Abstracts

RATIONALE: Temporal lobe epilepsy is often refractory to medical treatment. Long term remission has been reported in only one-third to one-half of the patients with temporal lobe epilepsy. Hippocampal sclerosis is the most common pathology in temporal lobe epilepsy. The presence of hippocampal sclerosis on MRI is known to correlate with medical intractability, however many cases of epileptics with hippocampal sclerosis on MRI have been well controlled by medical treatment. Thus, there is still a debate about the influence of hippocampal sclerosis on the response to medial treatment of temporal lobe epilepsy. The purpose of this study is to test whether the presence of hippocampal sclerosis has an influence on drug response and other clinical features of temporal lobe epilepsy.
METHODS: Forty-nine epileptics with complex partial seizures were enrolled. Their follow-up periods with antiepileptic medication were more than 2 years. To exclude the complex partial seizures confined to frontal lobe, the patients having brief seizure duration and no postictal confusion were excluded. MRI examinations were performed at 1.5 Tesla with axial T1- and T2-weighted images and coronal T2-weighted oblique thin-section images. Two neurologists and one radiologist independently determined the presence of hippocampal sclerosis on MRI by visual evaluation and then the patients were divided into hippocampal sclerosis group and non-hippocampal sclerosis group. We compared drug response, seizure frequency, sex, seizure onset, disease duration, secondary generalization, electroencephalographic abnormality, previous medical insult, and polytherapy between the two groups. The patients with seizure free for more than 1 year were considered as the tractable epileptics.
RESULTS: Fifteen patients were in hippocampal sclerosis group and 34 patients were in non-hippocampal sclerosis group. Ten of 15 (66.7%) in hippocampal sclerosis group and 11 of 34 (32.4%) in non-hippocampal sclerosis group were intractable and the difference of intractability between the two groups was significant (p [lt] 0.05). Eleven of 15 (73.3%) in hippocampal sclerosis group and 8 of 34 (23.5%) in non-hippocampal sclerosis group had seizures more than 4 times per year and the difference was significant (p [lt] 0.01). Other clinical features were not different between the two groups (p [gt] 0.05).
CONCLUSIONS: Our results showed that medical intractability in temporal lobe epilepsy was significantly associated with hippocampal sclerosis. Hippocampal sclerosis itself may be a crucial factor determining drug response of temporal lobe epilepsy.