Anti-seizure medications and DRESS in paediatric patients: A FAERS disproportionality and time-to-onset analysis

Anti-seizure medications (ASMs) are in the list of the most common drug culprits of drug reaction with eosinophilia and systemic symptoms (DRESS). However, no study has systematically evaluated DRESS associated with ASMs in a large paediatric cohort. This study aimed to investigate association of DRESS with ASMs and to characterize the onset pattern of ASM-related DRESS in paediatrics using FDA Adverse Event Report System (FAERS) database.

FAERS data from the first quarter of 2004 to the second quarter of 2025 were analysed using R software (version 4.5.1). For signal detection of ASM-related DRESS in paediatric patients, disproportionality analysis was performed with four methods, consisting of Reporting Odds Ratio, Proportional Reporting Ratio, Bayesian Confidence Propagation Neural Network and Multi-Item Gamma Poisson Shrinker. Cumulative incidence was estimated using Kaplan-Meier method, and time-to-onset (TTO) analysis was conducted with Weibull distribution test.

Nine hundred forty-one reports of ASMs-related DRESS in paediatric population from FAERS database were analysed. Positive signals for DRESS were identified for lamotrigine, carbamazepine, levetiracetam, phenytoin, valproic acid, oxcarbazepine, phenobarbital, zonisamide, clobazam, ethosuximide, perampanel, fosphenytoin and eslicarbazepine. Ninety-eight cases were included in the TTO analysis. Lamotrigine, carbamazepine, levetiracetam and zonisamide demonstrated a wear-out failure-type pattern. Phenytoin, valproic acid and oxcarbazepine presented a random failure-type profile.

Significant signals of DRESS associated with 13 ASMs were identified in paediatric populations. Notably, signals were detected for the first time for ethosuximide, perampanel, fosphenytoin and eslicarbazepine by disproportionality analysis. TTO analysis demonstrated wear-out or random failure-type patterns in the occurrence of ASM-associated DRESS.