Abstracts

Rationale: Temporal lobe epilepsy (TLE) is the commonest partial epilepsy of adulthood. Although widely perceived as a symptomatic disorder, several forms of familial non-lesional TLE have been reported. Descriptions of familial mesial TLE have varied from a benign syndrome with prominent deja vu, and without antecedent febrile seizures (FS) or MRI abnormalities, to heterogeneous but generally more refractory epilepsies, commonly with a history of FS and with hippocampal atrophy and high T2 signal on MRI. The explanation for this heterogeneity between studies is not clear. The high monozygotic twin concordance rates observed in familial TLE provide compelling evidence of a genetic aetiology. Although two large, dominant, benign mesial TLE families have been reported, one with linkage to chromosome 4q, no genes have been identified and the usual mode of inheritance remains unknown. Methods: Families were ascertained in which two or more first degree relatives had a history of TLE and no potentially epileptogenic abnormality was detectable on MRI. Probands and all available first and second degree relatives were interviewed, including completion of a validated, structured epilepsy questionnaire. Source clinical records, EEG details and brain imaging were scrutinized. Goodness of fit to Mendelian dominant inheritance models was tested, with models varying in penetrance (60-100%) and the epilepsy phenotypes included (from TLE alone to all epilepsy, febrile seizures or single seizures). Results: 20 new families, including 50 affected individuals, were studied. Median seizure onset age was 17 years (range 3-46). 6/50 (12%) had antecedent FS. Mesial semiology predominated, particularly psychic features, including deja vu in 36/50 (72%). The epilepsies were generally mild, with only 23/50 (46%) ever having generalized tonic-clonic seizures. 18/50 (36%) were on no treatment at the time of the study. 28/50, (56%) had good seizure control with a single antiepileptic drug, while only 4 individuals (8%) required more than one drug. MRI was normal in 32/34 (94%), with none showing MRI correlates of hippocampal sclerosis. Hippocampal pathology was normal in one case who became seizure free after temporal lobectomy. Inheritance patterns in probands’ relatives were analyzed in these families pooled with 19 other TLE families previously reported by us. Observed frequencies of TLE in first degree relatives were significantly lower than predicted by any of our dominant models, arguing against dominant inheritance as the usual mode of inheritance in this condition. Conclusions: This new series provides further evidence of familial TLE as a relatively benign syndrome with frequent déjà vu and without hippocampal sclerosis. Segregation analysis reveals significantly fewer affected relatives than predicted by even a low (60%) penetrance dominant model. This strongly suggests that complex inheritance, similar to that widely accepted in idiopathic generalized epilepsies, is the usual mode of inheritance in this condition.