REM Not Correlated With Increased Risk of SUDEP

January 9, 2021
Marco Meglio
Marco Meglio

Marco Meglio, Associate Editor for NeurologyLive, has been with the team since October 2019. Follow him on Twitter @marcomeglio1 or email him at

Although the percentage of total seizures arising from REM was low, it is notable that 2 out of 26 patients (7.7%) had REM-onset seizures.

Data from a small sample study suggested that there is no correlation between rapid eye movement (REM) sleep onset seizures and the incidence of sudden unexpected death in epilepsy (SUDEP).1

Among a cohort of 26 patients who died of SUDEP, the frequency of seizures arising from REM sleep was 1.9% (2 of 107), comparable to the 1% frequency previously reported. The investigators noted that the lack of association may be in part because of the cohort was too small to find such association. The results were presented at the American Epilepsy Society (AES) Virtual Meeting, December 6–10, 2020.

Though the percentage of total seizures arising from REM was low, there were 2 patients (7.7%) who recorded REM-onset seizures. Each of the 2 patients who recorded REM-onset seizures had a single seizure arising from REM sleep and both had additional seizures arising from other states.

Senior author Michael Sperling, MD, director, Jefferson Comprehensive Epilepsy Center, and Jefferson Clinical Neurophysiology Laboratory, and colleagues drew parallels to the MORTEMUS study, which found 12.5% of patients whose sleep stage was known at the time of terminal seizure was in REM sleep at the time of death.

READ MORE: Novel Associations Between Functional Seizures and Cerebrovascular Disease Identified

Patients included in Sperling et al’s work had undergone video-encephalogram (EEG) monitoring and were recorded on each seizure and the state in which it arose. There was a male/female ratio of 1:1 among 26 patients, with a mean age at the time of monitoring and at the time of death of 31 (range, 14–55) and 34 (range, 18–57), respectively.

Data on the percentage of patients with seizure arising from each vigilance state showed that 42% were during sleep and wakefulness, 31% wakefulness only, 19% sleep only, and 8% with no seizures recorded. There were 22 (85%) patients who had a history of tonic-clonic seizure and 22 (85%) with a history of nocturnal seizures. The mean number of seizures per patient was 4.1 (range, 0–13) among a total cohort of 107 seizures recorded.

The study did have notable limitations, including the impact that inpatient hospitalization and antiepileptic medication taper on the natural sleep pattern, as well as the lack of control patients in this sample.

"In a power analysis, we calculated that several hundred control patients would be required to demonstrate a statistically significant difference, which was not feasible. Additional work in larger cohorts and utilization of ambulatory EEG data may be useful in further addressing the potential association of REM-onset seizures and SUDEP,” Sperling et al concluded.

Another presentation at AES 2020 found the occurrence of SUDEP to be higher than what was previously reported. The overall rate of mortality was 0.26 per 100,000 infants and children compared to literature which has suggested that the mortality rate is 0.16 per 100,000 infants and children. Additionally, the data from the National Institutes of Health (NIH) and Centers for Disease Control and Prevention (CDC) Sudden Death in the Young (SDY) Case Registry revealed that black and multiracial infants and children (38%; 0.32 per 100,000) have a SUDEP mortality rate 1.5 times higher than white infants and children (58%; 0.22 per 100,000).2

1. Goldstein L, Resekhi RT, Sperling M, Nei M. The prevalence of REM sleep onset seizures in a cohort of patients who died of SUDEP. Presented at AES 2020 Virtual Meeting: December 4–8, 2020. Abstract 258.
2. Whittmore V, Burns KM, Shaw E, Faulkner M, Tian N. Population-based surveillance of sudden unexpected death in epilepsy using the sudden death in the young case registry. Presented at AES 2020 Virtual Meeting: December 4–8, 2020. Abstract 405.