Each year, about one in every thousand people with epilepsy will die suddenly, usually during sleep and immediately following a seizure. Sudden unexpected death in epilepsy, or SUDEP, was a little-studied and poorly understood phenomenon until large clinical trials of antiepileptic drugs in the 1990s revealed a significant incidence. Deaths first feared to be linked to the study drugs turned out to be related to disease severity – in particular, the presence of generalized tonic-clonic seizures (Drug Saf. 2003;26[10]:673-83). Patients with SUDEP are most often found prone in their beds, and in rare cases where they have been observed before dying, they are seen having a seizure followed by depressed or halted breathing. A detailed understanding of who is at risk for SUDEP and translation of that knowledge into clinical practice have taken decades to develop, and both re-searchers and patient advocates say there is far more work to be done. In 2017, the American Academy of Neurology and the American Epilepsy Society took a major step to highlight SUDEP risk by a publishing clinical guideline urging clinicians to proactively disclose risk. The guideline reported that SUDEP risk is generally low at only 1.2 in 1,000 adult patient-years, and advised clinicians to counsel patients and families that, therefore, 999 patients will not die in any given year for each one who does (Neurology. 2017 Apr;88[17]:1674-80). But for adult patients experiencing three or more generalized tonic-clonic seizures per year, the guideline also notes that the risk is 15-fold greater: 18 patients in 1,000 will die every year. Among children with epilepsy, meanwhile, the 2017 AAN/AES guideline estimates risk at only 0.22 cases per 1,000 patient-years. But that seemingly reassuring figure has been challenged by two studies published since that found children’s SUDEP risk to be on par with that of adults at about 1 in 1,000 patient-years.

Broaching the subject The need for SUDEP risk to be addressed frankly with patients and families “has been kind of a growing consensus. But SUDEP is something [clinicians] don’t like to talk about because it’s uncomfortable,” said neurologist Christopher DeGiorgio, MD, a SUDEP researcher at the University of California, Los Angeles. “SUDEP is a rare event, but it’s not rare in the person with drug-resistant epilepsy. I think that’s important to emphasize.” Elizabeth Donner, MD, a neurologist at the Hospital for Sick Children in Toronto who studies SUDEP in children and adolescents, said that, traditionally, “neurologists did not advise their patients, whether children or adults, that SUDEP exists. People didn’t talk about it. When I make an argument for why physicians should counsel about SUDEP, one of the first things I point out is that, when we are talking to parents of children with epilepsy, the question of whether their child could die of a seizure is the elephant in the room. And if a family isn’t asking you, ‘Could my child die?’ it’s just because they’re terrified.” Dr. Donner, who is a coauthor on the 2017 AAN/AES guideline, said clinicians “don’t necessarily need to present the data in the positive,” as the guideline suggests. “I think that for most people living with epilepsy in their family and caregivers, we can present it as it stands as one in a thousand.” Tom Stanton, the director of the Danny Did Foundation, which is based in Evanston, Ill., and was started in 2010 to raise awareness of SUDEP, said in an interview that “the topic of death in epilepsy has historically been underrepresented. And I think some of the reasons are valid: You want to propose a lifespan and a lifestyle that is positive and healthy, that you can do X, Y, and Z with epilepsy. Introducing the idea of sudden death into that narrative is tricky.”

Dr. Christopher DeGiorgio Dr. Elizabeth Donner

The publication of the 2017 AAN/AES guideline “was a big deal in our world. They were a long time coming,” Mr. Stanton said. “We might have worded them a little bit more strongly and been more definitive about certain aspects of how to communicate with patients and parents, but on the other hand, I think it was a victory that they came out at all.” While many specialists are making SUDEP discussions a routine part of their clinical practice – and stressing that good seizure control is essential to lower-ing risk – the conversation isn’t happening everywhere, Mr. Stanton noted. “We’re surrounded by activist doctors who are really great leaders, but when we see survey results we realize there’s a lot of ground to cover. We are talking to families on a daily basis who have still not heard about SUDEP. Some of them have had epilepsy in their lives for many years.” SUDEP in Adults: Etiology, biomarkers, and risk In 2017, Dr. DeGiorgio and colleagues ranked the top risk factors for SUDEP and found that three or more generalized tonic-clonic seizures per year was No. 1, followed by frequent seizures of any type, no treatment, younger age at onset of disease, being on three or more antiepileptic drugs, and low IQ (Front Neurol. 2017 Sept 21. doi: 10.3389/fneur.2017.00473). “The top risk factor by a huge margin is three or more grand mal seizures in the last year. And some patients will have had hundreds,” Dr. DeGiorgio commented. Biomarker studies, meanwhile, have revealed more about the etiology of SUDEP and how repeated seizures can damage brain structures associated with breathing – which suggests a causative role for tonic-clonic seizures. A 2017 study led by U.K. researchers used functional and structural MRI to detect differ-ences between the brains of epilepsy patients at high risk for SUDEP and those of patients classified as low risk based on clinical factors (Front Neurol. 2017 Oct 16. doi: 10.3389/fneur.2017.00544). High-risk patients, the investigators found, had evidence of impaired connectivity among brain regions that regulate cardiovascular and respiratory control. In a related study published this year, the same researchers reviewed neuroimaging results from patients who later succumbed to SUDEP and found that they had evidence of the kinds of structural and functional changes previously identified in the high-risk patients (Front Neurol. 2019 Mar 5. doi: 10.3389/fneur.2019.00185). credit: Zinkevych/Getty Image

