Continum Epilepsy.2019 PDF

APRIL 2019
VOL. 25 NO. 2 Epilepsy

Guest Editors: Cynthia L. Harden, MD; Erik K. St. Louis, MD, MS, FAAN
304 Editor’s Preface

Editor-in-Chief: Steven L. Lewis, MD, FAAN
REVIEW ARTICLES
306 Epilepsy Overview and Revised Classification of Seizures

and Epilepsies
Alison M. Pack, MD, MPH
322 Evaluation of Seizure Etiology From Routine Testing to Genetic

Evaluation
Stephan U. Schuele, MD, MPH, FAAN
343 Optimizing Management of Medically Responsive Epilepsy

Derek Bauer, MD; Mark Quigg, MD, MSc, FANA, FAES
362 Identification and Treatment of Drug-Resistant Epilepsy

Ji Yeoun Yoo, MD; Fedor Panov, MD
381 Antiepileptic Drug Treatment of Epilepsy in Children

Ahsan N. V. Moosa, MD
408 Treatment of Women With Epilepsy

Mona Sazgar, MD, FAES
431 Electroencephalography in Epilepsy Evaluation

Hai Chen, MD, PhD; Mohamad Z. Koubeissi, MD, FAAN
DENOTES CONTINUUM
454 Epilepsy Emergencies: Status Epilepticus, Acute Repetitive
AUDIO INTERVIEW
Seizures, and Autoimmune Encephalitis
DENOTES VIDEO Stephen VanHaerents, MD; Elizabeth E. Gerard, MD
CONTENT
477 Counseling and Management of the Risks of Living
With Epilepsy
Katherine Noe, MD, PhD, FAAN
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

492 Nonepileptic Episodic Events
Jennifer L. Hopp, MD, FAAN
508 Update on Antiepileptic Drugs 2019

Bassel W. Abou-Khalil, MD, FAAN
MEDICOLEGAL ISSUES
537 Driving and Epilepsy: Ethical, Legal, and Health Care Policy
Challenges
Joseph S. Kass, MD, JD, FAAN; Rachel V. Rose, JD, MBA
PRACTICE ISSUES
543 Tips and Resources for Medication Reconciliation

Anant M. Shenoy, MD, FAAN; Amy Bennett, JD;
Alan Z. Segal, MD, FAAN
SELF-ASSESSMENT AND CME
296 Learning Objectives and Core Competencies
551 Instructions for Completing Postreading Self-Assessment and CME

Test and Tally Sheet
553 Postreading Self-Assessment and CME Test
565 Postreading Self-Assessment and CME Test—Preferred Responses
576 Erratum
577 Index
List of Abbreviations (Back Cover)

CONTRIBUTORS
Cynthia L. Harden, MD, Bassel W. Abou-Khalil, MD,

Guest Editor FAAN
Head of Clinical Epilepsy Professor of Neurology,
Development, Xenon Vanderbilt University Medical
Pharmaceuticals Inc, Burnaby, Center, Nashville, Tennessee
British Columbia, Canada;
Relationship Disclosure: Dr Abou-Khalil has
Attending Neurologist, served on the editorial board of Clinical
14th Street Medical Arts Center, Neurophysiology and has received
research/grant support from Biogen,
New York, New York the National Institute of Neurological
Disorders and Stroke, SK-Pharma, Sunovion
Relationship Disclosure: Dr Harden
Pharmaceuticals Inc, and UCB SA.
is a full-time employee of Xenon
Pharmaceuticals Inc and holds stock
Unlabeled Use of Products/Investigational
options of unknown value in Xenon
Use Disclosure: Dr Abou-Khalil discusses the
Pharmaceuticals Inc.
unlabeled/investigational use of primidone
for the treatment of essential tremor,
valproate for the treatment of generalized
Use Disclosure: Dr Harden reports
myoclonic and generalized tonic-clonic
no disclosure.
seizures, gabapentin for the treatment of
headache and sleep disorders, lamotrigine
as a first-line treatment for epilepsy,
zonisamide as initial monotherapy for
Erik K. St. Louis, MD, MS, FAAN, epilepsy, and cannabidiol and clobazam for
Guest Editor the treatment of focal seizures.
Co-director, Mayo Center for
Sleep Medicine; Associate
Professor of Neurology, Derek Bauer, MD
Consultant in Medicine and Assistant Professor, Department
Neurology, Mayo Clinic College of Neurology, University of
of Medicine and Science, Virginia, Charlottesville, Virginia
Rochester, Minnesota
Relationship Disclosure: Dr Bauer is a site
Relationship Disclosure: Dr St. Louis has served subinvestigator for GWEP-1521, a double-blind,
as a consultant for Axovant Sciences, Inc randomized, placebo-controlled study to
and has received research/grant support investigate the efficacy and safety of
from Axovant Sciences, Inc; Mayo Clinic cannabidiol as add-on therapy in patients
Center for Clinical and Translational Science; with tuberous sclerosis complex who
the Michael J. Fox Foundation; the National experience inadequately controlled seizures.
Institutes of Health/National Institute on
Aging; National Heart, Lung, and Blood Unlabeled Use of Products/Investigational
Institute; and Sunovian Pharmaceuticals Inc. Use Disclosure: Dr Bauer reports
no disclosure.
Use Disclosure: Dr St. Louis reports
no disclosure.
Amy Bennett, JD
Manager, Quality Improvement,
American Academy of Neurology,
Minneapolis, Minnesota
Relationship Disclosure: Ms Bennett reports
no disclosure.

