Transcript Dr Kathryn Swoboda

Sudden Unexpected Death in
Epilepsy
Kathryn J. Swoboda, MD, FACMG
Matthew Sweney, MS, MD
AHCF International Family Conference
San Francisco, CA
June 28th, 2012
What is SUDEP?
 Sudden, unexpected, witnessed or unwitnessed, nontraumatic, and non-drowning death of a subject
affected by epilepsy, with or without evidence for a
seizure
 Autopsy does not reveal a structural or toxicological
cause of death.
Why talk about SUDEP
 The International AHC community has unfortunately witnessed
several deaths of AHC children and adults over the past decade
 Some of these deaths were entirely unexpected, not preceded
or accompanying unusually severe events, and sometimes
occuring during sleep
 Individuals with neurologic and neurodevelopmental disorders
have an increased risk for injury or death or illness due to their
disabilities, ie aspiration or accidental trauma
 Understanding when, where and why death or injury occurs in
some children and/or adults is critical for our community in
order to minimize this problem and to understand it
Historical Perspective
 “Sudden death in a fit” has been noted in medical
literature since the mid-19th century
 Posed as potential cause of death for Gustave Flaubert,
Prince John of Great Britain, and ancestors of Julius
Caesar
 Mid 20th century
 “As far as longevity is concerned, the patient should
definitely understand that epilepsy per se rarely causes
death and that there is no reason why an epileptic should
not live as long as he would if he did not have epilepsy”—
Dr. Samuel Livingstone, 1963
Epidemiology
 Mortality (Death) in Epilepsy
 Often categorized by epilepsy type and age category
 Standardized Mortality Ratio (SMR) for epilepsy = ratio of
observed deaths to expected deaths, ranges 1.0 - 7.0 in
epileptic patients Lancet Neurol 2006; 5: 481–87
 Challenges of Bias Epilepsy & Behavior 10 (2007) 363–376
 x cases of death in epilepsy/y total cases of death from other
causes
 Inaccurate numerator (e.g. inaccurate x Mis- or missed diagnosis
of epilepsy)
 Inaccurate denominator (e.g inaccurate y total causes of death
in general)
 Y is derived from census data, which has its own host of
problems
 Taking confusion one step further…
Epidemiology
 SUDEP pitfalls




Epilepsy & Behavior 10 (2007) 363–376
Lack of autopsy means cause of death is uncertain
Death certificate reliability is often questionable
Cultural and religious sensitivity in reporting death
The role of the specialist center
 Excess of severe and rare cases
 Longevity selects out early deaths
 “lost to follow up” and database linkage
 Study design (prospective vs retrospective, cohort, casecontrol, cross-sectional)
Epidemiology
Lancet Neurol 2008 (7):1021-1031
SUDEP in children
 Considerably less attention paid
 Probably less frequent, some estimate cases range 12/10,000 patient years (roughly 1/10 of adult SUDEP)
 Rarity makes organized studies challenging
 Emphasis on identifying underlying heart problems,
promoting medication compliance (see following slides)
 Children with seizures and neurologic handicap have
higher mortality rate in general, but influence on SUDEP
rate is not clear
Camfield and Camfield. Sem in Pedi Neuro 2005 (12):10-14
Risk Factors
(typically from an adult standpoint)
Epilepsy & Behavior 2009(14):280-287
Epilepsia 2011. DOI: 10.1111/j.1528-1167.2010.02952.x
Pathophysiologic Mechanisms
(Grasping at straws)
 Winter weather?
 Hibernators have unique protective cardiovascular characteristics—what can
we learn from them?. Med Hypotheses 2008;70(5):929-32.
 Lunar phase?
 SUDEP most common during full moon (70%) vs waxing (20%) and new moon
(10%) Epilepsy Behav. 2009 Feb;14(2):404-6
 409 probable SUDEP showing some evidence supporting link between
temperatures and season in SUDEP Epilepsia 2010 May;51(5):773-6
 Geomagnetic forces?

