Revision as of 21:07, 11 July 2017

Asadi-Pooya AA and Sperling MR (2009) Clinical features of sudden unexpected death in epilepsy. J Clin Neurophysiol Uno:Ein (E–pub)

Link to Article

Abstract: People with epilepsy may die unexpectedly without a clear structural or pathologic cause. This condition is called sudden unexpected death in epilepsy (SUDEP), and it accounts for a large proportion of deaths among people with epilepsy. SUDEP incidence rates vary with the cohort studied, ranging from 0.35 per 1,000 person-years of follow-up in population-based studies to 9.3 per 1,000 person-years in patients with refractory epilepsy. Although many studies have been performed, the causes of SUDEP are not understood. However, even without precise knowledge of the underlying pathogenic mechanism(s), SUDEP prevention could start with the identification of the most prominent risk factors. SUDEP seems to occur more commonly during sleep and it preferentially affects young adults with medically intractable epilepsy (especially tonic-clonic seizures), individuals who also have neurologic comorbidity, and patients receiving antiepileptic drug polytherapy. This article reviews the clinical features associated with SUDEP and suggests preventive measures for this condition.

Keywords: incidence, prevention, risk factors, SUDEP

Comments and Context

A general review, focusing on possible preventive measures.

@@ Line 1: / Line 1: @@
 ''Asadi-Pooya AA and Sperling MR (2009) Clinical features of sudden unexpected death in epilepsy. J Clin Neurophysiol Uno:Ein (E–pub)''
+'''[https://insights-ovid-com.ezp.welch.jhmi.edu/pubmed?pmid=19713869 Link to Article]'''
 '''Abstract:''' People with epilepsy may die unexpectedly without a clear structural or pathologic cause. This condition is called sudden unexpected death in epilepsy (SUDEP), and it accounts for a large proportion of deaths among people with epilepsy. SUDEP incidence rates vary with the cohort studied, ranging from 0.35 per 1,000 person-years of follow-up in population-based studies to 9.3 per 1,000 person-years in patients with refractory epilepsy. Although many studies have been performed, the causes of SUDEP are not understood. However, even without precise knowledge of the underlying pathogenic mechanism(s), SUDEP prevention could start with the identification of the most prominent risk factors. SUDEP seems to occur more commonly during sleep and it preferentially affects young adults with medically intractable epilepsy (especially tonic-clonic seizures), individuals who also have neurologic comorbidity, and patients receiving antiepileptic drug polytherapy. This article reviews the clinical features associated with SUDEP and suggests preventive measures for this condition.
-=Article=
+'''Keywords:''' incidence, prevention, risk factors, SUDEP
-People with epilepsy, particularly when their seizures are not controlled by medication, sometimes die unexpectedly without a clear structural or pathologic cause. This phenomenon is called sudden unexpected death in epilepsy (SUDEP), and it accounts for a large proportion of deaths among people with epilepsy, especially young people. The incidence of sudden unexplained death among young adults with intractable epilepsy is many times higher than that of the general population (Annegers, 1997; Forsgren et al., 2005b; Lhatoo and Sander, 2005; Nashef et al., 2007). This article focuses on the clinical aspects of SUDEP and reviews the incidence rate, risk factors, and preventive measures.
-==Case Report==
-A 32-year-old man began having seizures at age 12 years. He had both complex partial and secondarily generalized seizures that usually began with an aura of dizziness. Seizures had occurred sporadically in his teen years and then increased to a rate of 3 to 5 per month by age 25. Remission did not occur with therapeutic doses of phenytoin, carbamazepine, lamotrigine, and topiramate. He had a history of complex febrile convulsion at the age of 18 months but no other medical illnesses. He was employed as an electronics technician, was married, and had three young children. History and physical examination findings were otherwise unremarkable except for mild memory difficulties. Outpatient EEG showed right frontotemporal sharp waves, and MRI was normal. Treatment options were discussed during his first visit at the epilepsy center. He declined surgical evaluation, opting to try another medication first. He was instructed to take levetiracetam, 1,000 mg daily, in addition to lamotrigine, which he was taking at the initial interview. He called 6 weeks later and reported continued seizures. The levetiracetam dose was increased to 2,000 mg daily.
-The patient’s wife called 4 weeks later and informed us of his death. After having a typical complex partial seizure at 1 AM, he had gone back to sleep. She awakened normally at 7 AM and noticed that he was not breathing. He was pronounced dead in his bed by paramedics. Other than the epilepsy, he had been in good health before dying.
-==Definition==
