Neurologists have always faced tough decisions when it comes to sudden unexpected death in epilepsy (SUDEP). Not the least of these is how to explain to bereaved family members what happened to their loved one and how it could have been prevented, because no one really knows.
However, recent research is shedding light on the factors that may put epileptic patients at increased risk for SUDEP. This knowledge is helping clinicians make better decisions about how much to tell patients and their families about this phenomenon.
It's a delicate subject, and some patients do not want to know that they are at risk for sudden unexpected death. The question is, does knowing serve a purpose? "When I talk to my epileptic patients, I highlight the risk of injury and accident with potentially fatal consequences," said Michael Sperling, MD, the Baldwin Keyes Professor of Neurology at Thomas Jefferson University and director of the Jefferson Comprehensive Epilepsy Center in Philadelphia. Sperling advises patients, for example, to avoid rock climbing and not to swim without a lifeguard nearby. "But to say, 'By the way, you could suddenly die, we won't know why, and there's nothing I can do about it'—I don't think that serves any useful purpose," he said.
COMING INTO FOCUS
When clinicians such as Sperling do broach the subject with patients, they can at least draw on a better understanding of what SUDEP is. Research suggests that it may be the cause of death in 7% to 17% of all epileptic patients and in up to half of patients with refractory epilepsy.1 Criteria for establishing a postmortem diagnosis in a patient with known epilepsy include that death occurred suddenly (if observed)—or in any case, unexpectedly—without obvious medical cause, and that status epilepticus did not occur regardless of whether there was evidence of seizure.1 Almost all victims of SUDEP die at home, usually in bed.
Research is helping clinicians profile which patients are likely to be at highest risk. Important contributing factors include a history of uncontrolled seizures as well as seizure type (tonic-clonic seizures present the gravest danger and are associated with at least 90% of SUDEP cases; complex partial seizures also increase risk, but absence seizures do not). Increased seizure frequency also appears to elevate the hazard, although patients who experience tonic-clonic seizures even once or twice a year are in greater danger than patients with less serious seizure types.1,2
Evidence suggests that epilepsy duration heightens risk (most persons who die of SUDEP have had epilepsy for 15 to 20 years), but some clinicians—including Sperling—discount this factor. Age at onset and gender also may contribute; one study found that the relative risk (RR) of SUDEP was 7.7 among men in whom epilepsy developed during childhood or adolescence, which was higher than the RR among men in whom epilepsy developed after age 45 years.3
The age of patients in reported cases ranges from 8 months to 83 years, and pediatric cases are more common than is often assumed2; one community-based study identified 11 SUDEP cases among 20 deaths directly attributed to epilepsy in children aged 1 to 11 years.4 Although studies have drawn different conclusions, most researchers now consider those between age 20 and 40 years to be at highest risk.2 Patients with concurrent disorders—particularly mental retardation—are in greater peril as well. A suggested association with male sex appears to have been a result of weak statistical analysis.2
Studies of the association between SUDEP and pharmacotherapy—particularly multidrug approaches—have raised as many questions as they've answered. A Swedish study published in Lancet in 1999 found that the number of concomitant antiepileptic drugs (AEDs) taken was the second strongest risk factor for SUDEP. Compared with patients receiving monotherapy, the RR for SUDEP in patients who took 3 AEDs was 8.3 Others studies have reached similar conclusions, but clinicians disagree about the meaning of the results. "My suspicion is that these findings probably reflect the severity of the underlying epilepsy, and that the medication plays a minor role, if any," said Sperling.
Research into one AED—carbamazepine—-created another controversy in the late 1990s. A study published in Seizure in 1998 found that 11 (79%) of a total of 14 patients who fell victim to SUDEP while receiving treatment at an epilepsy clinic in Wales had been taking carbamazepine.5 The study authors examined possible causes and called for examination of the drug as a factor in SUDEP. In time, after further inquiry by various researchers, the possible connection was deemed "tenuous."6
There also had been suggestions that victims of SUDEP were more likely to be noncompliant with medication regimens. This theory, however, also was dismissed when an Australian team performed a comparative retrospective study of postmortem AED levels in 44 SUDEP cases and in 44 control cases that represented epileptics who died of causes other than epilepsy.7 The researchers reported that the plasma drug level profiles in both groups were the same and that there was no evidence to substantiate the hypothesis that poor drug compliance contributed to SUDEP.