Use Disclosure: Ms Bennett reports no
disclosure.
298 APRIL 2019

Hai Chen, MD, PhD Jennifer L. Hopp, MD, FAAN
Assistant Professor of Associate Professor of
Neurology, George Washington Neurology, University of
University, Washington, DC Maryland School of Medicine;
Director, Epilepsy Division,
Relationship Disclosure: Dr Chen reports
no disclosure. University of Maryland Medical
Center, Baltimore, Maryland
Use Disclosure: Dr Chen reports Relationship Disclosure: Dr Hopp has received
no disclosure. grant/research support as a site principal
investigator of the Established Status
Epilepticus Treatment Trial from the National
Institute of Neurological Disorders and
Elizabeth E. Gerard, MD Stroke and from SAGE Therapeutics. Dr Hopp
Associate Professor of has received personal compensation as a
speaker for J. Kiffin Penry Epilepsy MiniFellow
Neurology, Feinberg School Network’s Epilepsy MiniFellowship and
of Medicine, Northwestern Residents Epilepsy Program. Dr Hopp
receives publishing royalties from UpToDate,
University, Chicago, Illinois
Inc and has given expert medical testimony
for Venable LLP.
Relationship Disclosure: Dr Gerard has
received personal compensation as a
lecturer for the Society for Maternal-Fetal
Use Disclosure: Dr Hopp discusses
Medicine, Society of OB/GYN Hospitalists,
the unlabeled/investigational use of
and UCB China, and has received research/
sertraline for the treatment of psychogenic
grant support from the Eleanor Wood-
nonepileptic seizures.
Prince Grant from the Woman’s Board
of Northwestern Memorial Hospital,
the National Institute of Neurological
Disorders and Stroke/National Institutes of
Health, SAGE Therapeutics, and Sunovion Joseph S. Kass, MD, JD, FAAN
Pharmaceutical Inc. Associate Dean, Office of
Student Affairs; Professor
Use Disclosure: Dr Gerard discusses of Neurology, Psychiatry,
the unlabeled/investigational use of and Medical Ethics; Director,
immunosuppressant medications for the
treatment of autoimmune encephalitis
Alzheimer’s Disease and
(cyclophosphamide, IV immunoglobulin, Memory Disorders Center,
IV methylprednisolone/corticosteroids, Baylor College of Medicine;
plasma exchange, and rituximab) and the
pharmacologic and nonpharmacologic Chief of Neurology,
therapies for the treatment of seizures Ben Taub General Hospital,
and status epilepticus (convulsive and Houston, Texas
nonconvulsive), which include diazepam,
fosphenytoin/phenytoin, levetiracetam, Relationship Disclosure: Dr Kass serves
lorazepam, midazolam, phenobarbital, as associate editor of medicolegal issues
and valproate sodium/valproic acid. for Continuum, as an associate editor for
Dr Gerard discusses the unlabeled/ Continuum Audio, as a neurology section
investigational use of several agents editor of Ferri’s Clinical Advisor for Elsevier,
for the treatment of refractory and super- and as co-editor of Neurology Secrets,
refractory status epilepticus, which include Sixth Edition. Dr Kass has received personal
corticosteroids/methylprednisolone, compensation for CME lectures from
electroconvulsive therapy, hypothermia, Pri-Med LLC.
isoflurane, IV immunoglobulin, ketamine,
ketogenic diet, midazolam, pentobarbital, Unlabeled Use of Products/Investigational
propofol, thiopental, transcranial magnetic Use Disclosure: Dr Kass reports no
stimulation, and vagal nerve stimulation. disclosure.
C O N T I N U U M J O U R N A L .C O M 299

CONTRIBUTORS (CONTINUED)
Mohamad Z. Koubeissi, MD, Alison M. Pack, MD, MPH

FAAN Associate Professor of
Professor, Department of Neurology, Columbia University
Neurology; Director, Epilepsy Irving Medical Center,
Center, George Washington New York, New York
University, Washington, DC
Relationship Disclosure: Dr Pack receives
Relationship Disclosure: Dr Koubeissi serves research/grant support from the National
on the editorial boards of Epilepsy Currents Institutes of Health/National Institute
and Functional Neurology and as the of Neurological Disorders and Stroke as
surgery and device editor of Epilepsy.com. co-investigator of a multicenter study and
Dr Koubeissi has received personal receives royalties from UpToDate, Inc.
compensation for serving on the speakers’
bureaus of Sunovion Pharmaceuticals Inc Unlabeled Use of Products/Investigational
and UCB SA, and has received publishing Use Disclosure: Dr Pack reports no
royalties from Springer for his book, Epilepsy disclosure.
Board Review.

Use Disclosure: Dr Koubeissi reports no Fedor Panov, MD
disclosure. Assistant Professor,
Department of Neurosurgery,
Icahn School of Medicine
Ahsan N. V. Moosa, MD (Ahsan at Mount Sinai; Associate
Moosa Naduvil Valappil, MD) Director, Adult Epilepsy
Staff Physician, Pediatric Program, Mount Sinai, New
Epilepsy, Epilepsy Center, York, New York
Cleveland Clinic; Assistant
Professor of Medicine, Lerner Relationship Disclosure: Dr Panov has
received personal compensation as a
College of Medicine, Case consultant for NeuroPace, Inc
Western Reserve University, and Zimmer Biomet.
Cleveland, Ohio Unlabeled Use of Products/Investigational
Relationship Disclosure: Dr Moosa reports
Use Disclosure: Dr Panov reports
no disclosure.
no disclosure.
Use Disclosure: Dr Moosa reports no
disclosure.
Mark Quigg, MD, MSc, FANA,
FAES
Professor, Department of
Neurology, University of Virginia,
Associate Professor of
Charlottesville, Virginia
Neurology, Mayo Clinic College
of Medicine, Phoenix, Arizona Relationship Disclosure: Dr Quigg has
received research/grant support as
Relationship Disclosure: Dr Noe has received principal investigator of studies from the
research/grant support from National Institutes of Health/National
NeuroPace, Inc. Institute of Neurological Disorders and
Stroke, the University of Virginia Brain
Unlabeled Use of Products/Investigational Institute, and ZETO Inc. Dr Quigg has
Use Disclosure: Dr Noe reports no received publishing royalties from Elsevier
disclosure. and has given expert medical testimony.