-Definite SUDEP is defined as the sudden and unexpected death of a patient with epilepsy when the death is witnessed or unwitnessed, nontraumatic and nondrowning, with or without evidence of a seizure (excluding documented status epilepticus), and without an identified structural or toxicologic cause after postmortem examination (Nashef, 1997). When an autopsy has been performed, the diagnosis is considered definite SUDEP. When an autopsy has not been performed, sudden death occurring in benign circumstances with no known competing cause for death is classified as probable SUDEP. In other words, probable SUDEP cases meet all the criteria of definite SUDEP but lack postmortem data. Possible SUDEP includes cases in which SUDEP cannot be ruled out, but there is insufficient evidence about the circumstances of the death and no postmortem report is available. Because many patients who die do not have a postmortem examination, investigations of SUDEP usually must include definite SUDEP cases and probable SUDEP cases (Annegers, 1997). Without an autopsy, the patient described in the above case fulfills the criteria for probable SUDEP.
-==Incidence==
-The incidence of sudden unexplained death has been studied in highly selected cohorts of patients with epilepsy. Therefore, the reported SUDEP incidence rates vary with the cohort studied and range from 0.35 per 1,000 person-years of follow-up in the population-based studies (Ficker et al., 1998) to 9.3 per 1,000 person-years in surgical candidates with refractory epilepsy (Annegers and Coan, 1999; Dasheiff, 1991; Jick et al., 1992; Klenerman et al., 1993; Lip and Brodie, 1992; Nashef et al., 1995b; Sperling et al., 1999; Tellez-Zenteno et al., 2005; Tennis et al., 1995) (Fig. 1). SUDEP accounts for almost 9% of the deaths of patients with epilepsy who die between ages 15 and 44 years, although the rate is as high as 50% in series reporting mortality among patients with refractory epilepsy after epilepsy surgery (Ficker et al., 1998; Sperling et al., 1999). SUDEP is thought to be relatively uncommon in children, but studies are limited and most involved only small numbers of patients (Nashef et al., 2007). In one study, four cases of SUDEP were identified during an 18-year period. The incidence of SUDEP was 0.4 per 1,000 patient-years. All patients received polytherapy and had refractory epilepsy, developmental retardation, and early-onset epilepsy. Children with well-controlled epilepsy or absence seizures seem not to be at risk (Weber et al., 2005).
-The prevalence of active epilepsy in Europe, the United States, and Asia is about 6 per 1,000. The prevalence is about 15 per 1,000 in sub-Saharan Africa and 18 per 1,000 in Latin America (Burneo et al., 2005; Forsgren et al., 2005a; Mac et al., 2007; Pruex and Druet-Cabanac, 2005). About 60% of patients with newly diagnosed epilepsy have a good response to antiepileptic drugs (AEDs) and become seizure free on a modest or moderate dose of the first or second choice of AED without intolerable side effects. However, about 40% of patients have epilepsy that is difficult to control (Kwan and Brodie, 2000). Therefore, in the total population, the risk of SUDEP is about 0.024% in developed countries and triple this percentage in developing countries; the incidence of SUDEP among patients with uncontrolled epilepsy is estimated to be approximately 5 to 10 per 1,000 person-years. The overall risk of sudden, unexplained death among young people with epilepsy, particularly among patients with intractable seizures, markedly exceeds that of the general population by approximately 24 to 40 times (Annegers and Coan, 1999; Ficker et al., 1998).
-==Risk Factors for SUDEP==
-SUDEP might be understood and prevented with the identification of the most prominent risk factors. Several studies have investigated the clinical characteristics of epilepsy patients who have died of SUDEP. The ascertained risk factors have varied depending on the methodology used, but some findings have emerged that are consistent among studies.
-The most common risk factor for SUDEP is the presence of uncontrolled tonic-clonic seizures. An analysis of risk factors for SUDEP (Tellez-Zenteno et al., 2005) has shown that in studies that used non-SUDEP deaths as controls (seven studies), the most consistent risk factors were the occurrence of a seizure preceding death and subtherapeutic levels of AEDs. SUDEP has been associated with younger age, use of a high number of AEDs, alcohol use, and perhaps a diagnosis of primary generalized epilepsy. Similar risk factors emerged in five studies that used patients living with epilepsy as controls. High-seizure frequency, use of a high number of AEDs, youth, long duration of epilepsy, use of psychotropic drugs, male sex, and diagnosis of primary generalized epilepsy were implicated in SUDEP (Tettez-Zenteno et al., 2005). The authors concluded that the type of risk factors uncovered by different studies depends on the questions asked. For example, comparisons with non-SUDEP deaths in epilepsy explore best the circumstances surrounding death (e.g., seizures preceding death, AED levels at the time of death), but comparisons with patients living with epilepsy explore best the lifestyle and clinical variables that may contribute to SUDEP (e.g., frequency and type of seizures, number of AEDs, use of other drugs). Therefore, different risk factors uncovered by studies with different comparators are not necessarily contradictory but rather complementary to the complex clinical profile of this condition.