In rats with limbic epilepsy, 10% died following exposure to sham EM field, 60% of
died after exposure to natural EM fields Int J Biometeorol 2005 Mar;49(4):256-61

Time of day, date, international geomagnetic indices showed no correlation to
SUDEP. Neurology 2000 Feb 22;54(4):903-8
Pathophysiologic Mechanisms
Lancet 2008(7):1021-1032
Pathophysiologic Mechanisms
Cardiac/Autonomic Factors
Seizure 2010 (19):455-460
Epilepsia 2010 (5):725-737
 Autonomic “storm” or influences on heart during an
epilpetic seizure
 Brain (i.e. seizure) related influences on heart rate
 R hemisphere helps contribute to fast heart rate
 Tachycardia is nearly universal pre, ictal, or post
 Associated with mesial temporal lobe epilepsy
 May predispose to ictal atrial fibrillation (heart rhythm
abnormality occurs during an actual seizure event)
 L hemisphere helps contribute to slow heart rate
 Bradycardia (reduced heart rate) during seizure <5% of pts
 Asystole (heart stops) during seizure <1% of pts
 May be underestimated, as most of the time, spontaneous
recovery occurs
Pathophysiologic Mechanisms
Cardiac Factors
Seizure 2010 (19):455-460
Epilepsia 2010 (5):725-737
 Long-standing (chronic epilepsy) influences on heart rhythm
 Inter-beat interval (QTc) prolongation/shortening
 Seizure drugs may affect QT variability (GBP, LTG may prolong?)
 Erratic inter-beat interval (i.e. large QT dispersion) places one at
risk for reentry arryhthmias (irregular heart beats)
 Heart Rate Variability (how nimble is your heart rate??)
 HRV lower in chronic epilepsy patients, resulting in loss of vagal
tone and possible increased likelihood of irregular heart rhythms
 Some specific AEDs may influence HRV (CBZ,PHT associated with
low heart rate during seizure)
 Vagal Nerve Stimulator and HRV—role is unclear
Pathophysiologic Mechanisms
Respiratory Factors
Nature Reviews Neurology 2009 (5):492-504
 The brain (i.e. seizure) influence on respiratory rate
 Saturations <90% in 20-30% of all seizures
 Hypoxemia seen with carbon dioxide retention in some sz
 May be protective to some degree—stimulates respiratory drive
 Brain-induced fluid in the lungs (Neurogenic Pulmonary
Edema)
 Massive adrenaline surge results in fluid accumulation in lungs
 Witnessed SUDEP typically occurs minutes rather than hours
following seizure (based on unfortunate cases in epilepsy
monitoring units
 Brain-induced airway tightening (i.e. Laryngospasm)
Pathophysiologic Mechanisms
Cerebral Factors
Epilepsia 2009 (5):916-920
Epilepsia 2010 (11):2344-2347
 Cerebral Electrical Shutdown—Brain flatline
 Reported exclusively in cases of LTM SUDEP (when EEG
monitoring records the death)
 Precedes heart rate slowing and respiratory depression
 May be more common following grand mal/generalized
tonic-clonic seizures
 Prone position (face down) may increase risk—akin to
positioning influence in sudden infant death syndrome
(SIDS)
Pathophysiologic Mechanisms
Genetic Factors
 Increased incidence of SUDEP in Dravet Syndrome, an
epilepsy syndrome caused by a sodium channel mutation
 2% of Dravet pts died from SUDEP based on IDEA League
report (cohort of 833 individuals) Epilepsia 2010 (5):1915-1918
 Identification of SCN5A mutation in SUDEP case
Seizure 2009
(2):158-160
 Expressed in both the brain and heart—the only case in which
a mutated channel is expressed in both locations
 Long QT type 2 (VGKC mutation = voltage gated potassium
channel) has higher association with epilepsy phenotypes
than LQT 1 and 3
Pathophysiologic Mechanisms
Animal Model Studies
Epilepsia 2010 (3):465-468
 Adenosine Mouse Model
 Adenosine is endogenous anticonvulsant produced by the body
under extreme duress
 Impaired adenosine clearance can potentially lead to
decreased ventilatory rate and apnea (cessation of breathing)
 Excess adenosine production combined with decreased
clearance has lead to death in mouse models