Use Disclosure: Dr Quigg reports no
disclosure.
300 APRIL 2019

Rachel V. Rose, JD, MBA Stephan U. Schuele, MD, MPH,
Attorney, Rachel V. Rose FAAN
Attorney at Law PLLC; Affiliated Professor of Neurology
Faculty, Baylor College of and Physical Medicine and
Medicine, Houston, Texas Rehabilitation, Feinberg School
of Medicine, Northwestern
Relationship Disclosure: Ms Rose serves
on the editorial board of BC Advantage and University, Chicago, Illinois
receives book royalties from the American
Bar Association. Relationship Disclosure: Dr Schuele has
received personal compensation for serving
Unlabeled Use of Products/Investigational on the speaker’s bureau of Eisai Co, Ltd and
Use Disclosure: Ms Rose reports no Sunovion Pharmaceuticals Inc. Dr Schuele
disclosure. serves on the board of directors of the
American Clinical Neurophysiology Society
and the Epilepsy Foundation Greater Chicago,
and on the editorial board of Journal of Clinical
Neurophysiology. Dr Schuele has received
research/grant support from the Danny Did
Clinical Professor, Department Foundation and the National Institutes of
of Neurology, University of Health and has been asked to provide expert
medical testimony in cases involving sudden
California, Irvine, Orange, unexpected death in epilepsy.
California
Relationship Disclosure: Dr Sazgar has Use Disclosure: Dr Schuele reports no
received personal compensation for serving disclosure.
on the scientific advisory board for UCB
SA and for serving on the speaker’s bureau
of Eisai Co, Ltd; Sunovion Pharmaceuticals
Inc; and UCB SA. Dr Sazgar has received Alan Z. Segal, MD, FAAN
grants or research support from Biogen,
Sunovion Pharmaceuticals Inc, and UCB SA
Associate Professor of Clinical
as the principal investigator for clinical trials Neurology, Weill Cornell
and has received publishing royalties from Medical College, New York,
Springer for her book Controversies in Caring
for Women with Epilepsy. New York
Unlabeled Use of Products/Investigational Relationship Disclosure: Dr Segal reports no

Use Disclosure: Dr Sazgar discusses disclosure.
the unlabeled/investigational use
of acetazolamide, clobazam, depot- Unlabeled Use of Products/Investigational
medroxyprogesterone acetate, and Use Disclosure: Dr Segal reports no
progesterone products for the treatment of disclosure.
women with catamenial epilepsy.
Anant M. Shenoy, MD, FAAN

Chief of Neurology, Mount
Auburn Hospital, Cambridge,
Massachusetts
Relationship Disclosure: Dr Shenoy reports
no disclosure.

Use Disclosure: Dr Shenoy reports no
disclosure.

Stephen VanHaerents, MD Ji Yeoun Yoo, MD

Assistant Professor of Assistant Professor, Icahn
Neurology, Feinberg School School of Medicine at Mount
of Medicine, Northwestern Sinai, New York, New York
University, Chicago, Illinois
Relationship Disclosure: Dr Yoo has
received publishing royalties from Elsevier
Relationship Disclosure: Dr VanHaerents
for the book Rowan's Primer of EEG and
has received research/grant support from
has received personal compensation as a
the Citizens United for Research in Epilepsy,
lecturer for the Korean Epilepsy Society
National Institute of Mental Health, National
and for serving on the advisory board
Institute on Aging, National Institutes of Health,
of Zimmer Biomet. Dr Yoo has received
and SAGE Therapeutics. Dr VanHaerents has
research/grant support from the NeuroNEXT
received travel honoraria from NeuroPace
Program of the National Institutes of Health/
and SAGE Therapeutics.
National Institute of Neurological Disorders
and Stroke.
Use Disclosure: Dr VanHaerents discusses
Use Disclosure: Dr Yoo reports no disclosure.
the unlabeled/investigational use of
immunosuppressant medications for the
treatment of autoimmune encephalitis
(cyclophosphamide, IV immunoglobulin,
IV methylprednisolone/corticosteroids,
plasma exchange, and rituximab) and the
pharmacologic and nonpharmacologic
therapies for the treatment of seizures
and status epilepticus (convulsive and
nonconvulsive), which include diazepam,
fosphenytoin/phenytoin, levetiracetam,
lorazepam, midazolam, phenobarbital,
and valproate sodium/valproic acid.
Dr VanHaerents discusses the unlabeled/
investigational use of several agents
for the treatment of refractory and
super-refractory status epilepticus, which
include corticosteroids/methylprednisolone,
electroconvulsive therapy, hypothermia,
isoflurane, IV immunoglobulin, ketamine,
ketogenic diet, midazolam, pentobarbital,
propofol, thiopental, transcranial magnetic
stimulation, and vagal nerve stimulation.
302 APRIL 2019

Self-Assessment and CME Test Writers
James W. M. Owens Jr, MD, PhD Allyson R. Zazulia, MD

Associate Professor of Professor of Neurology and
Neurology, Adjunct Associate Radiology, Associate Dean for
Professor of Pediatrics, Continuing Medical Education,
University of Washington Washington University,
School of Medicine, Seattle, St. Louis, Missouri
Washington
Relationship Disclosure: Dr Zazulia reports
Relationship Disclosure: Dr Owens serves as no disclosure.
CME co-editor for Neurology and receives
publishing royalties from UpToDate, Inc. Unlabeled Use of Products/Investigational
Use Disclosure: Dr Zazulia reports no
Unlabeled Use of Products/Investigational disclosure.
Use Disclosure: Dr Owens reports no
disclosure.

EDITOR’S PREFACE
Cloudy With a 60% Chance of

Having Another Seizure in the
Next 10 Years
This issue of Continuum is devoted to the diagnosis and management
of our patients with epilepsy. I am delighted that both Drs Cynthia L.
Harden and Erik K. St. Louis served as guest editors of this issue. At
the same time, I feel that readers should be aware of the particular
circumstances that led to having these two outstanding experts
oversee this issue. About midway through the planning and creation of this issue,
Dr Harden became Head of Clinical Epilepsy Development for a pharmaceutical
company (see the “Contributors” page of this issue); in order to avoid even the
perception of potential conflicts of interest, I, upon consultation with and at the
recommendation of the Continuum Editorial Board, asked epileptologist Dr St. Louis, a
member of the editorial board, to take over the guest editorship of this issue, which was
immediately agreed to by Dr Harden. We hope readers concur that the resulting issue,
representing the initial planning and guest editorship of Dr Harden, the subsequent
editorial work by Dr St. Louis, and the authorship of each article by the remarkable
team of invited epileptologists, provides a clinically practical and up-to-date overview
of modern management of our patients with seizures and epilepsy.
The issue begins with the article by Dr Alison M. variety of strategies by which we can optimize the
Pack, who provides a clear overview of the management of our patients who have medically
epilepsies and the important details of the revised responsive epilepsy.
classification of seizures and epilepsies, including In the following article, Drs Ji Yeoun Yoo and
that epilepsy can be diagnosed if a person has one Fedor Panov review the identification and treatment
unprovoked or reflex seizure and has a probability of of drug-resistant epilepsy and the important role
at least 60% of having another seizure within the all neurologists share in referring these patients
next 10 years. This subjective but practical to comprehensive epilepsy centers for further
forecasting now plays an important part in clinical diagnostic evaluation and potential epilepsy surgery.
epilepsy diagnosis. Dr Ahsan N. V. Moosa next provides a thorough
Dr Stephan U. Schuele provides a contemporary review of the nuances of antiepileptic drug (AED)
overview of the evaluation of our patients’ seizure treatment for children with epilepsy. Additional
etiology, beginning with routine testing to the recent Continuum articles on childhood epilepsy and
increasing role of genetic evaluation. In the next epilepsy syndromes from the Child Neurology issue
article, Drs Derek Bauer and Mark Quigg discuss a are referenced in the “Beyond the Page” section of
304 APRIL 2019

