If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Epilepsy Monitoring Unit, Department of Neurology, Hospital del Mar, Passeig Marítim, 25-29, Barcelona, 08003, SpainIMIM (Hospital del Mar Medical Research Institute), Carrer del Dr. Aiguader, 88, 08003 Barcelona. Spain
Epilepsy Monitoring Unit, Department of Neurology, Hospital del Mar, Passeig Marítim, 25-29, Barcelona, 08003, SpainIMIM (Hospital del Mar Medical Research Institute), Carrer del Dr. Aiguader, 88, 08003 Barcelona. Spain
Epilepsy Monitoring Unit, Department of Neurology, Hospital del Mar, Passeig Marítim, 25-29, Barcelona, 08003, SpainIMIM (Hospital del Mar Medical Research Institute), Carrer del Dr. Aiguader, 88, 08003 Barcelona. Spain
This is the first case series of SUDEP performed in Spain.
•
SUDEP incidence was 1.3 per 1000 patient/year, comprising 0.56% of all deaths.
•
In Spain SUDEP is underdiagnosed and not recognized as a relevant clinical issue.
•
SUDEP risk biomarkers are not consistently assessed in our patients.
•
Health system and administrative issues difficult the register of SUDEP.
Abstract
Purpose
SUDEP is the first cause of mortality related to epilepsy. However, in Spain there are no published cases or series from Epilepsy Monitoring Units that could expose the characteristics of SUDEP in our population.
Method
We reviewed all patients treated at our Spanish Epilepsy Reference Centre who died between 2010–2018. SUDEP cases were classified as definite, probable, possible or near-SUDEP. Epilepsy type, demographics and case detection issues were described.
Results
From 1250 evaluated patients, 102 died during the study period. Seven patients were diagnosed with SUDEP or near-SUDEP: two definite SUDEP, one definite SUDEP plus, two probable SUDEP and two near-SUDEP. Specific problems for detection and registration of SUDEP inherent to the Spanish healthcare system and the legal framework were defined. Only 43% of cases were known by the referral neurologist. SUDEP incidence was 1.3 per 1000 patient/year, comprising 0.56% of all deaths in our cohort. Two cases were female, the average age was 36 years (18–61). All patients had focal epilepsy and suffered from generalized tonic-clonic seizures. All witnessed cases occurred after a focal to bilateral tonic-clonic seizure. Four cases occurred during sleep and all non-witnessed cases were found in prone position. One case occurred during video-EEG monitoring.
Conclusions
Our casuistic represents the first Epilepsy Monitoring Unit based case series of SUDEP conducted in Spain. The incidence in our population agrees with the reported in other countries. However, in our population, SUDEP is probably underdiagnosed due to administrative and legal issues.
Worldwide, almost 1% of the population suffers from epilepsy and its annual incidence for all ages is approximately 50–55 cases per 100,000 inhabitants [
Practice guideline summary: Sudden unexpected death in epilepsy incidence rates and risk factors: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology and the American Epilepsy Society.
]. There are different categories of diagnostic accuracy: possible, probable, definite SUDEP, near-SUDEP, and a subcategory "plus" when there is another possible cause of death but its implication has not been established [
]. Its incidence varies according to the group of patients analysed: between 0.1–2.3 cases per 1000 individuals/year in all cases; between 1.1–5.9 cases per 1000 individuals/year in patients treated in specialized epilepsy centres, many of whom have drug-resistant epilepsy; and between 6.3–9.3 cases per 1000 individuals/year in epilepsy surgery candidates, or those presenting persistent seizures after surgery [
The burden of premature mortality of epilepsy in high-income countries: a systematic review from the Mortality Task Force of the International League Against Epilepsy.
]. In a recent SUDEP case series analysis in our country, drawn from an epidemiological study of epilepsy-related mortality in a Spanish population (Málaga), the incidence of SUDEP was 0.56 per 1000 epileptic patients/year of follow-up [
]. However, due to the population analyzed, this report offered results similar to those of the general population and not to refractory epilepsies treated in Epilepsy Reference Centers, that is to say, individuals at increased risk.
