Interactions between riluzole and conventional antiepileptic drugs -- a comparison of results obtained in the subthreshold method and isobolographic analysis.

The exact types of pharmacodynamic interactions between riluzole and conventional antiepileptic drugs were evaluated in two available ways, the subthreshold and isobolographic analysis. Maximal electroshock test in mice was used as an animal model for generalized tonic-clonic convulsions. In the first method, riluzole (1.25-2.5 mg/kg) significantly raised the electroconvulsive threshold in mice. The drug administered at its subprotective dose of 0.3125 mg/kg enhanced the antiseizure activity of carbamazepine and phenobarbital, while, when applied at the higher dose of 0.625 mg/kg, it potentiated also the action of valproate and diphenylhydantoin. Riluzole (0.625) alone and in combinations with antiepileptic drugs did not produced any motor or log-term memory deficit. Results obtained from isobolographic analysis determined pure additive interaction between riluzole and all used conventional antiepileptic drugs. Since riluzole did not change plasma concentrations of co-administered antiepileptics, pharmacokinetic interactions, at least in terms of their free plasma levels, do not seem probable. The results of the present study confirm significant antiseizure properties of riluzole in the model of generalized tonic-clonic epileptic attacks. Moreover, comparison of effects obtained from both methods evaluating drug interactions strongly indicates a crucial role of the isobolographic analysis in verification and supplementation data achieved from the subthreshold method.

Glutamate receptor antagonists differentially affect the protective activity of conventional antiepileptics against amygdala-kindled seizures in rats.

LY 300164 (5 mg/kg), a selective non-competitive antagonist of AMPA/kainate receptors, exerted a significant anticonvulsant effect in amygdala-kindled rats, being ineffective at 2 mg/kg. LY 235959 (1--5 mg/kg), a selective competitive antagonist of NMDA receptors, failed to modify behavioral and electrographic correlates of kindled seizures. Amygdala-kindled seizures were inhibited by conventional antiepileptics, their lowest effective doses were: 20 mg/kg for carbamazepine and phenobarbital, 50 mg/kg for diphenylhydantoin, and 100 mg/kg for valproate magnesium. The combined treatment of the AMPA/kainate antagonist (2 mg/kg) with valproate at sub-effective doses (25--75 mg/kg) resulted in the reduced severity and duration of kindled seizures. Also, a clear-cut protective effect was observed when LY 235959 was co-administered with diphenylhydantoin (40 mg/kg). Any interaction at the pharmacokinetic level can be excluded because neither LY 300164 nor LY 235959 interfered with the free plasma levels of valproate or diphenylhydantoin, respectively. The combination of the AMPA/kainate receptor antagonist (2 mg/kg) with valproate (75 mg/kg) did not impair performance of rats in the rotorod test (motor co-ordination) or passive-avoidance task (long-term memory). Conversely, the NMDA receptor antagonist alone or in combination with diphenylhydantoin, produced significant mnemonic deficits. The results indicate that AMPA/kainate receptor antagonists might be of importance as adjuvant antiepileptic drugs in patients treated with valproate. A possible use of NMDA receptor antagonists may be questionable.