this issue, and interested readers will find these issue, therefore, represents a unique circumstance in
articles of great additional value. With regard to the my tenure as editor-in-chief of Continuum. I had
AED terminology, the Continuum editorial office is been so impressed by the encyclopedic and
aware that drugs used in the management of seizures well-organized article “Update on Antiepileptic
and epilepsy are not necessarily “antiepileptic,” Drugs,” by Dr Bassel W. Abou-Khalil in the February
but we have decided to interchangeably use the 2016 Epilepsy issue, that I requested he update it to
terms AEDs or antiseizure medications, depending be relevant to readers in 2019. The resulting article,
on the preference of the article author. “Update on Antiepileptic Drugs 2019,” represents a
Dr Mona Sazgar then discusses the important remarkable and completely up-to-date compendium
issues relevant to the treatment of women with of practical information of essentially every AED
epilepsy, including issues related to menses, currently available in our armamentarium.
pregnancy, and breast-feeding. In the Medicolegal Issues article, Dr Joseph S. Kass
Drs Hai Chen and Mohamad Z. Koubeissi provide and Ms Rachel V. Rose provide a hypothetical but
us with a concise but overarching and nicely illustrated highly relatable case to inform us about the ethical
overview of the role of EEG in epilepsy evaluation, and medicolegal issues we face daily when counseling
which will be of value to readers irrespective of our patients with epilepsy about driving.
their degree of familiarity with EEG. Drs Stephen In the Practice Issues article, Drs Anant M. Shenoy
VanHaerents and Elizabeth E. Gerard then review and Alan Z. Segal and Ms Amy Bennett provide
the critically important epilepsy emergencies, practical tips and resources for medication
including status epilepticus, acute repetitive seizures reconciliation, so important for many of our patients,
(and the newest terminology for these syndromes), including our patients with epilepsy.
and the autoimmune encephalitides. After reading the issue and taking the Postreading
Dr Katherine Noe next provides a very practical Self-Assessment and CME Test written by Drs James
guide to help us counsel and manage our patients W. M. Owens Jr and Allyson R. Zazulia, you may
regarding the risks of living with epilepsy. Of course, earn up to 20 AMA PRA Category 1 CreditsTM toward
not all events that look like seizures are epileptic self-assessment and CME or, for Canadian participants,
in origin, and Dr Jennifer L. Hopp discusses the a maximum of 20 hours toward the Self-Assessment
recognition, diagnosis, and management of these Program (Section 3) of the Maintenance of
nonepileptic episodic events and the distinction Certification Program of the Royal College of
between nonepileptic episodic events of physiologic Physicians and Surgeons of Canada. Additional credit
(but nonepileptic) origin from psychogenic can be obtained by listening to Continuum Audio
nonepileptic seizures. interviews associated with this and other Continuum
An important aim in the creation of each Continuum issues, available to all subscribers, and completing
issue is that it presents a fresh view of the topic, tests on the Continuum Audio web platform or app.
including brand new review articles authored by a Continuum Audio is also accredited by the Royal
new group of authors. The final review article in the College of Physicians and Surgeons of Canada.
My sincere thanks to Dr Harden for her skillful
and dedicated planning of this issue and for her initial
guest editing, and to Dr St. Louis, for also so capably
and responsively taking on the seemingly never-ending
tasks as subsequent guest editor. I also extend my
My sincere thanks to Dr Harden for thanks to the many expert epileptologists who
her skillful and dedicated planning provided such remarkable contributions to this issue,
of this issue and for her initial guest all to benefit the diagnosis and management of our
patients with epilepsy.
editing, and to Dr St. Louis, for also
so capably and responsively taking —STEVEN L. LEWIS, MD, FAAN
EDITOR-IN-CHIEF
on the seemingly never-ending tasks
as subsequent guest editor. © 2019 American Academy of Neurology.
CONTINUUMJOURNAL.COM 305

REVIEW ARTICLE

Epilepsy Overview and
C O N T I N UU M A UD I O
I NT E R V I E W A V AI L A B L E
ONLINE
Revised Classification of
Seizures and Epilepsies
By Alison M. Pack, MD, MPH
ABSTRACT
PURPOSE OF REVIEW: The classification of seizures, epilepsies, and epilepsy
syndromes creates a framework for clinicians, researchers, and patients
and their families. This classification has evolved over the years, and in
2017 the International League Against Epilepsy (ILAE) published an
operational classification of seizures and epilepsies. Understanding
this classification is important in the diagnosis, treatment, and
understanding of seizures and epilepsies, including epilepsy
incidence.
RECENT FINDINGS: The 2017 ILAE classification system builds on newly