The objective of our study was to analyse the characteristics of SUDEP and near SUDEP cases diagnosed in a tertiary epilepsy centre in Spain. We aimed to describe the epidemiological features and local constraints with respect to the identification and registering of SUDEP cases in the Spanish health system. To the best of our knowledge, this is the first analysis of SUDEP cases reported in an Epilepsy Reference Center in Spain.
2. Methods
2.1 Study design
An observational, retrospective study was performed. The publication of this manuscript was in accordance with the Clinical Research Ethics Committee of the IMIM-Hospital del Mar, Barcelona.
2.2 Study population
All patients attended for epilepsy between October 2010 and October 2018 in our Epilepsy Centre (Hospital del Mar, Barcelona) were included. This is a national reference centre (CSUR) for refractory epilepsy, nominated by the Spanish National Health Ministry and part of the European Reference Network (EPI-Care). In our Epilepsy Monitoring Unit (EMU), 24/7 surveillance is provided by trained nurses with exclusive dedication, in addition, an epileptologist is 24/7 on call. We offer specialized care for a reference population of 300,000 inhabitants and receive patients referred throughout Catalonia and Spain. The population base for this study were those patients from the centre who had died during this time period.
2.3 Definition of clinical variables
SUDEP was defined according to currently accepted criteria [
]: sudden, unexpected, witnessed/unwitnessed, nontraumatic and nondrowning death, occurring in benign circumstances, in an individual with epilepsy, with/without evidence for a seizure, and excluding documented status epilepticus in which post-mortem examination does not reveal cause of death. Classification was as following [
]: 1) Definite SUDEP: meets all criteria; 2) Probable SUDEP: meets all criteria, but lacks post-mortem data; 3) Possible SUDEP: SUDEP cannot be ruled out although there is insufficient evidence regarding the circumstances of death and no post-mortem report is available; 4) Near-SUDEP: an epileptic patient survives resuscitation for more than 1 h after a cardiorespiratory arrest that has no identified structural cause; and 5) The previous categories are classified as the subcategory “Plus” when evidence indicates that a pre-existing condition, established before or after the autopsy, could have contributed to death.
]: 1) At least two unprovoked (or reflex) seizures occurring >24 h apart; 2) One unprovoked (or reflex) seizure and a probability of further seizures ≥60% occurring over the following 10 years; and 3) Diagnosis of an epilepsy syndrome. The diagnosis of epilepsy in all the study participants was carried out in our outpatient clinic by certified epileptologists and posteriorly reviewed by at least two epileptologists (ASL, AP, ML, and RR). Patients who did not meet these criteria were excluded from the study.
Classification of epilepsies and seizures was performed according to the current guidelines (2017) of the International League Against Epilepsy (ILAE) [
Operational classification of seizure types by the international league against epilepsy: position paper of the ILAE commission for classification and terminology.
We retrospectively reviewed the electronic clinical records of all patients diagnosed with epilepsy attended in our Epilepsy Centre, either in the outpatient clinics and/or admitted for video-EEG monitoring (VEEGM). The following data were collected from all the participants: age, sex, epilepsy type, aetiology of epilepsy, treatment of epilepsy, and dead/alive status at the time of the review. All causes of death during the study period were evaluated by examining hospital reports, primary health systems, and emergency consultations. When an SUDEP case was suspected the patient's relatives were requested to respond to a telephone interview in order to ascertain the circumstances. According to the information obtained through the clinical records and the telephone interview, we confirmed or rejected SUDEP diagnosis. The following additional data were collected: age at seizure onset, age at death, family history of epilepsy or sudden death, personal history of seizure-related apnoea, seizure type, seizure frequency for each type in the previous year, and predominance of day/night-time seizures. EEG/video-EEG monitoring recording abnormalities, neuroimaging abnormalities, antiepileptic drug (AED) therapy, epilepsy surgery, comorbidities, comedication, risk factors for seizures (alcohol or other substance abuse, sleep deprivation, poor AED adherence), and the circumstances of death (approximate hour of death, awake or asleep when death occurred, seizure witnessed before death, place of death, situation found). For this study, if the death occurred out-of-hospital and a judicial autopsy was performed, we requested the patient's relatives’ permission to obtain the autopsy report through the Institutes of Legal Medicine and the judge responsible for each civil registry. In Spain, the cause of out-of-hospital death is determined by autopsy on the part of the local forensic pathology service. However, this evaluation is only carried out if the cause of death is not clear, depending on the opinion of the physician who completes the death certificate.
