Continuing Education Activity

Sudden unexpected death in epilepsy (SUDEP), defined as “death in a patient with epilepsy that is not due to trauma, drowning, status epilepticus, or other known causes but for which there is often evidence of an associated seizure,” represents a leading cause of death in patients with epilepsy. This activity reviews the definition, epidemiology, physiology, risk factors, and treatment of sudden unexpected death in epilepsy (SUDEP) and highlights the role of the interprofessional team in stratifying and mitigating the risks of SUDEP in patients with epilepsy.

Objectives:

• Describe the definition of sudden unexpected death in epilepsy.
• Identify the modifiable risk factors for sudden unexpected death in epilepsy.
• Summarize the findings leading to a diagnosis of sudden unexpected death in epilepsy.
• Review strategies employed by an interprofessional team to modify sudden unexpected death in epilepsy risk in high-risk patients.

Introduction

Epilepsy, defined as the tendency for unprovoked seizures, is a common neurologic disease affecting up to 1 in 26 individuals.[1] While the spectrum of comorbidity varies significantly among patients with epilepsy, there is an overall increased mortality risk in this population. Causes of death in individuals with epilepsy include non-epilepsy-related conditions such as suicide, cancers, and cardiovascular disease, as well as epilepsy-related provoked causes, including status epilepticus, anti-seizure drug effects, motor vehicle and bicycle accidents, alcohol withdrawal, aspirational pneumonia, and drowning.[2]

Sudden unexpected death in epilepsy (SUDEP) is defined as “death in a patient with epilepsy that is not due to trauma, drowning, status epilepticus, or other known causes but for which there is often evidence of an associated seizure” and represents a leading cause of death in patients with epilepsy.[3] A study on childhood-onset epilepsy long-term mortality showed that out of 60 deaths, 33 (55%) were related to epilepsy, including sudden, unexplained death in 18 subjects (30%). While the risk of sudden, unexplained death was 7% at 40 years.[4] As SUDEP is often unwitnessed, diagnosis is typically made post-mortem. Efforts to distinguish SUDEP from other causes of mortality in epilepsy have led to the sub-categorization of SUDEP into four major categories:

1. Definite,
2. Probable,
3. Possible,
4. Unlikely SUDEP[3] 

These categorizations have allowed for more precise investigation regarding the risks, mechanisms, and preventative strategies. Increasing awareness of SUDEP amongst health professionals and the general public can allow for improved patient counseling and potential modification of risk factors. 

Etiology

Several potential mechanisms have been studied concerning the pathogenesis of sudden unexpected death in epilepsy. Retrospective observational patient data, hospital data from patients with sudden unexpected death in epilepsy in hospital epilepsy monitoring units (EMU), importantly, the MORTEMUS trial (MORTality in Epilepsy Monitoring Unit Study), as well as eyewitness accounts, have identified respiratory abnormalities, cardiac arrhythmia, and attenuation of cerebral function as commonly observed peri-ictal phenomena.[5] 

Multiple studies have evaluated respiratory factors that may contribute to SUDEP. In most reported SUDEP cases, the patient was discovered in a prone position, likely during sleep, suggesting suffocation [5]. MORTEMUS trial reported 29 cardiorespiratory arrests, including 16 SUDEP (14 at night), nine near SUDEP, and four deaths from other causes. It also concluded that EMU patients should have close monitoring at night as cardiac arrhythmia and respiratory distress can occur in the early postictal state leading to SUDEP.

Although cardiac arrhythmias have also been identified, signs of respiratory abnormalities (apnea, abnormal breathing patterns, hypoxia) are more commonly associated with brain dysfunction and were more commonly observed in witnessed and monitored cases of SUDEP. As the heart rhythm is generated independent of brain function, cardiac arrhythmia present in cases of SUDEP may suggest a secondary mechanism of the disease, such as channelopathy.[6] Furthermore, in cases where video electroencephalography (EEG) monitoring data was available, clinically significant arrhythmias were not identified.[5] Cerebral suppression has also been suggested as the primary event in a cascade terminating in SUDEP. Generalized suppression of brain function includes suppression of respiratory centers in the brainstem, resulting in severe acidosis. This suppression will also result in hypercapnia and acidosis, and laryngospasm.[7]