formulated definitions of seizures and epilepsy. Seizure classification
begins by determining whether the initial manifestations of the seizure
are focal or generalized. If the onset of the seizure is missed or unclear,
the seizure is of unknown onset. Focal seizures are classified according
to the individual’s level of awareness, the most prominent motor or
nonmotor features of the seizure, and whether the focal seizure evolves
CITE AS:
CONTINUUM (MINNEAP MINN)
to a bilateral tonic-clonic seizure. Similarly, generalized seizures are
2019;25(2, EPILEPSY):306–321. classified according to motor or nonmotor manifestations. Motor seizures
are either tonic-clonic or other motor seizures. Nonmotor generalized
Address correspondence to
seizures primarily refer to absence seizures. Similar to seizure
Dr Alison M. Pack, Columbia
University, Department of classification, the epilepsies can be classified as focal or generalized.
Neurology, 710 W 168th St, In addition, the new classification system recognizes two new categories:
New York, NY 10032,
ap390@cumc.columbia.edu.
combined generalized and focal epilepsy and unknown epilepsy.
The concept of an epilepsy syndrome has been introduced under the
RELATIONSHIP DISCLOSURE: new classification system and refers to a cluster of features incorporating
Dr Pack receives research/grant
support from the National seizure types, EEG, imaging, and other features including genetics.
Institutes of Health/National The new classification system emphasizes the etiology of seizures
Institute of Neurological and epilepsies.
Disorders and Stroke as
co-investigator of a multicenter
study and receives royalties from SUMMARY: The recent ILAE seizure and epilepsy classification system
UpToDate, Inc.
aims to create a framework to better classify seizures and the epilepsies.
UNLABELED USE OF Universal adoption and implementation of this system will enable patients,
PRODUCTS/INVESTIGATIONAL their families, clinicians, and researchers to better define and treat
USE DISCLOSURE:
Dr Pack reports no disclosure.
the epilepsies. Incidence studies have not generally classified seizures
and the epilepsies, and use of this classification system, which
emphasizes etiology, will lead to a better understanding of epilepsy
© 2019 American Academy
of Neurology. incidence.
306 APRIL 2019

INTRODUCTION
S
eizure and epilepsy classification systems have been used in clinical
practice and research since the 1970s. Over the years, multiple
revisions have been implemented, the most recent of which is the 2017
International League Against Epilepsy (ILAE) operational epilepsy
classification system.1–3 This system aims to better define seizures and
epilepsies by classifying them using key clinical features, EEG findings, imaging,
and genetics. This article reviews the history of epilepsy classification and the
details of the 2017 system, with an emphasis on the importance of classification in
epilepsy incidence studies.
HISTORICAL OVERVIEW OF SEIZURE, EPILEPSY, AND EPILEPSY

SYNDROME CLASSIFICATION
Seizure and epilepsy classification has evolved over the years. Prior to the first
modern seizure classification by Gastaut in 1969,4,5 seizures and epilepsy types
were not distinctly recognized. Although initially met with resistance, this
system gained international recognition after 1970 and was widely used. In 1981,
the ILAE, informed by advances in technology—notably video recording with
simultaneous EEG—published a classification of seizures6 followed by a proposal
of epilepsy classification in 1985,7 which was then revised in 1989,8 wherein the
concept of an epilepsy syndrome was introduced.
The ILAE then began an effort to revise seizure terminology and the
organization of seizures and the epilepsies.9 This effort culminated in the 2017
published ILAE operational classification of seizure types1,2 and epilepsies.3
The 1981 ILAE seizure classification system dichotomized seizures into either
partial or generalized seizures. Partial seizures were defined as an epileptic
seizure in which “the first clinical and EEG changes indicate initial activation
of a system of neurons limited to part of one cerebral hemisphere.”6 Partial
seizures were subdivided by level of consciousness: simple partial seizures were
associated with no impairment of awareness, and complex partial seizures
were associated with impairment of awareness. A third partial seizure type
included seizures that evolved to a secondary generalized convulsion. A
generalized seizure was defined as a seizure “in which the first clinical changes
indicate initial involvement of both hemispheres.”6 Six generalized seizure types
were identified: absence, myoclonic, clonic, tonic, tonic-clonic, and atonic.
The 1985 epilepsy classification was also a dichotomized system dividing
epilepsies into either idiopathic or symptomatic. The term idiopathic derives
from the Greek idios, meaning self, own, and personal. Idiopathic epilepsies and
syndromes were described as disorders “not preceded or occasioned by another.”
In these disorders, no underlying cause is present other than a possible hereditary
predisposition.7 Notable idiopathic epilepsies included juvenile myoclonic
epilepsy and childhood absence epilepsy. Symptomatic epilepsies occurred
because of a known disorder or lesion. A revision of the epilepsy classification in
1989 added cryptogenic epilepsies, which were likely symptomatic, but a cause
was not identified.8
Since the seizure and epilepsy classification systems were developed in the
1980s, many advances in neuroimaging, genetics, and molecular biology have
occurred. Although the newly published seizure classification primarily
incorporates the signs and symptoms of seizures and EEG findings, the
classification of epilepsy type and syndrome encompasses these new advances.1–3

CLASSIFICATION OF SEIZURES AND EPILEPSIES
The current classification systems build on newly formulated definitions of

seizures and epilepsy. The 2005 updated seizure definition is a “transient
occurrence of signs and/or symptoms due to abnormal excessive or synchronous
neuronal activity in the brain.”10 Further, in 2014 the ILAE redefined epilepsy as
a disease and not a disorder to emphasize the importance and impact of epilepsy.
Epilepsy occurs when an individual has an epileptic seizure and his or her
“brain…demonstrates a pathologic and enduring tendency to have recurrent
seizures.”11
Previously, epilepsy was diagnosed when an individual had at least two
unprovoked or reflex seizures more than 24 hours apart. Reflex seizures are
stimulus-sensitive or sensory-evoked seizures. Although the new definition of
epilepsy includes this presentation, epilepsy is also diagnosed if a person has one
unprovoked or reflex seizure and has a probability of at least 60% of having
another seizure within the next 10 years.11 The probability of at least 60%
was chosen for this definition because this is the lower limit of the confidence
interval for someone with two unprovoked seizures having another seizure
within 10 years.12
In addition, epilepsy is diagnosed when an individual has an epilepsy
syndrome. An analysis of retrospective cases supports the applicability of
this new definition.13 In patients with a single unprovoked seizure, a second
seizure occurred in more than 80%. The definition, however, does not clarify
what metrics should be used to identify patients at risk for a recurrent seizure.
Just as multiple scenarios lead to an initial diagnosis of epilepsy, multiple
circumstances can contribute to epilepsy being considered resolved. Epilepsy is
considered resolved when a patient with an age-dependent epilepsy syndrome
is older than the age in which this syndrome was active or when a patient has
been seizure free for 10 or more years and has been off all antiepileptic drugs for
5 or more years.14
CLASSIFYING SEIZURES, EPILEPSIES, AND EPILEPSY SYNDROMES
The report by the Institute of Medicine, “Epilepsy Across the Spectrum,”
emphasizes the heterogeneity of seizures and the epilepsies.15 Making sense of
this heterogeneity requires a system to provide a common language. To classify
means to arrange something into classes or categories according to shared
qualities or characteristics. The prior and current classification systems aim to
group seizures according to clinical presentation and brain region onset and to
group epilepsies according to seizure type, age of onset, probability of remission,
EEG findings, radiologic findings, and genetics. Classification of seizures,
epilepsies, and epilepsy syndromes creates a framework for patients, their
families, clinicians, and researchers.
For patients, the universal common language of the classification system
provides a name and diagnosis, which improves understanding and recognition
of the disease. Moreover, this common language enhances communication
between patients, their families, and providers.
Similarly, clinicians can use the language to better communicate with patients
and colleagues. The classification systems allow clinicians to consider history
and include data from new technologies when making a diagnosis, choosing
treatment, and assessing prognosis.
For researchers, these systems enable standardized investigation of seizure
and epilepsy presentation, etiologies, and drug or surgical treatments. For
example, when studying incidence patterns of epilepsy, using a common
308 APRIL 2019