 Adenosine receptor antagonist (caffeine) can extend survival
time.
 Could be synergistic to drugs we use to control seizures
 No studies of this in humans
Pathophysiologic Mechanisms
Animal Model Studies
Epilepsia 2006 (1):21-26
 DBA/2 Mice and audiogenic (sound-induced) seizures
 88% with respiratory arrest following audiogenic seizure (remaining
12% can be induced with cyproheptadine—a serotonin blocker)
 Mice treated with SSRIs (pro-serotonin drugs) resulted in decreased
rates of respiratory arrest and increased rates of survival
 Highlights the potential interaction between seizures and
respiratory/arousal centers, which are dependent on serotonin
 No studies in humans regarding this
 Dr. Chugani’s work has demonstrated some involvement of the
serotonergic system in AHC
Pathophysiologic Mechanisms
Hypotheses
Epilepsia 2011 (Suppl. 1):28-38
 Hypothesis for post-ictal death
 Post-ictal state may be due in part to stunning of serotonin-related
systems
 Depression of serotonin-generating neurons could lead to ictal
and/or postictal hypoventilation (decreased respiratory rate/effort)
 Genetic susceptibilities with bad luck (e.g. prone position) may
prevent compensatory response (respiratory drive or arousal)
 Seizure/depression/SUDEP phenotype?—may be associated with
activity of serotoninergic systems in brain and/or brainstem
 Some have hypothesized whether SSRIs may prove protective—
unfortunately no answers
What about SUDEP and AHC?
 Support for diagnosis of epilepsy in ~50% of AHC patients
 Many children/adults have infrequent seizures which are
easily controlled with standard epilepsy medications,
but status epilepticus also occurs
 Unclear what role specific mutations relevant in AHC
may play in risk for SUDEP, cardiac arrhythmias, or
neurotransmitter regulation
 Catastrophic deaths in AHC have other causes too
 details of such cases may prove helpful in understanding
potential risk factors, but broad generalization of
individual observations is VERY challenging
Management Issues
 Medication Compliance—suggests the importance of adhering to prescribed
treatments for epilepsy, and early treatment intervention for generalized seizures
 Need for Bed Monitors/Supervised sleeping/“Back to sleep”—unclear; however,
sleep study to rule out sleep-disordered breathing may be indicated in some cases
 Anticonvulsant medication selection and potential interaction with genetic factors
 Pacemakers indicated in rare cases identified with ictal bradycardia or asystole
associated with episodes (slowing or stopped heartrate)
 VNS protective or harmful?—trade off between improved seizure control and
disruption of vagal tone
 SSRIs? Caffeine?—unproven
 Provision of oxygen with prolonged seizure activity or spells in which oxygen levels
drop below 90% for prolonged periods

Consider asking your doctor about need to monitor oxygen levels if your child has
color changes around lips or obvious changes in breathing patterns with any type of
spells, or in association with trials of new medications
Future directions
 Work with medical advisory board members and other
specialist consultant to create specific guidelines for
AHC patients
 Diagnostic workup including laboratory and genetic
evaluations
 Standard of care guidelines
 Definitions for types of episodes
 How to grade episodes and recommendations for interventions
including monitoring safety for feeding and breathing during
episodes
 How to prepare for seizures, and understand risks of associated
seizure disorders/status and age of your child
Acknowledgements
 Patients and Families with AHC
 Matthew Sweney, MS, MD; Aga Lewelt MD; Sandy Reyna
MD; Abby Smart RN; Tara Newcomb, MS, LCGC
 The national and international AHC foundations, who
are the primary liasons to families and the key to our
future success in effectively diagnosing, managing, and
treating the many symptoms of AHC