2.5 Statistical analysis
Descriptive analyses for all variables (means, standard deviation (SD) and frequencies) were evaluated. The proportional mortality for the different causes of death was calculated dividing the number of deaths by the number of patients with epilepsy in our cohort. We calculated the incidence of SUDEP as the number of SUDEP cases in our cohort during the observation time divided by individual-years at risk. All statistical procedures were performed using the Statistical Package for Social Sciences, Windows (SPSS 22.0. IBM, Chicago, USA).
3. Results
The cohort included 1250 patients whose characteristics are depicted in Table 1. There were 5203 individual/years follow-up during the study period and 102 participants died. The proportional mortality of the different causes of death is described in Table 2. Seven patients were diagnosed with SUDEP/ near-SUDEP which represented 6.8% of all deaths and affected 0.56% of all the participants. SUDEP incidence in our cohort was 1.3/1000 patients per year. It is notable that only 3 of the 7 cases of SUDEP (43%) were reported to the Epilepsy Unit since they were in the process of active pre-surgical diagnosis. The other 4 were identified only after a retrospective review of clinical records. In none of these cases was the information transmitted from the general practitioner to the reference hospital.
Table 1Characteristics of the study population (n = 1250 patients).
The epidemiological features and comorbidities of the SUDEP patients are described in Table 3. Two patients were female. The average age when SUDEP occurred was 36 years (range 18–61). The mean duration of epilepsy at the time of death was 16.6 years (range 4–51). None of the participants had a personal history of status epilepticus, cardiac arrhythmia, or family history of sudden death. One patient (case 7) had previously presented post-ictal apnoea. All the subjects suffered from focal epilepsy. Four cases had temporal lobe epilepsy, two frontal lobe epilepsy, and one posterior quadrant epilepsy. The aetiology of epilepsy was structural in 3 cases, immune in one case, and unknown in the remaining 3 (Table 3). It is noteworthy that some characteristics were shared by all seven patients including refractory focal to bilateral tonic-clonic seizures (FBTCS), the most frequent seizure type in all cases. All patients had at least one seizure per month whilst five (71%) suffered more than one. All of them presented nocturnal seizures but only one case evidenced a predominance of such distribution (Table 3).
Table 3Clinical characteristics of SUDEP and near-SUDEP cases.
The AEDs used at the moment of SUDEP are described in Table 3, and factors of adherence and/or past reported AED low levels in Table 4. Two patients were on monotherapy, two received two AED, and the other three were under polytherapy with more than two AED. We reviewed the blood AED levels evaluated during recent years in our SUDEP patients and found results for patients 3, 5, 6, and 7. Patient 3 had normal blood levels of valproate acid (VPA) and phenobarbital (PB) in a blood test six months prior to SUDEP. Patients 5 and 6 had normal VPA levels a year before SUDEP took place. Patient 7 had several levels of VPA below the therapeutic range in routine exams in the past although the latest was within the normal range. In addition, we checked our electronic records and confirmed that in 6 out of 7 cases there were clear reports of poor adherence to treatment.
Table 4Characteristics of SUDEP / near-SUDEP episodes, autopsy results, and final diagnosis.