language and classification enables comparison across multiple populations KEY POINTS
and studies.
● In 2014 the International
UPDATED SEIZURE CLASSIFICATION League Against Epilepsy
redefined epilepsy as a
With the exception of neonatal seizures, the new seizure classification applies to disease and not a disorder to
adults and children. The new classification addresses the limitations of the 1981 emphasize the importance
seizure classification, which include the following: (1) some seizure types can and impact of epilepsy.
have either focal or generalized onset, (2) lack of knowledge about seizure onset Epilepsy occurs when an
individual has an epileptic
makes a seizure unclassifiable and difficult to place within the 1981 system,
seizure and his or her “brain
(3) retrospective seizure descriptions often do not include the level of demonstrates a pathologic
consciousness, (4) terms used in the 1981 seizure classification such as complex and enduring tendency to
partial or simple partial are difficult to understand, and (5) some seizure types have recurrent seizures.”
are not included in the 1981 classification.1
● Prior and current
To address needs of different clinicians and researchers, both basic classification systems aim to
(FIGURE 1-1) and expanded (FIGURE 1-2) versions of seizure classification were group seizures according to
created. The basic version of the seizure classification is a contracted form of the clinical presentation and
expanded classification and is intended to be more useful for pediatricians, brain region onset and
epilepsies according to
non-neurologists, general neurologists, physicians in general practice, nurses, seizure type, age of onset,
and health care workers. The expanded version is more detailed and will likely be probability of remission,
used more by epileptologists/neurophysiologists and researchers. EEG findings, radiologic
Seizure classification starts with whether the initial manifestations of the findings, and genetics.
seizure are focal or generalized. Focal seizures originate within a neuronal
● The 2017 International
network limited to one hemisphere that may be discretely localized or more League Against Epilepsy
widely distributed, whereas generalized seizures originate at some point within seizure classification
the brain and rapidly engage bilateral distributed networks.9 If the onset of the addresses limitations of the
1981 seizure classification,
seizure is missed or is unclear, the seizure is of unknown onset.
which include the following:
Beginning in 2010 the term focal seizure officially replaced partial seizure.9 (1) some seizure types can
The term focal seizure had been used prior to the 1981 classification system, at have either focal or
which time partial seizure had replaced focal because of concern that its use was generalized onset, (2) lack of
confusing and inferred onset in a very restricted area of brain (some seizures may knowledge about seizure
onset makes a seizure
involve large parts of a hemisphere). It is, however, now felt that the term focal unclassifiable and difficult
seizure is more widely understood. to place within the 1981
system, (3) retrospective
Focal Seizures seizure descriptions often
Focal seizures are classified according to the patient’s level of awareness and the do not include a level of
consciousness, (4) terms
first most prominent motor or nonmotor features of the seizure. These early used in the 1981 seizure
classification such as
complex partial or simple
partial are difficult to
understand, and (5) some
seizure types are not
included in the 1981
classification.
FIGURE 1-1
Basic version of 2017 International League Against Epilepsy seizure type classification.
Reprinted with permission from Fisher RS, et al, Epilepsia.1 © 2017 John Wiley and Sons.

FIGURE 1-2
Expanded version of 2017 International League Against Epilepsy seizure type classification.
prominent features are important to consider when localizing the seizure onset or
epileptogenic zone. The final feature used in classification of focal seizures is
whether the focal seizure evolves to a bilateral tonic-clonic seizure. The term
secondary generalized tonic-clonic seizure is no longer used because the term
focal seizure more completely differentiates this type from generalized seizures.
Awareness is defined as knowledge and understanding that something is
happening or exists. When a person is having a focal seizure, his or her awareness
is determined by whether the person knows who they are and what is going on
in his or her surroundings during the seizure; it does not refer to awareness of the
seizure occurring. Awareness is also distinct from responsiveness. If awareness
is impaired for any portion of the seizure, then the seizure is classified as a focal
seizure with impaired awareness.
Awareness may be considered a surrogate for consciousness. Impaired
awareness or consciousness during a seizure is likely secondary to depressed
subcortical arousal systems, leading to deep sleep activity in widespread
neocortical regions, hence the involvement of both subcortical and cortical
structures.16 A focal aware seizure replaces the previously termed simple partial
seizure, and a focal impaired awareness seizure replaces the term complex partial
seizure.1 If unknown, the level of awareness does not need to be included.
Focal motor seizures can be more specifically defined. Motor-onset
manifestations include automatisms, epileptic spasms, and atonic, clonic,
hyperkinetic, myoclonic, or tonic seizures. Automatisms are coordinated,
purposeless, repetitive motor activities that may appear normal in other
circumstances. Examples include oral automatisms such as lip smacking and
manual automatisms including repetitive hand movements such as patting (CASE 1-1).
310 APRIL 2019

