepperdine University, 13University of Gothenburg, 14University of Calgary
Objective:
To determine outcomes associated with antiseizure medication (ASM) in people with post-stroke seizure (PSS).
Background:
It is unclear which ASMs are most effective in treating PSS.
Design/Methods:
We systematically searched electronic databases for studies in PSS patients on ASM. Our outcomes were mortality, adverse events, seizure recurrence, and ASM discontinuation. We assessed risk-of-bias using ROB-2 and ROBINS-I. Using levetiracetam as the common comparator, we conducted frequentist network meta-analyses and validated our results using a Bayesian approach. We report pooled odds ratio (OR) and 95% confidence intervals (CI) and ranked treatment benefits using p-scores. (PROSPERO CRD42022363844)
Results:
Our search yielded 15 studies (3 randomized, 12 non-randomized, N=18676, 60% male, mean age 69 years) comparing 13 ASMs. Three studies had moderate, and 12 had high risk-of-bias. Using frequentist approach, compared to levetiracetam, phenytoin (OR 8.3; CI 5.7-11.9), valproic acid (VPA) (4.7; 3.6-6.3), oxcarbazepine (3.3; 1.7-6.5), and carbamazepine (2.1; 1.7-2.7) were associated with greater mortality. Phenytoin (5.2; 1.2-22.9) and carbamazepine (1.9; 1.0-3.5) were associated with greater adverse events. Carbamazepine (1.8; 1.5-2.2), VPA (1.3; 1.0-1.6), and phenytoin (1.9; 1.4-2.8) had greater ASM discontinuation rates. Despite ASM use, 37.4% (23.2-52.8%) (10 studies; N=4783) had a recurrent seizure(s). Carbamazepine (3.0; 1.7-5.3), VPA (4.2; 2.1-8.3), lamotrigine+VPA (5.8; 2.1-16.1), and phenytoin (7.3; 3.7-14.5) were associated with greater seizure recurrence and potential confounding by indication. P-scores indicated lowest mortality with levetiracetam, fewest adverse events and drug discontinuation with lamotrigine, and fewest seizure recurrence with eslicarbazepine. Findings were consistent using the Bayesian approach.
Conclusions:
Levetiracetam and lamotrigine may be preferred ASMs for PSS. Despite ASM use, seizure recurrence rate remains high. Cautious interpretation is necessary due to moderate-to-high bias risk in reported data and potential confounding by indication. An analysis of methodically collected prospective multicentric ASM data in PSS patients is critically needed to determine safety, efficacy, and the mechanisms of drug resistance.