Abstracts

Nervous system adverse events (AEs) are the most common AEs associated with levetiracetam (LEV, Keppra[reg]) in a controlled trial of adjunctive therapy (up to 60 mg/kg/day) in pediatric patients (age 4 to 16 years old) with refractory partial seizures (PS). A number of analyses were undertaken to better understand the nature of the reported psychiatric events in these patients. The relative risk (RR) of psychiatric / behavioral events (PBEs) between patients treated with LEV and placebo was estimated for the 198 evaluable patients. Patients were categorized for analysis by prior psychiatric or cognitive history, and by response status. The results were then compared to a similar analysis completed for adult patients treated with LEV for refractory PS. Two hundred sixteen patients were randomized; 198 patients provided evaluable data (LEV: 101; placebo: 97).
Although the incidence of PBEs in children with refractory PS who are treated with LEV is 38.6% versus 18.6% in adults, the RR is similar in adults and children, as there is also a higher incidence of PBEs in children treated with placebo as compared to adults (27.8% vs. 10.5%). The modestly elevated risk for PBEs in children is 1.39 (95% CI: 0.93-2.08); while that in adults is 1.77 (95% CI 1.30-2.42). The majority of PBEs in children are in the category of non-psychotic mood/anxiety/behavioral symptoms, RR of 2.03 [95% CI: 1.25-3.30]. Those individual terms for which there was a two-fold or greater RR in LEV treated pediatric patients as compared to placebo were agitation (5.76), nervousness (4.32), and depression (2.88). Hostility also tended to occur with a greater RR (1.92) in pediatric patients randomized to LEV.
There is no greater risk of PBEs in pediatric patients with psychiatric histories (1.05: 95% CI 0.63-1.72) or with cognitive impairment (1.01; 95% CI: 0.56-1.82). The incidence of these events in LEV-treated pediatric patients is similar whether or not they have a prior psychiatric history (39.2% vs. 38.0%, respectively). The RR of emotional lability (2.94 vs. 0.94), hostility (3.43 vs. 1.18), and personality disorder (3.92 vs. 0.63) are higher in pediatric patients without a history (possibly attributable to a low incidence amongst placebo patients).
Upon comparison of the responders (patients who experienced a [ge]50% reduction in weekly seizure frequency) to the non-responders, there does not appear to be a preponderance of psychiatric adverse events in either response category. The psychiatric AE profile in children with refractory PS is similar to that in the currently approved LEV labeling for adults. The relative risk for PBEs is comparable for adults and children. There is no greater risk of PBEs in pediatric patients with psychiatric histories or with cognitive impairment. (Supported by UCB S.A.)