Epilepsy in Rett syndrome: Association between phenotype and genotype, and implications for practice

PURPOSE To investigate the association between genotype (methyl-CpG-binding protein 2 (MECP2 gene mutation)) and epileptic seizure phenotype in Rett syndrome. METHODS We used the British Isles Rett syndrome survey to identify 137 subjects with one of the nine most frequent MECP2 gene mutations and invited their parents or carers to participate in a postal questionnaire and telephone interview. The questionnaire recorded information about epileptic seizure types, non-epileptic vacant spells and treatments. Two investigators conducted telephone interviews and three epileptologists classified their epileptic seizures. RESULTS 89 subjects (65%) responded. The epilepsy prevalence was 67%, and 74% had non-epileptic vacant spells. The epilepsy prevalence within specific genotypes ranged from 47% (mutation C-terminal deletion, downstream of the Transcription Repression Domain) to 100% (mutation p.R270X, c.808C>T). The prevalence of non-epileptic vacant spells within genotypes ranged from 50% (mutation p.R306C, c.916C>T) to 100% (mutation p.R106W, c.316C>T). The epileptologists differed considerably in their classification of events, particularly of non-epileptic vacant spells. CONCLUSIONS The large majority of people with Rett syndrome have epilepsy. Most have multiple epileptic seizure types, although generalised tonic-clonic seizures are the most common. There were no significant clinical differences between genotypes. The clinical differentiation of non-epileptic vacant spells is difficult. Discordance in epileptic seizure classification between clinicians suggests that caution is needed, since the clinical history alone cannot adequately classify the epileptic seizure type in Rett syndrome.