Symptomatic temporal lobe epilepsy typically develops in three phases: brain damage --> epileptogenesis --> spontaneous seizures (epilepsy). The challenge is to prevent epileptogenesis after injury. We hypothesized that alleviation of damage by caspase inhibitors will reduce epileptogenesis or at least have disease-modifying effects (less severe epilepsy, milder cognitive decline). Epileptogenesis was triggered by amygdala stimulation-induced status epilepticus (SE) in rats and spontaneous seizures were monitored with video-electroencephalography (EEG). First, we tested the neuroprotective effect of a 1-week treatment with caspase 1, 3 or 9 inhibitors (3 micro g/d/i.c.v., started 3 h after the beginning of SE). The least damage to the hippocampus was observed in animals treated with the caspase 3 inhibitor (z-DEVD-fmk) which reduced the enzyme activity to 6% of that in the vehicle group. Thus, z-DEVD-fmk was chosen for long-term studies, in which the treatment regime remained the same except the dose was doubled (6 micro g/d/i.c.v.). Video-EEG monitoring was performed for 3 to 4 weeks, starting either 8 or 14 weeks after SE. One group of animals was tested in water-maze and fear-conditioning tests, and all animals were perfused for histological analysis. Treatment with the caspase 3 inhibitor neither prevented the development of epilepsy, nor had any disease-modifying effects. Mossy fibre sprouting, however, was reduced. The present data indicate that administration of z-DEVD-fmk monotherapy was not antiepileptogenic despite its short-term neuroprotective effects. These findings challenge the idea that prevention of cell death is the primary target for the development of antiepileptogenic compounds.