“We treat epilepsy not only to reduce seizures but to reduce death and morbidity: falls, injuries, burns, drowning,” Dr. DeGiorgio said. “We’re also treating epilepsy to prevent the chronic brain injury that occurs with these frequent seizures. SUDEP risk comes from progressive injury to structures that normally would terminate seizures and also support respiration.” Cardiac biomarkers seen associated with SUDEP risk include the QT interval, which is frequently prolonged in patients at risk, and reduced heart rate variability, which may indicate damage to brain and brain stem structures governing the heart (Front Neurol. 2018;9:484).

“When you have a severe seizure, your heart rate goes way up, but at the same time your oxygen levels are dropping, creating a gap between the oxygen needs of the heart and how much oxygen you can provide it because you’re not breathing effectively,” Dr. DeGiorgio said. “Normally, the heart can handle these kinds of stresses. But if the brain is not healthy and can’t modulate the heart rate effectively, it puts the heart at risk for a lethal arrhythmia.” Postictal suppression, a dramatic drop in brain activity following seizures that can be captured on EEG, is also seen associated with SUDEP. These cardiac, imaging, and EEG biomarkers are seldom used in clinical practice because risk is more readily characterized by clinical factors, such as frequency of grand mal seizures (Epilepsia. 2013;54:e127-30).

credit: Child Neurology Foundation

SUDEP in children: A window on heterogeneity Dr. Donner, who has studied SUDEP in children for nearly two decades, was the lead investigator on one of two recent studies that found a much higher incidence of pediatric SUDEP than was known at the time of the AAN/AES guideline. “For a long time, we’ve seen quite significant barriers to case identification in that, when you review death certificates, many deaths are not identified as SUDEP,” she said. “The fact that the deceased had epilepsy often didn’t make it onto the death certificate. The term SUDEP for sure didn’t make it on, and if there’s any other comorbid condition, the death was often attributed to that.” Dr. Donner and colleagues were among the researchers contributing to the earlier incidence estimates of about 1 in 4,500 patient-years for children with epilepsy, a figure which derived from searching public autopsy records. “I knew even then that this had to be an underestimate,” she said. Fortunately, this is changing, Dr. Donner said, citing a recent white paper from the National Association of Medical Examiners with recommendations for the investigation of deaths in people with epilepsy (Epilepsia. 2018 Mar;59[3]:530-43). In 2018, Dr. Donner and colleagues scoured for cases in data from a pediatrician surveillance program, child neurologist reports, and forensic autopsies in Ontario. They found SUDEP deaths among patients younger than 18 years to be 1.11 per 1,000 patient-years, similar to adult rates (Neurology. 2018 Jul;91[2]:e107-11). The findings dovetailed with those from a large population-based study from Sweden the previous year that also found SUDEP incidence among people younger than 16 years to be 1.11 per 1,000 patient-years (Neurology. 2017 Jul;89[2]:170-7). Specific risk factors for children and adolescents remain to be defined, although frequent tonic-clonic seizures are considered a top risk in children, as well as adults. Whether the kind of cumulative damage to brain structures identified in adults contributes to SUDEP in children also remains to be studied. Some severe childhood-onset syndromes, such as Dravet syndrome, may be associated with a higher risk of SUDEP. In one unusual case study published by Dr. Donner’s research group, an 8-year-old girl with drug-resistant seizures experienced “near-SUDEP” while undergoing EEG monitoring. Though she experienced depressed breathing and brain activity and required resuscitation, the event was not preceded by a seizure (Epilepsia Open. 2018;3[1]:98-102). The more researchers learn about SUDEP, “we’ll find that not all SUDEPs will be the same,” Dr. Donner predicted. Rather, the understanding of what happens when a person dies of SUDEP “will be heterogeneous, the way epilepsy itself is heterogeneous. You can have epilepsy due to genetic factors, for example, or due to a stroke or other brain injury.” While some people who die of SUDEP “are on many meds at the same time, have many seizures every day, have brain malformations and very complicated epilepsy, there are other people who die of SUDEP who have relatively uncomplicated epilepsy, who are taking one med, who are complying with their regimens.” A number of prevention strategies have been proposed for SUDEP in children and adults, including nocturnal monitoring devices and having patients sleep accompanied. The Danny Did Foundation publishes information on commercial devices available to help alert patients and family to seizures, which may help reduce the risk of SUDEP. But good seizure control is, for now, the only proven way patients and physicians can collaborate to reduce SUDEP risk. Carefully underscoring the seriousness of that risk, rather than downplaying it, may be the place to start. credit: Phanie/Science Source