Many anticonvulsants block Sodium (Na+) channels, Calcium (Ca2+) channels, AMPA receptors or NMDA receptors. Some anticonvulsants inhibit the metabolism of GABA or increase its release.
Some anticonvulsants have shown antiepileptogenic effects in animal models of epilepsy. That is, they either prevent the expected development of epilepsy or can halt or reverse the progression of epilepsy. However, no drug has shown this effect in human trials.
The usual method of achieving approval for a drug is to show it is effective when compared against placebo, or that it is more effective than an existing drug. In monotherapy (where only one drug is taken) it is not ethical to conduct a trial with placebo on a new drug of uncertain efficacy. This is because untreated epilepsy leaves the patient at significant risk of death. Therefore, almost all new epilepsy drugs are initially approved only as adjunctive (add-on) therapies. Patients who's epilepsy is currently uncontrolled by their medication (i.e., it is refractory to treatment) are selected to see if supplementing the medication with the new drug leads to an improvement in seizure control. Any reduction in the frequency of seizures is compared against a placebo.
Once there is confidence that a drug is likely to be effective in monotherapy, trials are conducted where the drug is compared to an existing standard. For partial-onset seizures, this is typically carbamazepine. Despite the launch of over ten drugs since 1990, no new drug has been shown to be more effective than the older set, which includes carbamazepine, valproate and phenytoin. The lack of superiority over existing treatment, combined with the lack of placebo-controlled trials, means that few modern drugs have earned FDA approval as initial monotherapy. In contrast, Europe only requires equivalence to existing treatments, and has approved many more. Despite their lack of FDA approval, the American Academy of Neurology and the American Epilepsy Society still recommend a number of these new drugs as initial monotherapy.