Epidemiology

Sudden unexpected death in epilepsy has been reported in many populations, from young children to the elderly. The highest prevalence of SUDEP is in people between the ages of 20 and 45.[8][9] A higher odds ratio for SUDEP has been reported in patients seen in an epilepsy center, as well as those with refractory epilepsy, those cared for in a residential care facility, and patients referred for epilepsy surgery. There is a higher odds ratio for SUDEP in males (OR = 1.42) and people with childhood-onset epilepsy.[10] 

Large-scale studies from the USA and Europe have identified higher rates of SUDEP in populations with socioeconomic barriers to care, including lack of employment, lack of access to medications and other treatments, and increased distance from appropriate healthcare providers.[9] SUDEP has been described in a variety of epilepsies, including generalized and focal epilepsy types, but based on recent population-based studies, an increased risk has not been observed in patients purely with absence or myoclonic seizures.[11]

A study by Sveinsson et al. in Sweden showed that patients with generalized tonic clinic seizures (GTCS) had a tenfold increased risk of SUDEP, and the majority of patients lived alone. A twofold increased risk of SUDEP was seen in individuals with a previous diagnosis of substance abuse or alcohol dependence.[12]

Pathophysiology

Although the exact mechanisms of sudden unexpected death in epilepsy are unclear, multiple mechanisms have been proposed, including electroencephalographic suppression (arousal dysfunction) after a generalized seizure and secondary cardiac arrhythmias (bradycardia and asystole) postictal apnea.[13] The generally accepted "final common pathway" for SUDEP is profound cardiopulmonary depression. Inpatient peri-ictal monitoring has provided valuable data regarding physiologic changes in the peri-ictal period, including oxyhemoglobin desaturation (as low as 40%) and severe bradycardia.[8]

Various hypotheses have been proposed which implicate postictal dysfunction of arousal centers in the brainstem, leading to secondary hypoventilation. The combination of hypercapnia and hypoxia accompanying postictal hypoventilation would potentially contribute to cardiac dysfunction and death.[5] Various risk factors, including channelopathies or genetic mutations causing epilepsy, are thought to increase a patient's susceptibility to this process.

Histopathology

Although the proposed mechanism of sudden unexpected death in epilepsy involves cardiorespiratory failure secondary to a seizure, relatively little is known regarding specific histologic changes occurring in patients who died from SUDEP. Identifying specific histochemical markers and other post-mortem findings may help distinguish SUDEP from death from alternative causes. Physical examinations of patients with SUDEP have revealed oral trauma (secondary to biting of the lip or tongue) and pulmonary edema.[14] 

Gross examination of the brain may reveal findings consistent with hypoxia, including mild gyral flattening consistent with brain edema; however, a significant mass effect has not been identified in SUDEP cases.[14] Various immunohistochemical staining methods have been employed to look for markers of inflammation, gliosis, disruption of the blood-brain barrier, and neuronal injury in patients with SUDEP vs. patients with non-epileptic sudden death. Such markers, including CD-163, HIF-(I-alpha), and HLA-DR reactivity in the medulla, hippocampus, and amygdala, have not shown significant differences in staining patterns compared with non-SUDEP controls.[7]

History and Physical

A number of historical factors have been associated with an increased or decreased incidence of sudden unexpected death in epilepsy. For example, the presence of generalized tonic-clonic seizures (GTCS), specifically greater than 3 in a year, are associated with 10 and 15-fold risk increases for SUDEP, respectively.[15] Furthermore, despite evidence that the addition of anti-seizure drugs in refractory epilepsy has a diminishing benefit with regard to seizure control, failure to add an anti-seizure drug increases the odds of SUDEP 6-fold.[15] In addition, a history of comorbid developmental delay and early-onset epilepsy have been identified as additional significant risk factors in the pediatric population.[16] 

