Sleep-related hypermotor epilepsy Long-term outcome in a large cohort

Objective: To assess the long-term outcome of sleep-related hypermotor epilepsy (SHE). Methods: We retrospectively reconstructed a representative cohort of patients diagnosed with SHE according to international diagnostic criteria, sleep-related seizures >= 75% and follow-up >= 5 years. Terminal remission (TR) was defined as a period of >= 5 consecutive years of seizure freedom at the last follow-up. We used Kaplan-Meier estimates to calculate the cumulative time-dependent probability of TR and to generate survival curves. Univariate and multivariate Cox regression analyses were performed. Results: We included 139 patients with a 16-year median follow-up (2,414 person-years). The mean age at onset was 13 +/- 10 years. SHE was sporadic in 86% of cases and familial in 14%; 16% of patients had underlying brain abnormalities. Forty-five percent of patients had at least 1 seizure in wakefulness lifetime and 55% had seizures only in sleep (typical SHE). At the last assessment, 31 patients achieved TR (TR group, 22.3%), while 108 (NTR group, 77.7%) still had seizures or had been in remission for <5 years. The cumulative TR rate was 20.4%, 23.5%, and 28.4% by 10, 20, and 30 years from inclusion. At univariate analysis, any underlying brain disorder (any combination of intellectual disability, perinatal insult, pathologic neurologic examination, and brain structural abnormalities) and seizures in wakefulness were more frequent among the NTR group (p = 0.028; p = 0.043). Absence of any underlying brain disorder (hazard ratio 4.21, 95% confidence interval 1.26-14.05, p = 0.020) and typical SHE (hazard ratio 2.76, 95% confidence interval 1.31-5.85, p = 0.008) were associated with TR. Conclusions: Our data show a poor prognosis of SHE after a long-term follow-up. Its outcome is primarily a function of the underlying etiology.