If tracking SUDEP risk factors has been confusing, the challenge of finding the syndrome's roots presents another difficulty. Sperling, together with a team that includes Maromi Nei, MD, assistant professor of neurology at Jefferson Medical College in Philadelphia, has been trying to identify the underlying causes of SUDEP.
"I think SUDEP has more than one cause," Nei said. "A lot of data suggest that it could have a cardiac etiology in some individuals and a pulmonary one in others." Support for the latter theory comes from autopsy reports of lung congestion, but Nei acknowledges that these could simply be artifacts of the death process itself. For now, a promising target relates to problems with cardiac rhythm. "A variety of rhythm disturbances have been seen in people with epilepsy, ranging from atrial fibrillation to ventricular tachycardia to asystole, where the heart actually stops beating," Sperling explained.
For example, the team's 2004 paper, published in Epilepsia, retrospectively compared EEG and ECG records of patients who died of confirmed SUDEP (n = 6) or probable SUDEP (n = 15) with those of living patients with refractory partial epilepsy (n = 43).8 Those who died of SUDEP were more likely to have experienced seizures during sleep than those in the comparison group, and there was evidence of seizure within 8 hours of death in a third of those in the SUDEP group. Although rates of ictal tachycardia were only slightly higher among patients who died of SUDEP compared with patients who had refractory epilepsy (94% vs 84%), the mean maximal heart rate among these patients was significantly higher (149 vs 126 beats per minute). Of the 16 patients in the SUDEP group for whom ictal ECGs were available, 56% had ictal rhythm or repolarization abnormalities. By comparison, such abnormalities were detected in 39% of patients in the control population.
The researchers noted increased autonomic activity associated with seizures in those who later died of SUDEP. They concluded that several factors may have been related to SUDEP, primarily the sleep state and its associated physiologic changes, seizures (particularly generalized tonic-clonic seizures) that might directly have caused death, and seizure clusters. They also found that seizures during sleep could cause sudden and extreme fluctuations in autonomic tone (from predominant vagal tone to extreme sympathetic tone), which might precipitate lethal cardiac arrhythmias. Such effects would be greatest in younger patients, which may explain why most victims of SUDEP are relatively young.
In a similar vein, a British study, published in Lancet in 2004, that used loop-recorded ECGs to determine the frequency of cardiac arrhythmias in patients with refractory seizures found that those patients with certain peri-ictal cardiac abnormalities had clinical characteristics similar to those of patients at greatest risk for SUDEP. Researchers speculated further that asystole might underlie many instances of SUDEP.9
"It's possible that the rate of asystole occurring in conjunction with seizures is higher than we suspected," commented Nei. "But you can have asystole in other states such as sleep apnea, so it's complicated. I think a lot of things are connected but that we don't have a full answer."
As if the picture weren't adequately complex, one study reported 3 SUDEP cases, recorded during video/EEG monitoring, that suggested a mechanism of cerebral hypofunction leading to brain stem failure and the cessation of cardiorespiratory activity.10
For Sperling, a central issue concerns autonomic alterations during sleep. "There are changes in circulating hormonal levels, in heart rhythm, and in respiratory regulation," he said. "It's possible that in some individuals this is a cardiac problem and in others it's respiratory. It may also be a combination of factors." Sperling noted further that cardiac ischemic changes common to epilepsy might predispose those with uncontrolled seizures to fatal arrhythmias.
FINE, BUT WHAT DO YOU TELL PATIENTS?
All this research may eventually help you identify which of your patients are at highest risk for SUDEP, but the question remains: What are you going to do about it? As already noted, Sperling is disinclined to alarm his patients unless he sees good reason to do so. One solid justification for discussing SUDEP with a patient is if he or she is a candidate for surgery but is vacillating about the procedure.