Few factors are strongly associated with a lowered odds ratio for SUDEP. Two such factors include the presence of another individual in the bedroom who could act as an observer, as well as the use of a monitor/listening device in the room of the child or adult at risk for SUDEP.[15] Although most cases of SUDEP are unwitnessed, systematic evaluation of SUDEP cases occurring in epilepsy monitoring units has allowed for increased patient data collection in the peri-ictal period, allowing for easier identification of risk factors.[10]

The American Academy of Neurology (AAN) 2017 Practice Guidelines on SUDEP studied over 20 other risk factors for SUDEP, including sex, age of seizure onset, a specific type, dose and duration of antiepileptic drug usage, type of epilepsy, comorbid medical and psychologic disorders, imaging and EEG characteristics. None of these were strongly associated with an increased or decreased risk for SUDEP.[15]

Additional SUDEP risk factors under investigation include the time of day during which the seizure occurs, as well as physiologic changes that occur during different times of the day and body position (prone vs. supine).[17] Other non-modifiable risk factors include severe epilepsy and early age of onset of epilepsy and genetic factors, including pathogenic variants of potassium and sodium channels (such as KCNQ1, SCN1A, LQTS, KCNH2, and SCN5A).[18] Identification of KCNH2 and SCNC5 variants, which have been implicated in cases of long QT syndrome, is vital due to the potential for ensuring appropriate medication intervention.[19]

Evaluation

Sudden unexpected death in epilepsy, by definition, is a sudden and unexpected death in a person with epilepsy for which no apparent alternative cause has been found. Therefore, evaluation for alternative causes of death is mandatory in making a definitive SUDEP diagnosis.[20]

The diagnosis of SUDEP is based mainly upon history and postmortem examination findings, such as the patient being found in a prone position. The multiple potential confounding factors contributing to a patient's death make the diagnosis difficult, and accurate identification of a cause of death has medico-legal and public health implications.[21] A classification system has been proposed, considering historical points and comorbidities that may contribute to death in an epileptic patient. Devinsky et al. (2018) suggested an 8-tiered classification system:

1. Definite SUDEP: Sudden, unexpected death in a patient with epilepsy. The death may be witnessed or unwitnessed, and there does not need to be evidence of a recent seizure. Postmortem examination does not reveal an alternative cause of death besides epilepsy.
2. Definite SUDEP plus comorbidity: This is the same as definite SUDEP with a comorbid medical condition that could contribute to death (but not a competing cause).
3. Probable SUDEP: Unexpected death in a patient with epilepsy for whom an autopsy was not performed.
4. Probable SUDEP plus comorbidity: Same as probable SUDEP with comorbidity that could contribute to death (such as a channelopathy).
5. Possible SUDEP: Sudden death in a patient with epilepsy for whom there is a likely alternative cause of death (such as a patient found dead in the water, but no forensic confirmation of drowning made).
6. Resuscitated SUDEP: A patient with epilepsy who is resuscitated and survives greater than 1 hour after cardiopulmonary arrest. No alternative cause of the arrest is discovered after evaluation.
7. Not SUDEP: A clear alternative cause of death was identified.
8. Unclassified: There is insufficient information on which to make a diagnosis.[3]

Examination findings supporting a diagnosis of SUDEP include prone position, as well as signs of recent seizures, including oral trauma, ecchymoses, conjunctival hemorrhage, or lacerations to protruding areas of the head or limbs. These findings are non-specific but may support the diagnosis of a recent seizure.[22]

Treatment / Management

As sudden unexpected death in epilepsy is one of the most feared and catastrophic complications of epilepsy, options for mitigating or modifying the risk are important. Several interventions for decreasing the risk of SUDEP have been explored, including optimizing epilepsy control, detecting seizures through observation and devices, positioning to decrease the risk of airway compromise, and pharmacologic interventions to reduce the possibility of hypoventilation.[23] The most common interventions are the reduction of risk factors thought to contribute to SUDEP (including reducing seizure burden, increasing compliance with medication, and referring appropriate patients for more advanced treatments such as surgery). In recent years, seizure-monitoring devices (alarms, heart rate monitors, bed sensors) have been a source of hope for families of epileptic patients, and evidence is increasing regarding their use.[24] 