-Finally, neurologic comorbidity is a clear risk factor for excess mortality in epilepsy, most likely including SUDEP (Forsgren et al., 2005b). Patients with epilepsy who also have mental retardation or cerebral palsy have significantly higher mortality rates than those with epilepsy alone. This could be related to a greater chance that these individuals are apt to have more severe epilepsy, or it could be related to an increased burden of neurologic disease.
-Major risk factors for SUDEP are summarized in Box 1. Evidence suggests that SUDEP preferentially afflicts young adults with medically intractable epilepsy and mental retardation who take multiple AEDs, experience frequent seizures, and have low AED serum levels (Nilsson et al., 1999; Tellez-Zenteno et al., 2005). Indeed, death in bed is a relatively common finding in SUDEP, with patients usually found in the prone position. The seizure type affects the risk of SUDEP, with the greatest risk ascribed to generalized tonic-clonic seizures (Tomson et al., 2005), although SUDEP has been reported with all seizure types except absence seizures. Other contributing factors are summarized in Box 2. The presence of a structural brain lesion, comedication with psychotropic drugs, and lack of postictal stimulation may increase risk. Because primary hypoventilation has been postulated as a possible mechanism, stimulation might cause breathing to resume, and lack of stimulation might lead to respiratory arrest. Seizure frequency is probably related to the risk of SUDEP (Nilsson et al., 1999), although there may be a ceiling effect that makes it difficult to characterize the relationship between seizure frequency and risk (Table 1). It should be noted that the above risk factors have not been confirmed by all studies (Hitiris et al., 2007), and caution should be exercised in interpreting the data.
-==Prevention of SUDEP==
-===Seizure Control===
-The most important preventive measure seems to be preventing seizures. Because SUDEP seems to occur in patients who continue to experience seizures, stopping seizures should be the best method of preventing SUDEP. Several approaches can be considered.
-===Drug Therapy===
-Drug treatment can be optimized to prevent seizures in many individuals. Putative risk factors relating to AED treatment, such as using appropriate medication, maintaining therapeutic drug levels, ensuring patient adherence to treatment, and avoiding polytherapy, are important and amenable to manipulation during routine management. Response rates to certain AEDs clearly differ depending on the underlying epilepsy syndrome, so it is most important to ensure that the correct AED has been prescribed. For example, the patient with juvenile myoclonic epilepsy who continues to experience seizures while taking a sodium channel blocker (e.g., carbamazepine or lamotrigine) should try valproate, which more often is effective for this condition.
-Subtherapeutic postmortem levels of AEDs and variable AED ingestion over time have been reported in SUDEP cases and taken as indicators of possible poor medication adherence (Leetsma et al., 1989; Williams et al., 2006). However, the finding of low serum levels in patients who died of SUDEP depends on the questions asked in any particular study, and noncompliance is not a consistent finding (Tellez-Zenteno et al., 2005). This issue is difficult to study, because accurate information about medication adherence shortly before death is not usually available. In addition, low postmortem phenytoin concentrations do not necessarily imply poor compliance. In an animal study, blood concentrations of carbamazepine were stable for 72 hours after death, but postmortem phenytoin blood levels decreased to 35% of the antemortem serum concentrations (Tomson et al., 1998). Consequently, low serum levels of phenytoin at postmortem examination may reflect instability of the drug concentration rather than poor compliance (Leetsma et al., 1989). However, one study documented greater variability of AED concentration in hair at autopsy, reflecting variable AED ingestion over time in patients who died of SUDEP compared with epilepsy outpatients and inpatients (Williams et al., 2006). Therefore, despite these limitations, careful monitoring of drug compliance and serum levels in patients with recurrent seizures may lead to changes in patient behavior and improved seizure control and may prevent SUDEP.
-Whether certain medications increase the risk of SUDEP is unclear. One study suggested that carbamazepine might increase the risk of SUDEP (Nilsson et al., 2001), but this finding has not been reliably reproduced. There is insufficient evidence to suggest that any particular AED might increase or decrease the risk of sudden unexpected death (Lathers and Schraeder, 2002; Walczak et al., 2001).
