Model Predicts Outcomes After AED Withdrawal

An evidence-based model allows neurologists to predict individual patient outcomes following the withdrawal of antiepileptic drugs (AEDs), according to research published in the July issue of Epilepsia. The model indicates a patient’s risk of relapse and chance of long-term seizure freedom. The model therefore might help physicians and patients make individualized choices about treatment, said the authors.

Patients with epilepsy who have achieved seizure freedom may want to discontinue their AEDs to avoid their associated side effects. Discontinuation raises the risk of seizure recurrence, however. Previous prognostic meta-analyses have been unable to calculate individual outcome predictors’ effect sizes because of the heterogeneous methods and reporting in the literature.

Analyzing Individual Participant Data

To overcome this limitation, Herm J. Lamberink, MD, a doctoral student at University Medical Center Utrecht in the Netherlands, and colleagues conducted a meta-analysis using individual participant data from previous studies. They reviewed PubMed and Embase for articles that reported on patients with epilepsy who were seizure-free and had started withdrawal of AEDs. Eligible articles contained information regarding seizure recurrences during and after withdrawal. The investigators selected 25 candidate predictors based on a systematic review of the predictors of seizure recurrence after AED withdrawal.

Herm J. Lamberink, MD

Dr. Lamberink and colleagues identified 45 studies that included 7,082 patients in all. The meta-analysis included 10 studies with 1,769 patients. The populations included selected and nonselected children and adults. Median follow-up time was 5.3 years. In all, 812 patients (46%) had seizure relapse, which was a higher rate than the average reported in the literature. Approximately 9% of participants for whom data were available had seizures in their last year of follow-up, which suggests that they had not regained lasting seizure control.

Model Had Stable Performance

Epilepsy duration before remission, seizure-free interval before AED withdrawal, age at onset of epilepsy, history of febrile seizures, number of seizures before remission, absence of a self-limiting epilepsy syndrome, developmental delay, and epileptiform abnormality on EEG before withdrawal were independent predictors of seizure recurrence. Epilepsy duration before remission, seizure-free interval before AED withdrawal, number of AEDs before withdrawal, female sex, family history of epilepsy, number of seizures before remission, focal seizures, and epileptiform abnormality on EEG before withdrawal were independent predictors of seizures in the last year of follow-up.

The adjusted concordance statistics for the model were 0.65 for predicting seizure recurrence and 0.71 for predicting long-term seizure freedom. Internal–external cross validation indicated that the model had good and stable performance in all cohorts.

One limitation of the study is that the population included only participants who attempted to withdraw AEDs. In addition, too few cases of epileptic encephalopathy and juvenile myoclonic epilepsy were included in the population to determine whether these disorders predict outcomes after AED withdrawal.

—Erik Greb

Suggested Reading

Lamberink HJ, Otte WM, Geerts AT, et al. Individualised prediction model of seizure recurrence and long-term outcomes after withdrawal of antiepileptic drugs in seizure-free patients: a systematic review and individual participant data meta-analysis. Lancet Neurol. 2017;16(7):523-531.