gms | German Medical Science

Objective: In patients with pharmacoresistant epilepsy caused by an epileptogenic focus in an eloquent brain area, operative resection may result in severe neurological deficits. Therefore we have developed a novel experimental technique by implanting valproate-containing osmotic minipumps with a catheter (1) or a gel with controlled release (2) into (1) or on the surface of (2) the epileptic hippocampus in a mouse model of pharmacoresistant epilepsy.

Methods: The well-described mouse model of unilateral intrahippocampal kainate injection was used that reflects pharmacoresistant temporal lobe epilepsy. In the first operation, 50 nL (20 mM) kainate was injected stereotactically into the left dorsal hippocampus followed by the development of limbic seizures within a period of four weeks. In these mice osmotic mini-pumps with catheter tips within the left hippocampus were implanted, delivering either saline or 10 mg valproate within 7 days. In a second group (2) valproate gel or placebo was applied on the surface of the hippocampus after corticotomy. Subsequently, intrahippocampal encephalography electrodes were implanted in all mice and long-term recordings were performed. Anticonvulsant effects were measured by a mathematical counting paradigm of high frequency oscillation (fast ripples).

Results: Clear antiepileptic effects of valproate micro-pumps and gels were observed compared to the placebo groups. Valproate micro-pumps decreased the fast ripple activity from 100% to 37.7%. Valproate gels decreased the fast ripple activity from 100% to 5% (p<0.05). The antiepileptic effect disappeared after 14 days in both groups and the previous epileptic activity was restored. After this local antiepileptic treatment, 10 mg valproate was given intraperitoneally in a single dose in both the verum and the placebo groups to test whether the animal model used was pharmacoresistant to conventional application of valproate (as described in the literature). Intraperitoneal valproate application, however, did not decrease fast ripple activity. All hippocampal specimens showed Ammon’s horn sclerosis as a neuropathological correlate of temporal lobe epilepsy.

Conclusions: We have demonstrated that local, but not systemic application of valproate has antiepileptic effects in a conventionally pharmacoresistant epilepsy model.