-Whether polytherapy with AEDs is a true risk factor is not certain. Although two cohort-based case-controlled studies found that treatment with more than two AEDs is a risk factor for SUDEP even after adjustment for seizure frequency (Langan et al., 2005; Nilsson et al., 1999), other unexamined factors may have been responsible for the increase in death rate, and increased risk has not consistently been associated with polytherapy (Hitiris et al., 2007). Polytherapy is usually prescribed when seizures are poorly controlled, and any risk associated with polytherapy may reflect differences in the severity of underlying illness rather than a causal effect of polytherapy. However, because AEDs might affect cardiac conduction and autonomic function, polytherapy may increase the risk of SUDEP. Until definitive information is available, prudence dictates exercising caution and avoiding the prescription of more than one drug.
-Overall, although medical treatment has the potential for side effects and is not free of risk, the evidence suggests that the risk-benefit ratio favors AED treatment, and there is an increased risk of SUDEP among persons never treated (Tomson et al., 2005). Improving seizure control with one or two appropriate AEDs while avoiding polytherapy with more than two AEDs (if possible) may be a way to reduce the risk of SUDEP.
-===Epilepsy Surgery===
-Because recurring seizures are the most important risk factor for SUDEP, all treatments that have the potential to stop seizures should be used. If two or three appropriate drugs have failed, a trial of additional drugs is unlikely to induce seizure remission (Kwan and Brodie, 2000) and a different therapeutic approach should be entertained. First, diagnostic testing should be performed to verify the diagnosis of epilepsy and to ensure that the patient does not have psychogenic seizures, a different type of seizure disorder that requires a novel approach to medical therapy, or some other medical disorder (e.g., cardiogenic seizures). If drug-resistant epilepsy is confirmed, several treatment options exist. Epilepsy surgery is the only nonmedical treatment that offers a good chance of stopping seizures completely, and the other options are generally palliative. Epilepsy surgery usually involves either excising the cortical area that causes seizures or disconnecting the epileptogenic regions of the brain from other areas. Removal of part of the temporal lobe (anterior temporal lobectomy or anteromedial temporal resection) is the most commonly performed operation, but surgery can be carried out in all other lobes of the brain as well. Disconnection procedures, such as a corpus callosotomy, also can be done but are nearly always palliative. Vagus nerve stimulation and the ketogenic diet are also mainly palliative treatments and are unlikely to completely and permanently abolish seizures.
-Successful epilepsy surgery is associated with decreased mortality from all causes and seems to decrease the incidence of SUDEP in medically refractory epilepsy. In two large studies, initially of 393 patients (Sperling et al., 1999) and later extended to include 583 patients (Sperling et al., 2005), mortality was decreased among patients who were seizure-free after epilepsy surgery. Eighteen late deaths unrelated to surgery occurred in patients with recurrent seizures, whereas only one death occurred among patients who were seizure-free after surgery. Mortality from all causes was decreased and SUDEP was observed only among patients who had persistent seizures after surgery. Even patients with rare seizures (a few per year or less) were at risk for SUDEP. In contradistinction, SUDEP did not occur after successful surgery. Mortality was indistinguishable between seizure-free patients and the general population, whereas the standardized mortality ratio was 4.6 (95% confidence interval [CI], 2.5–8.2) for patients with recurrent seizures; half of late postoperative deaths were attributed to SUDEP. In these two studies, the mortality rate for patients who continued to experience seizures after surgery was similar to the mortality rate for patients with medically refractory seizures who did not have surgery. Another study of 299 patients who had temporal lobe epilepsy surgery also showed that surgery decreased the overall mortality associated with chronic epilepsy (Hennessy et al., 1999). Whether surgery truly decreases the excess mortality associated with epilepsy can still be questioned (Ryvlin et al., 2006). The predictors of surgical failure in temporal lobe epilepsy, such as tonic-clonic seizures, might also be risk factors of SUDEP. Thus, much of the preoperative SUDEP burden observed in refractory epilepsy could be carried by patients for whom surgery will eventually fail, whereas patients who will be cured surgically may have a low risk of SUDEP even before the operation. This is a reasonable possibility, but the similarity in mortality rates between surgical failures and medically treated patients is consistent with the concept that surgery can decrease mortality from SUDEP in epilepsy.