ID
Witnessed
Time found (h)
Sleep
Possible Seizure Triggers
Situation found
CPR
Autopsy result
SUDEP
1
No
8-16
Unknown
History of low AED levels
Prone position
No
Unknown
Probable
2
Yes, FBTCS
24-8
Yes
Toxic consumption, Poor adherence
–
Yes, ineffective
Without cause. Mild pulmonary oedema. Long QT-syndrome. Toxicologic: negative for drugs; LEV detected (not quantified)
Definite Plus
3
No
16-24
Yes
Poor adherence
Prone position, abnormal posture
No
Without cause. Mild pulmonary oedema. Mild aortic valve sclerosis. Toxicologic: negative for drugs; PB 9.8mcg/mL, VPA 17.2mcg/mL
Definite
4
No
8-16
Yes
Poor adherence
Prone position, enuresis
No
Without cause. Tongue bite. Mild pulmonary oedema. Asymptomatic 75% stenosis LAD. Toxicologic: negative for drugs; OXC detected (not quantified)
Definite
5
No
24-8
Yes
Poor adherence
Prone position, abnormal posture
No
Not performed
Probable
6
Yes, FBTCS
16-24
No
Toxic consumption, Poor adherence
–
Yes, effective
Not performed
Near-SUDEP
7
Yes, FBTCS
24-8
No
AED stopped for VEEGM
Supine position
Yes, effective
–
Near-SUDEP
AED: antiepileptic drugs. CPR: cardiopulmonary resuscitation. FBTCS: focal to bilateral tonic-clonic seizure. LDA: left anterior descending artery. LEV: levetiracetam. OXC: oxcarbazepine. PB: phenobarbital. SUDEP: sudden unexpected death in epilepsy. VEEGM: video-electroencephalographic monitoring. VPA: sodic valproate.
Six patients (86%) had definite drug-resistant epilepsy and two were considered appropriate candidates for epilepsy surgery (Table 3). In case 5 a selective amygdalohippocampectomy on the basis of hippocampal sclerosis (confirmed by anatomopathological examination) was performed, but seizures persisted. Due to a behavioural disorder an invasive study was excluded, and subsequently the subject underwent a second surgery by means of an anterior temporal lobectomy. Nevertheless, seizures continued suggesting the possibility of an insular epilepsy. SUDEP took place three months after the second surgery. Case 4 was on the waiting list for epilepsy surgery when SUDEP occurred. In four patients a pre-surgical study was started (57%). No autonomic features were observed in the clinical semiology review of recorded seizures. The ECG analysis during the recorded seizures showed: no changes in auras, mild sinus tachycardia (below 95 bpm in all cases) in focal impaired awareness seizures, and marked sinus tachycardia when focal seizures evolved to bilateral tonic-clonic seizures. In case 4, after a FBTCS some supraventricular extrasystoles, and a single ventricular extrasystole could be observed lasting for 1 min. Case 7 occurred during a VEEGM. AED were completely discontinued when cardiorespiratory arrest (CRA) took place. Seizure commenced in the left posterior quadrant and evolved to a FBTCS. Subsequently, a prolonged (8 min) post-ictal generalized electroencephalographic suppression (PGES) was observed (Fig. 1). In the early post-ictal, respiratory rate was 12 rpm, followed by a terminal apnoea 39 s after seizure end. Heart rate persisted 2 min after seizure end, but dramatically dropped after that time, leading subsequently to asystole. Cardiopulmonary resuscitation (CPR) was started within the first minute after asystole, and spontaneous breathing and heartbeat were recovered after 2 min of CPR without consequences. No other respiratory dysfunctions during or after seizures were observed in the other SUDEP patients.
Fig. 1Prolonged post-ictal generalized electroencephalographic suppression after a focal to bilateral tonic-clonic seizures in case 7.
Factors related to death are shown in Table 4. Death occurred during sleep in 4 cases (57%). Three of them slept alone and were found dead by relatives in prone position and abnormal position in bed; case 4 showed a tongue bite and enuresis, suggesting that a seizure had probably occurred before decease. Remarkably, all the non-witnessed cases of SUDEP were found in prone position. In all witnessed cases cardiorespiratory arrest (CRA) occurred after a FBTCS (Table 4). Case 7 occurred in the EMU and CPR administered by sanitary personnel. In cases 2 and 6 a CPR was started by the witnesses and continued by the emergency team, being effective in case 6, but not in case 2. Unfortunately, in the latter the time in asystole was prolonged, and after 72 h the patient was diagnosed with brain death due to severe anoxic brain injury. Patients 6 and 7 met the criteria of near-SUDEP (Table 4). Five patients (71%) revealed risk factors for uncontrolled seizures: in one case AED therapy was discontinued during VEEGM, three cases showed poor treatment adherence (case 3 had infra-therapeutic AED levels in the autopsy toxicologic test), and in two cases recreative drug abuse was confirmed (Table 4).