"If someone has uncontrolled epilepsy and medicine isn't working, then they should think about alternatives such as surgical treatment," Sperling said. "It's a scary notion; you can go into surgery and die, but you can also die of epilepsy. The data suggest that surgery saves lives and that the risk is probably lower than the risk of uncontrolled seizures. I discuss SUDEP in that context to put things into perspective."
His colleague Nei is similarly reticent but believes that a frank talk about SUDEP is warranted if a patient isn't paying attention to drug regimens or otherwise seems overly blas about the situation. "If the person is noncompliant with medications, then they need to know what risks they are taking," Nei explained. "If the person asks me about all the risks of having seizures, I always talk about SUDEP as well, but if they're trying everything and they still have uncontrolled seizures, that's harder, because it can scare them. Without anything to recommend, it's a difficult situation."
Nei noted that the family members of patients who die of SUDEP occasionally express regrets. "They'll sometimes say they wish they'd known all the risks because then they would have been more careful about medications," Nei said. "I don't truly know if that would make a difference in those cases, but I think it's a natural human tendency to wonder if there was more they could have done." Nei typically briefs the parents of pediatric patients and the caregivers of mentally impaired patients about SUDEP to ensure that someone in the household is aware of the risk. In this regard, it should be noted that a fairly recent study by a team led by Yvonne Langan, MD, noted expert on SUDEP and consultant neurologist at the Royal Victoria Infirmary in Newcastle, UK, found that nighttime supervision protected patients against SUDEP.11
Elson So, MD, professor in the Department of Neurology at the Mayo Clinic College of Medicine in Rochester, Minnesota, also adapts his decision to the individual patient but favors providing the information. "Patients with a history of uncontrolled, generalized tonic-clonic seizures are at highest risk," So said. "Those are the ones I pay the closest attention to and counsel about SUDEP. I think they need to know."
So acknowledged that the information sometimes increases patients' anxiety but said that if it encourages medication compliance or serious consideration of surgery when it is warranted, it is worth it. "There also are practical measures that can be taken for some patients," he adds. "For example, people who have just had a generalized tonic-clonic seizure need prompt attention." In those with less severe epilepsy, however, So, like Sperling and Nei, is disinclined to raise the issue.
The SUDEP controversy has gained attention overseas, in part because countries with national health care systems have experienced clashes between government-promulgated guidelines and actual clinical practice.
In the United Kingdom, controversy erupted when the National Institute for Clinical Excellence decreed that all patients treated for epilepsy should be told about SUDEP at the time of diagnosis. Susan Duncan, MD, a consulting neurologist at the Greater Manchester Neurosciences Centre at Hope Hospital in Salford, Manchester, joined with colleagues to propose a clinical trial of this protocol by randomly assigning patients in whom epilepsy had just been diagnosed into 2 groups. One group was to be told about SUDEP at diagnosis and the other at follow-up several months later.
"We wanted to know whether telling everyone at diagnosis might raise more feelings of anxiety and depression," Duncan said. Unfortunately, the proposal was turned down by the hospital's ethics committee in part because of the opinion of an outside consultant who wrote that most neurologists in the United Kingdom already told their patients about SUDEP. "I said to the committee, 'I beg to differ; this is not true,'" Duncan said. "And I set out to prove that the majority of my colleagues in neurology do not tell their patients about SUDEP."
The result was a survey, to which 82% of consultant neurologists in the United Kingdom responded. It revealed that fewer than 5% of survey respondents consistently discuss SUDEP with all the patients they serve who are receiving treatment for epilepsy.12 Twenty-six percent discussed SUDEP with "a majority" of their patients, 61% with "a few" of their patients, and 7.5% with "none" of their patients. Physicians reportedly brought up SUDEP for some of the reasons already noted: if patients asked about risks directly, if they were at high risk, or if they were noncompliant with medications. Duncan noted that survey responses indicated a strong feeling that telling some patients about SUDEP was counterproductive.