-One study suggested that vagus nerve stimulation may decrease SUDEP (Annegers et al., 2000). After device insertion, 1,819 individuals were observed for 3,176.3 person-years. When the vagal nerve stimulation experience was stratified by duration of use, the rate of SUDEP was 5.5 per 1,000 during the first 2 years but only 1.7 per 1,000 thereafter. However, seizures persist after vagus nerve stimulator use; persistent remission is rare. Hence, this finding requires confirmation because it defies other research findings.
-===Modification of Sleep Position===
-In one study, many SUDEP victims died in the prone position (Kloster and Engelskjon, 1999), and some investigators have questioned whether a prone position predisposes to suffocation from the pillow. Although change in sleep position from the prone to the supine position has decreased sudden infant death syndrome, it is not known whether avoidance of the prone position reduces SUDEP risk, and formal studies of sleep position are warranted. Nonetheless, lacking firm data, it may not be unreasonable to advise patients to try to avoid sleeping in the prone position.
-===Supervision===
-Although most deaths are unwitnessed, studies suggest that supervision and attention to recovery after a seizure could help prevent SUDEP. A case-control study reported 154 cases of SUDEP in which a postmortem examination was performed (Tomson et al., 2005). Each case had four controls with epilepsy from the community, matched for age and geographic location. The risk of SUDEP was increased with a history of generalized tonic-clonic seizures in the previous 3 months (odds ratio, 13.8; 95% CI, 6.6–29.1). The presence of supervision at night (defined as the presence of an individual of normal intelligence at least 10 years old in the same bedroom) or the use of special precautions (regular checks throughout the night or the use of a listening device) was found to be protective. The odds ratio was 0.4 (95% CI, 0.2–0.8) when a supervising individual shared the same bedroom and 0.1 (95% CI, 0.0–0.3) when special precautions were used. Supervision has emerged as a protective factor independent of seizure control, suggesting that it is not simply a surrogate marker for epilepsy control.
-Another study involved a residential school for children with severe epilepsy and learning difficulties and included 310 patients (Nashef et al., 1995a). The follow-up period (4,135 person-years) included time in residence at the school as well as time after leaving. The age- and sex-standardized overall mortality ratio was 15.9 (95% CI, 10.6–23.0), with 20 of 28 deaths considered to be related to epilepsy. The incidence of sudden death was 1 in 295 per year. All 14 sudden deaths occurred when the pupils were not under close supervision, and most sudden deaths were unwitnessed.
-Further studies are needed to determine the effectiveness of supervision or warning devices in preventing SUDEP. However, caregivers should be advised of the actions to take if they witness a seizure, such as placing the person in the recovery position, stimulating the person, and checking on respiration. Most sudden epilepsy deaths are unwitnessed. When witnessed, most occur in association with a generalized tonic-clonic seizure, and respiratory compromise is a prominent feature. Positioning or stimulation of respiration after a seizure may help to prevent death (Langan et al., 2000).
-==Counseling Patients and Families==
+=Comments and Context=
-One final issue remains to be addressed: what should patients and their families be told? Should people with epilepsy be informed about the possibility of sudden death? It seems reasonable that patients should be educated about their illness and told about potential complications. However, should a physician discuss a condition like SUDEP, whose occurrence cannot be predicted, whose etiology is unknown, and for which no proven treatments exist? The National Institute for Health and Clinical Excellence (NICE, 2007) guidelines from the United Kingdom state the following: “Tailored information on the individual’s relative risk of SUDEP should be part of the counseling checklist for people with epilepsy and their families and/or carers.” However, in a study in the United Kingdom, among all neurologists only 5% discussed SUDEP with all patients, 26% with a majority, 61% with a few, and 7.5% with none (Morton et al., 2006). The commonest reasons for SUDEP to be discussed were the patient asking about it and the neurologist counseling people with known risk factors for SUDEP. Because the risk of SUDEP is probably not increased in patients with newly diagnosed epilepsy who respond well to treatment (Mohanraj et al., 2006), especially with AED monotherapy, SUDEP need not be discussed with this group of patients unless they specifically ask about it. However, when seizures fail to respond to medical therapy, it is reasonable to discuss the mortality risks and preventive strategies with patients or caregivers (or both). This may help patients decide whether to pursue surgical treatment and may encourage medication adherence.
+*A general review, focusing on possible preventive measures.

Difference between revisions of "Clinical features of sudden unexpected death in epilepsy"

Revision as of 21:07, 11 July 2017

Comments and Context

Navigation menu

Search