Abstracts

Rationale: Most reports of seizure aggravation by antiepileptic drugs (AEDs) are anecdotal, uncontrolled, not quantitative and involve only small numbers of patients. Randomized, placebo-controlled clinical trials provide an opportunity to examine the possibility of seizure aggravation in a quantitative manner using similar methodology to traditional AED efficacy trials but measuring seizure increase rather than seizure reduction. The mirror image of the standard measure of drug responsiveness (50% or greater reduction in seizures compared to baseline) is a 100% or greater increase in seizures, as both represent a 2-fold change. Seizure aggravation by AEDs is believed to be more common in generalized epilepsies. Two clinical trials have recently shown efficacy for levetiracetam (LEV) against primary generalized tonic-clonic (PGTC) and myoclonic seizures, respectively, in idiopathic generalized epilepsy (IGE). Methods: Data from two randomized, double-blind, placebo-controlled, parallel-group studies (Berkovic et al, Neurology, 2007; 69:1751-60. Noachtar et al, Neurology, 2008; 70:607-16) in 283 patients with refractory IGE receiving either adjunctive LEV 3000 mg/day (n=139) or placebo (n=144) were pooled and analysed for evidence for or against seizure aggravation by LEV. The proportions of patients experiencing an increase in seizure frequency of any degree and of >100% during treatment (4-week up-titration period, followed by a 12- or 20-week evaluation period) compared to an 8-week baseline were calculated. As in the efficacy analyses of each individual study, the analyses were performed on the intention-to-treat population, and the seizure frequency measures were myoclonic seizure days per week, PGTC seizures per week and seizure days (any type) per week. Results: Seizure increases were more common with placebo than LEV for all severities and for PGTC, myoclonic and all seizures. The proportion of patients experiencing any increase in myoclonic seizure days was 18.8% for placebo and 5.8% for LEV, while PGTC seizure frequency increased in 18.1% of patients taking placebo and 9.4% of patients taking LEV; for all seizures, the corresponding proportions with increases in seizure days were 29.9% and 10.8%, respectively. A greater than doubling of myoclonic seizure days occurred in 3.5% with placebo and 0.7% with LEV, while a greater than doubling in PGTC seizure frequency occurred in 3.5% with placebo and 2.2% with LEV; for all seizures, a greater than doubling occurred in 5.6% with placebo and 2.2% with LEV. The difference between placebo and LEV was statistically significant for any increase in myoclonic seizure days (p=0.0016), PGTC seizure frequency (p=0.0366) or all seizure days (p=0.0001). The difference between rates of greater than 100% increase did not reach statistical significance but patient numbers in this group were small. Conclusions: There is no evidence that adjunctive treatment of IGE with LEV produces aggravation of myoclonic or PGTC seizures.