Abstracts

RATIONALE: Intractable temporal lobe epilepsy with hippocampal damage is often linked to a precipitating insult during childhood. The data available suggests that once epilepsy has emerged recurrent seizures may cause further structural and functional changes. However, longitudinal follow-up studies are needed to determine the course of epileptic process and to identify surrogate markers for disease modification.
METHODS: We designed a prospective follow-up study of 112 patients with newly diagnosed partial epilepsy who were allocated to treatment with either carbamazepine (CBZ; n=54) or with tiagabine/vigabatrin (TGB/VGB; n=40/18). The aim of the study was to investigate the occurrence of hippocampal damage and to assess whether the changes progress in the long-term follow-up. Specifically, we wanted to compare the hippocampal volumes in patients treated with either standard sodium-blocking agent or with newer gabaergic agents, and to examine whether volumetric measures correlate with clinical outcome. The initial magnetic resonance imaging (MRI) was performed before the antiepileptic medication was started, and later, after 1 year, 2-3 years and 5 years of follow-up. We used Cavalieri method of modern design stereology for volume estimations of the hippocampus. A group of 20 healthy subjects served as control population for MRI. The clinical outcome of the patiens was determined by assessing the efficacy and retention rate of the initial treatment.
RESULTS: The mean left and right hippocampal volumes did not differ between controls and patients studied at baseline, 1 year, 2-3 years and 5 years of follow-up. At least a 2 SD reduction in the volume of the hippocampus was observed in altogether 6 patients. In the general linear model of repeated measures there was a trend of volume decrease when left and right hippocampal volumes of each patient were compared during 2-3 years of follow-up but the change was not significant. When patients were divided into two groups according to the treatment (CBZ vs TGB/VGB) or seizure control (seizure-free vs seizures) no difference was observed in the serial hippocampal volumes. Accordingly, the efficacy of treatment in intial treatment groups was similar. Comparing the overall effectiveness of the treatments, CBZ proved to be more effective than TGB/VGB.
CONCLUSIONS: This is first prospective quantitative MRI study that systematically follows large number of newly diagnosed partial epilepsy patients. The study shows that hippocampal damage at the time of diagnosis is rare. No significant progressive change in the hippocampal volume was observed during follow-up. In the future most useful way for evaluating the course of epileptic process and effects of therapy might be combining several measures of outcome such as seizure-control, imaging data and cognitive data.
[Supported by: Kuopio University Hospital EVO grant]