The 'number needed to treat' with Levetiracetam (LEV): comparison with the other new antiepileptic drugs (AEDs).

K. Rijckevorsel,P. Boon
DOI: https://doi.org/10.1053/SEIZ.2001.0497
IF: 3.414
2001-04-01
Seizure
Abstract:Levetiracetam (LEV, Keppra) is a new antiepileptic drug for the add-on treatment of partial onset seizures. Pivotal studies have shown a significant efficacy for all tested doses (pooled data: 28% of responders with 1000 mg per day, 32% of responders with 2000 mg per day and 41% of responders with 3000 mg per day). Its safety profile is also quite encouraging: in pivotal studies, adverse events were mostly central nervous system related and manifested as somnolence, asthenia and dizziness; often being mild to moderate in severity. No life-threatening adverse events related to the study drug were described. Furthermore, LEV has a very straightforward pharmacokinetic profile: no protein binding, no potential or significant drug interactions, no hepatic metabolization, as well as displaying linear kinetics and no active metabolites. Because neurologists have received numerous new therapeutic possibilities over the last few years, it is important to compare this new antiepileptic drug (AED) to the others in terms of efficacy and potential new advantages. A first attempt to compare AEDs for efficacy and/or safety was made by Marson et al.1 using the odds ratios from meta-analysis of double-blind placebocontrolled studies. This analysis did not show significant differences between the new AEDs. This could be due to the fact that the odds ratio does not take into account the placebo response and the variability of the therapeutic answer between the different studies. Other comparative methods used include the Star Rating system of Brodie 2 and the improvement rates at recommendedoses suggested by Cramer et al.3. The first method is partially subjective while the second one could give some interpretations about the notion of ‘recommended’ doses. The number needed to Table 1: Comparative number needed to treat to obtain one addition (to placebo) responder.
What problem does this paper attempt to address?