Only in 4 cases (57%) was an autopsy performed; the results are summarized in Table 4. In all SUDEP cases, the patient's relatives responded to the telephone interview, but in case 1 the local court responsible for the case refused to provide the autopsy report. In case 6 the family turned down the necropsy, but during admission a complete study had been performed, ruling out other causes for CRA apart from the previous seizure. None of the performed autopsies revealed a cause for the deaths, and in none was SUDEP mentioned as a possible diagnosis, although in all cases the pathologist knew the antecedent of epilepsy. It is noteworthy that in all cases congestive pulmonary oedema was described, as well as abundant secretions in the upper airways. Case 4 had a 75% stenosis of the anterior descending artery, considered asymptomatic since no myocadiac infarct was found. In summary, 2 patients presented definite SUDEP, 1 definite SUDEP plus, 2 probable SUDEP, and 2 near-SUDEP.
4. Discussion
We reviewed the causes of death of all patients evaluated for epilepsy in our EMU during the previous eight years. Seven cases met the criteria for SUDEP or near-SUDEP. SUDEP was the only direct epilepsy-related cause of death in our cohort, representing 6.8% of all deaths and affecting 0.56% of the study participants. The incidence of SUDEP was 1.3/1000 patients per year, in agreement with previous studies conducted in multidisciplinary epilepsy units [
]. Currently, knowledge concerning this issue is increasing among both physicians and the general population. Nevertheless, there is a lack of SUDEP epidemiological studies in southern European countries, probably due to multifactorial causes. For instance, although there was an initiative from the Spanish Society of Neurology to create a national SUDEP registry of SUDEP [
] it did not come into practice. Our study confirms that in our country SUDEP cases are not adequately reported as reflected by the fact that the referring physician was unaware of 57% of cases. Moreover, unless the patients’ relatives contacted the referring neurologists (which is quite infrequent [
]) they would not be informed of a case of sudden death. Other observed restrictions included considerable bureaucratic obstacles to obtain autopsy records from courts and medical-legal institutes. This suggests that the Spanish legal system is dissociated from the healthcare structure.
Our analysis supported the importance of several risk factors related to SUDEP. Those present in all our cases were early onset of epilepsy (average, 19.4 years old), poor seizure control, generalized tonic-clonic seizures (GTCS) (0.5–3 per month), poor adherence to treatment, and long-lasting epilepsy (average 16.6 years), in accordance with previous studies [
Practice guideline summary: Sudden unexpected death in epilepsy incidence rates and risk factors: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology and the American Epilepsy Society.
]. Six of our 7 cases (86%) had confirmed poor adherence to treatment, or had presented several low AED levels in routine examinations in the past. In the only patient in whom post-mortem AED levels were available, these were below the therapeutic range, and in case 7 the near-SUDEP episode occurred after AED reduction during VEEGM. Therefore, our data suggest that fluctuations in AED levels may be a relevant risk factor for SUDEP. Data related to the body (posture in bed, enuresis, lingual bite, and sialorrhea) indicated that a seizure probably occurred prior to death. Specifically, mild pulmonary oedema and abundant secretions in the upper airways were present in all the performed autopsies, which are common findings after GTCS [
Sudden unexplained death in epilepsy (SUDEP) following previous seizure-related pulmonary oedema: case report and review of possible preventative treatment.
]. However, and this is remarkable, none of the autopsies performed considered SUDEP as a possible diagnosis of death. It is possible that pathologists in our milieu are not familiar with this specific entity.