Of note, automatisms can also be seen in absence seizures, which are discussed KEY POINTS
in the next section.
● Focal seizures originate
Focal atonic seizures are characterized by loss of tone in one body part. within a neuronal network
Clonic seizures are repeated, regularly spaced stereotypical jerking movements. limited to one hemisphere
Epileptic spasms were previously only considered generalized seizures. that may be discretely
Clinically, epileptic spasms present in young children with flexion of the waist localized or more widely
distributed, whereas
and flexion or extension of the arms, usually in clusters. If epileptic spasms
generalized seizures
occur in infants or early in life, they can be referred to as infantile spasms. originate at some point
Differentiating focal epileptic spasms from generalized epileptic spasms may within the brain and rapidly
require careful observation of clinical and electrographic features. Hyperkinetic engage bilateral distributed
networks.
or excessive muscular movement seizures can have variable features clinically,
including thrashing or pedaling. Focal myoclonic seizures present with jerking ● Focal seizures are
but, in contrast to clonic seizures, the jerking is irregular and not rhythmic. Tonic classified according to the
seizures refer to motor seizures with increased tone or stiffening of the limb patient’s level of awareness
or neck. and the first most prominent
motor or nonmotor features
Focal seizures with nonmotor symptoms as the first prominent feature of the seizure. These early
include autonomic, behavior arrest, cognitive, emotional, or sensory seizures. prominent features are
Autonomic seizures present with changes in heart rate, blood pressure, sweating, important to consider when
skin color, piloerection, or gastrointestinal sensations. Behavioral arrest seizures localizing the seizure onset
or epileptogenic zone. The
are characterized by cessation of movement, which should be the dominant final feature used in
feature throughout the entire seizure and not just a brief part of the seizure; classification is whether the
clinical symptoms include a blank stare and cessation from talking or moving. focal seizure evolves to a
Patients with nonmotor cognitive seizures can experience changes in language bilateral tonic-clonic
seizure.
function, thinking, or associated higher cortical functions; more specific
examples include déjà vu, jamais vu (a feeling of unfamiliarity), or hallucinations. ● Generalized seizures are
Emotional seizures appear with clear emotional changes such as dread, fear, classified according to
anxiety, or pleasure. Focal sensory seizures are classified according to changes in motor or nonmotor
sensory phenomena such as taste, smell, hearing, vision, pain, numbness, manifestations. Broadly,
motor seizures are either
or tingling. tonic-clonic or other motor
Focal seizures can be further classified as to whether they evolve to a bilateral seizures. Nonmotor
tonic-clonic seizure. As discussed previously, this classification replaces generalized seizures
secondary generalized tonic-clonic to avoid any confusion between generalized primarily refer to absence
seizures.
and focal seizures. These seizures start in one area of the brain (as with all focal
seizures) and then spread to both sides of the brain. This spread is typically
clearly seen on EEG.
Generalized Seizures
Similar to focal seizures, generalized seizures are classified according to motor
or nonmotor manifestations. Broadly, motor seizures are either tonic-clonic
or other motor seizures. Nonmotor generalized seizures primarily refer to
absence seizures.
Motor onset more specifically includes tonic-clonic, clonic, tonic, myoclonic,
myoclonic-tonic-clonic, myoclonic-atonic, atonic, or epileptic spasms. Generalized
tonic-clonic seizures generally last 1 to 3 minutes and result in immediate loss of
awareness or consciousness. The initial tonic phase is a stiffening of all limbs. The
patient may groan or cry in the beginning as air is forced past the vocal cords.
The tongue may also be bitten during this phase.
The clonic phase occurs after the tonic phase and is characterized by sustained
rhythmic jerking of the limbs. If the person has impaired breathing, he or she
may look dusky. Incontinence of either bladder or bowel occurs when the body

relaxes. A generalized clonic seizure is characterized by bilateral and sustained

rhythmic jerking. A patient with generalized tonic seizures will have stiffening of
all limbs.
In contrast to generalized clonic seizures, generalized myoclonic seizures are
associated with irregular and not necessarily synchronous bilateral jerking of
limbs, face, eyes, or eyelids. Myoclonic-tonic-clonic seizures are a new seizure
CASE 1-1 A 27-year-old man presented after experiencing an episode of loss of

awareness. His friend reported that he had been eating dinner when he
acutely stared off, which was followed by lip smacking, chewing
movements, and clenching of his
left hand, lasting a total of
90 seconds. He then appeared
confused and was back to baseline
approximately 10 minutes after
the episode began.
The patient’s past medical history
was notable for a prolonged febrile
seizure at age 18 months but was
otherwise unremarkable. He took no
medications and had no family
history of seizures. He drank two to
three glasses of wine a week. He
denied tobacco or illicit drug use.
Physical and neurologic examination
were unremarkable. Brain MRI
revealed atrophy of the right mesial FIGURE 1-3
temporal region and increased T2 Imaging of the patient in CASE 1-1.
signal in the right hippocampus Coronal MRI reveals atrophy of the
right mesial temporal region and
(FIGURE 1-3). EEG revealed right increased signal in the right
temporal slowing and epileptiform hippocampus consistent with mesial
sharp waves (FIGURE 1-4). temporal sclerosis.
COMMENT This patient’s clinical presentation was consistent with a focal seizure. His
awareness was impaired, and early prominent features included nonmotor
oral automatisms followed by motor left hand dystonia. This seizure
would, therefore, be classified as a focal impaired awareness seizure with
automatisms. The seizure combined with the MRI and EEG findings
suggests a focal epilepsy with at least a 60% risk of a having a second
unprovoked seizure within 10 years. Specifically, this patient has right
temporal lobe epilepsy, with right mesial temporal sclerosis as the etiology.
Treatment with an antiepileptic medication should be recommended, and
if he continues to have seizures despite treatment with two antiepileptic
medications, he would be considered refractory, and referral to an
epilepsy center for epilepsy surgery should be recommended.
312 APRIL 2019

designation and begin with irregular jerking on both sides followed by a
tonic-clonic seizure. Myoclonic-tonic-clonic seizures are common in juvenile
myoclonic epilepsy (CASE 1-2). Myoclonic-atonic seizures are also a new seizure
designation and are characterized by an initial irregular jerking followed by
loss of tone on both sides. These seizures are common in epilepsy with
myoclonic-atonic seizures (Doose syndrome).
FIGURE 1-4
EEG of the patient in CASE 1-1. Longitudinal bipolar montage EEG recording shows right
temporal slowing and anterior temporal spike-and-slow-wave epileptiform discharge.