A 2012 Cochrane Review, however, has shown that while generalized tonic-clonic seizures (GTCS) may be accurately detected with such methods, there has not been enough data to suggest that they would prevent SUDEP. Similarly, the efficacy of interventions such as “safety pillows” or pharmacologic intervention with selective serotonin reuptake inhibitors to reduce central hypoventilation is not supported by data.[23]

Differential Diagnosis

The differential diagnosis for sudden unexpected death in epilepsy includes a number of alternative causes of death and typically includes entities that can be identified with an autopsy.[22] Common alternative causes of death in patients with epilepsy include:

• Drowning
• Cardiomyopathy
• Cardiac arrhythmia
• Drug overdose
• Pulmonary disease
• Trauma
• Intracranial hemorrhage

Prognosis

Although sudden unexpected death in epilepsy, by definition, is a fatal condition, the prognosis of patients with epilepsy, in general, can range from poor to excellent. The prognosis regarding developing SUDEP in a patient with epilepsy can be assessed using a variety of tools, including the SUDEP-7 inventory. This determines a patient's risk based on multiple factors, including:

• Three or more tonic-clonic seizures in the past year
• One or more tonic-clonic seizures in the past year
• One or more seizures of any type in the past year
• More than 50 seizures per month in the past year
• Duration of epilepsy greater than or equal to 30 years
• Treatment with three or greater anticonvulsants
• Intellectual disability with a measured IQ of less than 70[25]

The presence of any of the above risk factors increases the risk score (with a maximum score of 12). However, the individual risk score has not been validated as a means to predict SUDEP. In general, the chance of dying from SUDEP is 1 in 1000, and patients with poorly controlled epilepsy have a greater risk.

Complications

SUDEP is the most catastrophic complication of epilepsy. Other complications include injuries related to seizures, aspiration pneumonitis, drowning, status epilepticus, and anti-seizure medication toxicity. Patients with epilepsy are also at risk for neuropsychiatric disorders, including learning disabilities, and mood disorders, including depression. Furthermore, patients are at risk of decreased physical activity and social stigmatization.[26]

Deterrence and Patient Education

Although sudden unexpected death in epilepsy is a feared complication of epilepsy, identification and modification of risk factors are vital for prevention. Discussion with families of patients at risk for SUDEP is likely to alter patient behavior to lower SUDEP risk. Although discussion may be a source of anxiety for patients and caregivers, studies support the notion that patients prefer information about SUDEP.[27] Professional societies such as the United Kingdom National Institute for Clinical Excellence (NICE) and the Scottish Intercollegiate Guideline Network (SIGN) recommend discussing SUDEP with patients after initial diagnosis.[28] Several tools exist to help stratify patients’ risk of SUDEP, including The SUDEP-7 Risk Inventory.[25] These tools can aid in patient discussions and help improve outcomes.

Enhancing Healthcare Team Outcomes

The diagnosis of sudden unexpected death in epilepsy is not always straightforward, and definitive determination requires collaboration between neurologists and pathologists. While the neurologist is primarily responsible for treating epilepsy, the pathologist is critical in ruling out alternative diagnoses. The accurate diagnosis of SUDEP is vital for medicolegal reasons and can provide closure to patients' families.

Collaboration between physicians and patient advocacy groups is also essential to ensure accurate and accessible information is available to the public. There is ample data to inform patients of their overall risk of SUDEP, as well as evidence to suggest that ongoing seizure management in patients with uncontrolled epilepsy can mitigate SUDEP risk.[Level 2] Furthermore, some evidence exists to support nocturnal monitoring for seizure detection and SUDEP prevention. [Level 4]

Lastly, several studies have identified modifiable risk factors in SUDEP.[Level 1] Collaborative efforts between the epilepsy treatment team (including neurologists, neurology nurses, and social workers) along with patient advocacy groups (such as the Danny Did Foundation, Epilepsy Foundation) can help to properly disseminate information amongst the public and create behavioral changes in patients to mitigate risk.[15]