The final pathophysiological mechanism of SUDEP is far from elucidation. Rather than being associated with a unique common mechanism, SUDEP cases may occur due to the combination of various factors and circumstances at a given moment. The SUDEP cases recorded during VEEGM in the MORTEMUS study [
] revealed cardiac and respiratory dysfunction, altered arousal and PGES pattern, common to all cases. Frequently, after a GTCS there is a period of autonomic dysregulation, initially dominated by a sympathetic hyperactivation coupled with parasympathetic suppression, while the later phase is characterized by an impaired vagal recovery [
]. The PGES pattern correlates with baro- and chemoreceptor centre dysfunction, the production of central apnoea and cardiac arrhythmias, and it is thought to be an SUDEP risk marker [
]. In our study, 2 of the 4 cases studied with VEEGM had a recorded FBTCS with a PGES pattern (Fig. 1). In case 7, near-SUDEP took place during VEEGM, allowing an electro-clinical analysis of the episode. The sequence of events was similar to that most frequently observed in the MORTEMUS study [
]: an early post-ictal neurovegetative breakdown leading to a terminal apnoea that preceded the terminal asystole by 2 min. This patient had a documented history of peri-ictal apnoea, which is a proposed SUDEP biomarker [
] (57% of our cases), suggesting that different factors associated with sleep (e.g. increased likelihood of being alone or having a non-witnessed seizure, deeper impairment of arousal after seizures, or a five-fold greater risk of PGES [
]) may be determinant for the occurrence of SUDEP.
There is a reported relationship between seizures, cardiac arrhythmias, and cardiac repolarization alterations, especially in the setting of channelopathies [
]. Several genetic mutations associated with epilepsy and cardiac repolarization abnormalities have been identified in SUDEP patients (SCN1A, KCNQ1, KCNH2, FBN1, HCN1, SCN4A, SCN5A, EFHC1, NOS1AP, CDKL5, CNTNAP2, GRIN2A, and ADGRV1) [
]. One of our patients (case 2) had a previously unknown genetic mutation related to long QT syndrome (LQTS), diagnosed at post-mortem. Two other patients (cases 5 and 6) had ECG abnormalities (right bundle branch block and left ventricular hypertrophy), and in one case (case 4), supraventricular and ventricular extrasystoles in the early post-ictal phase were observed. Therefore, in 4 of the 6 accessible ECG recordings some abnormality was determined (Table 3).
Such findings taken together suggest that, in concordance with previous reports, in our case series the pathophysiological risk factors associated with SUDEP were active GTCS, fluctuations in AED levels, seizures during sleep, prone position, male sex, early-onset and long-lasting epilepsy. Moreover, post-ictal neurological depression (represented through PGES pattern) and cardiac conduction abnormalities probably play a crucial role in SUDEP pathophysiological mechanism.
5. Study limitations
This study has major limitations. First, there is a selection bias determined by the character of a referral centre. Most of the patients presented hard-to-treat epilepsies and represented a sample from a reference centre, but not of the general population. Moreover, as a monocentric study it additionally suffers from selection bias. Because of the retrospective nature of the study, we cannot rule out the fact that some patients could have been overlooked. Finally, the lack of monitoring during most of the SUDEP episodes limits the etiologic and pathophysiologic conclusions that can be obtained from the report.
6. Conclusions
This is the first EMU-based case series analysis of SUDEP conducted to date in Spain. Our study suggests that this entity is highly underdiagnosed, a finding that concurs with other authors. Moreover, we confirmed the presence of most the SUDEP risk biomarkers currently described, and established that these biomarkers are not consistently assessed.
Based on this evaluation, we highlighted several characteristics in our health system that might preclude the correct identification of SUDEP. One of them is that this entity may be underappreciated by pathologists leading to an underestimation of final diagnosis of death. In addition, we verified the existence of legal, administrative difficulties to obtain autopsy results, even for the medical specialists responsible for the patient.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflicts of interest
None.
References
Forsgren L.
Beghi E.
Õun A.
Sillanpää M.
The epidemiology of epilepsy in Europe - A systematic review.
Practice guideline summary: Sudden unexpected death in epilepsy incidence rates and risk factors: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology and the American Epilepsy Society.
The burden of premature mortality of epilepsy in high-income countries: a systematic review from the Mortality Task Force of the International League Against Epilepsy.
Operational classification of seizure types by the international league against epilepsy: position paper of the ILAE commission for classification and terminology.
(form (http://www.sen.es/attachments/article/834/REGISTRO%20SUDEP.pdf) in http://www.sen.es/investigacion/id/renmuertesubita. Consulted last time at June)2018
Sudden unexplained death in epilepsy (SUDEP) following previous seizure-related pulmonary oedema: case report and review of possible preventative treatment.
30032-9&id=gr1.jpg){kind=link}