"In the United Kingdom we have this notion that every clinical contingency should be covered by a policy," Duncan said. "I think that detracts from the true essence of medicine—that you treat each patient as an individual, not according to what the latest guidelines say." Australian neurologist Andrew Black, MD, recently wrote in the journal Medicine and Law that although raising the issue of SUDEP could cause patients anxiety, it could also save lives.13 "A choice will involve a judgment about how we perceive an optimal benefit or alternatively a minimal harm," Black wrote. "Central to this ethical question is the autonomy of the patient in deciding his or her course of action in pursuing a health outcome. No person can make such a choice [without] the necessary information." Black went on to note, however, that autonomy also implied that a patient might prefer not to have certain kinds of information.
Black offered several scenarios in which physicians face uncertainty about discussing SUDEP and made suggestions that generally agree with those already mentioned, indicating that a professional consensus may be forming. Black's recommendations include the following:
• At the first visit, when the patient is trying to come to terms with the epilepsy diagnosis and the volume of information given about his or her condition, the clinician should judge whether it is the appropriate time, based on the patient's specific needs and tolerance, to broach the question of SUDEP risk.
• Patients who ask directly about fatality risks should be told about SUDEP. If a patient is accompanied by a family member who asks, however, gauging whether the patient is prepared for an honest answer "can be tricky."
• Patients who are noncompliant with medications should probably be told about SUDEP in an effort to boost lagging motivation. By contrast, those with uncontrolled epilepsy who are careful about their medications may be needlessly upset by information about SUDEP.
• Clinicians should consider warning patients in long-term remission who are considering withdrawal from AEDs that withdrawal has been linked to increased risk of SUDEP. Again, the needs and tolerance of the individual patient must be assessed when deciding whether to broach the subject.
This more nuanced approach, reflecting real-world clinical practice, is likely to make sense to neurologists regardless of where they practice.
1. Sperling M. Sudden unexplained death in epilepsy. Epilepsy Cur. 2001;1:21-23.
2. Tomson T, Walczak T, Sillanpaa M, Sander JW. Sudden unexpected death in epilepsy: a review of incidence and risk factors. Epilepsia. 2005;46(suppl 11):54-61.
3. Nilsson L, Farahmand BY, Persson PG, et al. Risk factors for sudden unexpected death in epilepsy: a case-control study. Lancet. 1999;353:888-893.
4. Harvey AS, Nolan T, Carlin JB. Community-based study of mortality in children with epilepsy. Epilepsia. 1993;34:597-603.
5. Timmings PL. Sudden unexpected death in epilepsy: is carbamazepine implicated? Seizure. 1998;7:289-291.
6. Walczak T. Do antiepileptic drugs play a role in sudden unexpected death in epilepsy? Drug Saf. 2003;26:673-683.
7. Opeskin K, Burke MP, Cordner SM, Berkovic SF. Comparison of antiepileptic drug levels in sudden unexpected deaths in epilepsy with deaths from other causes. Epilepsia. 1999;40:1795-1798.
8. Nei M, Ho RT, Abou-Khalil BW, et al. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia. 2004;45:338-345.
9. Rugg-Gunn FJ, Simister RJ, Squirrell M, et al. Cardiac arrhythmias in focal epilepsy: a prospective long-term study. Lancet. 2004;364:2212-2219.
10. Bird JM, Dembny KAT, Sandeman D, Butler S. Sudden unexplained death in epilepsy: an intracranially monitored case. Epilepsia. 1997;38(suppl 11):S52-S56.
11. Langan Y, Nashef L, Sander JW. Case-control study of SUDEP. Neurology. 2005;64:1131-1133. 12. Morton B, Richardson A, Duncan S. Sudden unexpected death in epilepsy (SUDEP): don't ask, don't tell? J Neurol Neurosurg Psychiatry. 2006;77:199-202.
13. Black A. SUDEP—whether to tell and when? Med Law. 2005;24:41-49.
CARY GRONER is a freelance writer in northern California.