Facing up to SUDEP.

Epilepsy is a chronic condition that affects approximately 2.3 million people in the United States. Approximately one-third of people with epilepsy continue to have seizures despite adequate medication trials and adherence to medical recommendations. People of all races and backgrounds, young and old, are affected by epilepsy from a variety of etiologies, and epilepsy carries a high burden of comorbidity including the risk of death. The overwhelming majority of the deaths are due to sudden unexpected death in epilepsy (SUDEP), which is closely related to a recent convulsive or tonic-clonic seizure, and thus may be a direct result of a seizure. SUDEP preferentially affects young people and carries a public health burden estimated in potential life years lost (1.16 cases per 1,000 patients with epilepsy) second only to stroke. The mechanism of SUDEP has been debated: CNS shutdown with impaired autonomic regulation, cardiac causes (arrhythmias), or respiratory etiology (mostly central apneas with hypoventilation) are the top choices, with recent insights from a video-EEG review of recorded SUDEP in the epilepsy monitoring unit. Understanding the mechanism of SUDEP may help to prevent it. Increasing research and funding are being devoted to SUDEP to prevent this feared consequence of uncontrolled epilepsy. The rapidly growing literature indicates that uncontrolled and unattended tonic-clonic seizures are likely the strongest risk factor, but other important risk factors, such as early onset of epilepsy, longer duration of illness, and prolonged postictal generalized EEG suppression (PGES), also correlate with SUDEP. Unfortunately, aside from optimized seizure control, these features are difficult to influence. One potential modifiable risk factor has been identified: being found in bed in the prone position; however, there were no large studies to confirm this. In this issue of Neurology®, Liebenthal et al. describe their systematic review and meta-analysis of the literature up to 2003 in which they found 25 studies (of 1,106 unique ones): 14 single case studies and 11 case series where body position was documented in relation to the SUDEP. These authors used the Nashef definition of SUDEP: witnessed or unwitnessed, nontraumatic and nondrowning, excluding status epilepticus or an identifiable cause of death at autopsy; cases were divided into definite (autopsy-proven), probable (no autopsy), or possible (other cause of death was listed). Assumptions were made about SUDEP occurring during sleep if the person was found dead in bed, and during wakefulness if found outside of the bed or bedroom. Of the 253/413 SUDEP cases reviewed in which body position was adequately documented, 73.3% were prone (p , 0.001). Of note, in the 11 cases that had SUDEP documented on video-EEG, 100% were found prone, had had a generalized tonic-clonic seizure, and had PGES. A subset analysis by age revealed that patients older than 40 years were less likely to be prone than those younger than 20 years. No difference was found for sex or for sleep/wake state at time of seizure onset. Discovering a possible modifiable risk factor for SUDEP, such as sleeping in the prone position, is clearly important because it implies the possibility of preventing a substantial number of deaths by having patients sleep on their back, as has been done with SIDS (sudden infant death syndrome). Animal models that link SUDEP to a serotonin mechanism (reduced drive to breathe) will be important. Furthermore, it will be critical to identify patients at high risk of SUDEP, such as those with PGES during their epilepsy monitoring unit stay, those with intellectual disability and frequent uncontrolled tonic-clonic seizures, or possibly carriers of specific potassium channel genes linked to sudden death in animals. There are several limitations of the study. From a very large number of publications, only 25 studies were used; this small proportion of studies that described a prone or supine position at time of SUDEP may have introduced a reporting bias. There was no control group of final position in patients with tonic-clonic seizures to determine the frequency of patients ending prone without associated SUDEP. The study also cannot answer whether sleeping in a prone position is more prevalent among patients with epilepsy who die of SUDEP, or