Comparing New Antiepileptic Medications

Source: Marson A, et al. New antiepileptic drugs: A systematic review of their efficacy and tolerability. BMJ 1996; 313:1169-1174.

Five new antiepileptic drugs (aeds) have appeared within the last three years. These include gabapentin and lamotrigine in the United States, Canada, and Europe; vigabatrin in Canada and Europe; and zonisamide in Japan. Topirimate was just released in the United States in January 1997. Marson and colleagues now report that there is no difference in antiepileptic efficacy and tolerability between these agents.

The authors carried out a meta-analysis of 28 clinical trials of gabapentin, lamotrigine, topirimate, vigabatrin, and zonisamide ("new AEDs"). All studies were randomized, placebo-controlled, add-on trials for partial-onset seizures. A total of 3883 patients were enrolled. Efficacy was assessed by comparing the percent of patients achieving a 50% reduction in seizure rate on the new AED to the percent of patients achieving a 50% reduction in seizure rate on the placebo. Tolerability was assessed by comparing the percent of patients withdrawing from the new AED treatment to the percent of patients withdrawing from the placebo. No significant difference between these new AEDs was seen on either the efficacy or tolerability measure.


Careful statistical meta-analyses such as that carried out by Marson et al are valuable to the neurologist who must compare new AEDs and select the best agent for a given patient. As these authors state, a systematic approach seems better than "reliance on the views of ‘opinion leaders.’" In particular, views of such leaders may be influenced by which new AED an investigator had the opportunity to work with in pre-marketing experimental clinical trials. It is human nature to feel most comfortable when one can recommend an already familiar medication. Marson et al conclude that gabapentin, lamotrigine, topirimate, vigabatrin, and zonisamide have similar efficacy and tolerability as add-on therapy against partial onset seizures.

The antiepileptic efficacy measure Marson et al employed (percent of patients who achieved a 50% reduction in seizure rate) is reasonable. However, their rather nonspecific tolerability measure (percent of patients who withdrew from the studies) does not give us potentially important details of the side effect profiles of these medications. In fact, the authors did undertake a detailed analysis of specific side effects reported in these new AED trials. However, they state that those findings ". . . are outside the scope of this article and will be reported separately." This is unfortunate. We are left making judgments by comparing putatively meaningful efficacy measures but knowing only less useful and meaningful tolerability measures.

If overall efficacies and tolerabilities of these new AEDs are similar, then the neurologist may wish to select the drug for a given patient based on specific side effects and costs. The table shows specific medication toxicities as noted in two recent reviews (Dichter M, Brodie M. N Engl J Med 1996;334:1583-1590; Wilder B. Can J Neuro Sci 1996;23 [Suppl 2]:S18-S23). The monthly costs of the doses studied in the meta-analysis of Marson et al are also presented in the table.


Side effects and prices on new AEDs

MedicatioSide Effects Monthly costs

GabapentiSomnolence $191

Lamotrigine Rash $119

Topirimate Cognitive difficulties $245

VigabatriPsychosis $239

This paper of Marson et al also illustrates and discusses the need for several other types of clinical trials of new AEDs. These include: 1) head-to-head comparisons of new agents with each other and older drugs, rather than against placebos, 2) monotherapy trials with patients with less severe epilepsy, and 3) trials with patients with primary generalized seizures. —drl