Abstracts

Levetiracetam (LEV) is FDA approved for treatment of focal seizures in adults and has been reported to be effective in the treatment of seizures in children. We sought to determine the efficacy of LEV at one year in children classified by seizure type, cognitive status, concomitant medications and prior anticonvulsant use.
This is a retrospective study of 59 children (ages 9 months to 23 years, mean age 11 years) with intractable seizures treated with LEV. Charts were reviewed for seizure type (focal, generalized, or mixed), cognitive function (no special education vs. special ed), concomitant anticonvulsant medications (range: 0 to 5), and number of prior anticonvulsant drugs (range: 1 to 12). Seizure frequency was determined before and after 12 months of LEV therapy.
Of 59 patients, 15 (25%) had focal seizures, 29 (49%) had generalized seizures, 13 (22%) had mixed seizure types, 2 were not classified. At 12 months 18 children were no longer on LEV due to lack of efficacy or side effects.
Good to excellent seizure control (50%-100% reduction) was attained in 6 (40%) of patients with focal seizures, 16 (55%) with generalized seizures and 8 (61%) with mixed seizures.
Good to excellent control was achieved in 7 of 15 (47%) patients with no/mild cognitive impairment and in 23 of 43 (53%) patients with moderate/severe cognitive impairment. (No cognitive information on 1 child).
Good to excellent control was accomplished in 16 of 24 (67%) patients with prior exposure to fewer than 4 anticonvulsant drugs and in 15 of 35 (43%) previously treated with 4 or more drugs.
Among those children on LEV and one other drug, 8 of 20 (40%) had excellent ([gt]90%) control and 3 of 20 (15%) had poor ([lt]50%) control. It is of note that none in the excellent control group were on zonisamide (ZNS) and all 3 in the poor control group were on ZNS. (Chi square value=11; p[lt]0.001.)
Adverse effects (AE[apos]s) were not a function of dose, as children without AE[apos]s had a higher mean dose than those with AE[apos]s.
In this group of children, LEV worked equally well on focal, generalized and mixed seizures. Efficacy of LEV was independent of cognitive status. Children were more likely to achieve good to excellent seizure control (50%-100% reduction) if they had failed fewer than four medications previously, presumably a less intractable group of patients. Adverse effects (AE[apos]s) were not a function of dose. This could be a function of different rates of metabolism (levels were not routinely obtained), or represent idiosyncratic responses to the medication.
Our finding that those children taking the combination of LEV and ZNS had a significantly worse outcome than those on LEV and a different drug warrants further study, both clinically and from the standpoint of mechanisms of action of LEV and ZNS and/or pharmacokinetic interactions between them.
[Supported by: UMDNJ Foundation Student Summer Fellowship for Marjorie Bunch]