Following ASM Discontinuation, Age at Epilepsy Diagnosis and Abnormal EEG Predict Future Pediatric Seizure Relapse
The median time to seizure relapse following antiseizure medication discontinuation was 8 months, although 16.7% of the cohort had relapse occur just 1 month after baseline.
Mirac Yildirim, MD
Findings from a study of seizure-free children who discontinued their antiseizure medication (ASM) indicated that being an adolescent age at diagnosis, having abnormal EEG findings following withdrawal, and having a high number of seizures while on ASMs were associated with a higher risk of seizure relapse. Additionally, advanced age at first seizure, presence of intellectual disability, and short ASM tapering were also associated with increased seizure relapse in only univariate logistic regression analyses.1
Among a cohort of 269 children who were followed up for a median of 46 months (range, 18-126; IQR, 29-61), the overall frequency of seizure relapse was 33.5%, with approximately 75% of relapses occurring within 2 years of withdrawal. In addition to the 6 identified predictive factors significantly related to relapse, data also suggested that abnormal MRI findings, such as malformations of cortical development and hydrocephalus, might be potential biomarkers for an increased risk. Those with genetic epilepsies were also shown to be at increased risk as well.
The study, conducted by lead investigator Mirac Yildirim, MD, department of pediatric neurology, Ankara University, and colleagues retrospectively evaluated demographic and clinical characteristics of these children, the rate of seizure relapse following ASM withdrawal, predictors of relapse, and clinical course after seizure relapse. A total of 195 (72.5%) children had idiopathic epilepsy, 67 (24.9%) had symptomatic/structural epilepsy, and 7 (2.6%) had genetic epilepsy. During the treatment of ASM, most of the children (87%; n = 234) were exposed to monotherapy, 30 (11.2%) to duotherapy, and 5 (1.9%) to polytherapy.
Across the entire cohort, the median time to seizure relapse was 8 months (range, 0.23-117; IQR, 2-25) after ASM withdrawal. Seizure relapse occurred in 16.7% of the 90 children at 1 month, 35.6% at 3 months, 45.6% at 6 months, 62.2% at 1 year, 74.4% at 2 years, 87.8% at 3 years, and 94.4% at 5 years. The remaining 5.6% had relapse after 5 years.
The univariate logistic regression analyses identified 6 factors significantly associated with relapse: age at first seizure, age at diagnosis of epilepsy, presence of intellectual disability, EEG findings after ASM withdrawal, the number of seizures on ASM, and ASM tapering time. Despite this only age at diagnosis of epilepsy (odds ratio [OR], 1.873), EEG after ASM withdrawal (OR, 3.797), and number of seizures on ASM (OR, 1.637) showed statistical significance (P <.05) on multivariate logistic regression analyses.
Brain MRI, performed in each patient, was shown to be abnormal in 23.8% (n = 64) of the cohort. Of the various abnormal MRI findings, encephalomalacia (n = 25), periventricular leukomalacia (n = 14), and cerebral and/or cerebellar atrophy (n = 13), were the most common, among others. Seizure relapse occurred in 40% (n = 10) of children with encephalomalacia, 28.6% (n = 4) of those with periventricular leukomalacia, 38.5% (n = 5) of children with cerebral and/or cerebellar atrophy, 57.1% (n = 4) of children with hydrocephalus, 16.7% (n = 1) of those with posttraumatic brain injury findings, and 66.7% (n = 4) of children with malformations of cortical development.
Those with self-limiting epileptic syndromes such as childhood absence epilepsy (n = 3), benign epilepsy with centrotemporal spikes (n = 3), and Panayiotopoulos syndrome (n = 1), showed favorable outcomes, with only 1 child had a seizure relapse. However, there were 3 children with juvenile myoclonic epilepsy (JME) in the study, and 2 (67%) of them had a seizure relapse. The remaining child with JME had no seizure relapse during the 3-year follow-up after ASM withdrawal.