Atonic seizures are brief and occur when there is bilateral loss of tone and the
muscles become limp. If the person is standing when the seizure occurs, he or she
will fall, often resulting in injury. Epileptic spasms are also brief and typically
occur in clusters with flexion at the trunk and flexion or extension of the limbs.
As with focal epileptic spasms, an EEG may be needed to distinguish whether
the seizure is generalized.
CASE 1-2 A 13-year-old girl presented for evaluation after experiencing a single
generalized tonic-clonic seizure that occurred in the morning upon
awakening. Her history revealed that for the past year, she would often
drop her toothbrush in the morning.
Her past medical history and family history were unremarkable.
Menarche had occurred at age 12. She took no medications and had no
history of alcohol, tobacco, or illicit drug use. Her physical and neurologic
examinations were unremarkable. Brain MRI was normal, and EEG
revealed generalized spike-wave discharges (FIGURE 1-5). The patient’s
parents questioned whether she had epilepsy and whether she should
be treated.
FIGURE 1-5
EEG of the patient in CASE 1-2. Longitudinal bipolar montage shows frontally predominant
3-Hz to 4-Hz generalized spike-wave discharges.
COMMENT Epilepsy is diagnosed if a person has one unprovoked seizure and at least a
60% risk of having a second unprovoked seizure within 10 years. This girl
presented with a single generalized seizure with no focal features and had
a history suggestive of morning myoclonus. The clinical presentation and
EEG were consistent with a generalized epilepsy syndrome, likely juvenile
myoclonic epilepsy. She was diagnosed with epilepsy, and antiepileptic
drug treatment was recommended.
314 APRIL 2019

Nonmotor or absence seizures include typical, atypical, myoclonic, or eyelid KEY POINTS
myoclonia. Typical absence seizures present with a sudden cessation of activity
● Seizures of unknown
sometimes with eye fluttering, head nodding, or other automatisms followed onset can be classified by
by an immediate recovery. EEG always reveals generalized spike-wave activity motor (tonic-clonic,
during the seizure. Atypical absence seizures are similar to absence seizures epileptic spasms) or
but have other features including slower onset, prolonged recovery, and more nonmotor (behavior arrest)
presentation. If information
pronounced changes in tone. A myoclonic absence seizure begins with a few
is inadequate or if the
irregular jerks followed by an absence seizure. Eyelid myoclonia is defined by seizure cannot be
jerks of the eyelids and upward deviation of the eyes. Light and closing the categorized, then the
eyes can precipitate these generalized seizures. Eyelid myoclonia with absence seizure is considered
unclassified.
seizures is seen in Jeavons syndrome.
● Similar to seizure
Unknown Seizures classification, the epilepsies
Seizures of unknown onset can be classified by motor (tonic-clonic, epileptic are classified as generalized
spasms) or nonmotor (behavior arrest) presentations. If information is or focal. The new
classification system
inadequate or if the seizure cannot be categorized, then the seizure is considered additionally recognizes two
unclassified. new categories: combined
generalized and focal
UPDATED EPILEPSY CLASSIFICATION epilepsy and unknown
epilepsy.
The second level of classification is the epilepsy type. This classification
assumes the patient has epilepsy as defined by the previously discussed updated ● Patients with generalized
definition.11 The epilepsy type is predominantly determined clinically; epilepsy have one or more
characteristic EEG findings provide supportive evidence. Similar to seizure of the generalized seizure
types, and their EEGs
classification, the epilepsies are classified as generalized or focal. The new
typically display generalized
classification system additionally recognizes two new categories: combined spike-wave activity. For
generalized and focal epilepsy and unknown epilepsy.3 individuals who have
Patients with generalized epilepsy have one or more of the generalized seizure generalized seizure types
types, and their EEGs typically display generalized spike-wave activity. For and a normal EEG, other data
are needed to determine
individuals who have generalized seizure types and a normal EEG, other data are whether the epilepsy is
needed to determine whether the epilepsy is generalized. Having myoclonic generalized.
jerks or a pertinent family history supports the diagnosis of a generalized
epilepsy type. ● Patients with one or more
focal seizure types have
Clinically, patients with one or more focal seizure types have focal epilepsy. focal epilepsy. These
These epilepsies can be either unifocal or multifocal. Although not always seen, epilepsies can be either
focal EEG findings such as focal slowing or epileptiform discharges support the unifocal or multifocal.
diagnosis of focal epilepsy. Concordant focal MRI findings are also supportive.
● Designation of combined
Designation of combined generalized and focal epilepsy is for patients with
generalized and focal
both focal and generalized seizures. EEG may reveal both focal and generalized epilepsy is for patients with
electrographic findings. Examples of combined generalized and focal epilepsy both focal and generalized
include Dravet syndrome and Lennox-Gastaut syndrome. seizures. EEG may reveal
When the patient has epilepsy as defined by the ILAE but it remains both focal and generalized
electrographic findings.
undetermined whether the patient has focal or generalized epilepsy, the Examples of combined
classification of unknown epilepsy type is used. Patients with this classification generalized and focal
may not have an available EEG or the EEG may be indeterminate. Other epilepsy include Dravet
supporting studies such as MRI and family history are also either not available syndrome and
Lennox-Gastaut syndrome.
or do not clarify the epilepsy classification.
NEW CLASSIFICATION OF EPILEPSY SYNDROME

The epilepsy syndrome is a new addition to the current classification system and
is defined as “a cluster of features incorporating seizure types, EEG, and imaging

features that tend to occur together.”3 Factors that contribute to epilepsy

syndrome include age of onset, remission, triggers, diurnal variation, intellectual
and psychiatric dysfunction, EEG findings, imaging studies, family history,
and genetics. The ILAE has never formally classified a list of epilepsy syndromes;
however, well-known and accepted syndromes are described, and some are
reviewed here. For a full list of ILAE-accepted epilepsy syndromes, refer to the
ILAE website.17
Previously, the term benign was used to describe some of the epilepsy
syndromes, but it is no longer used as it infers the epilepsy has minimal effect
on the patient. It is now more clearly understood that any epilepsy can have
social effects and can be associated with other comorbidities such as learning
disorders or psychiatric conditions. The term self-limiting is now used.
Idiopathic or Genetic Generalized Epilepsy Syndromes

Idiopathic generalized epilepsies include childhood absence epilepsy, juvenile
absence epilepsy, juvenile myoclonic epilepsy, and generalized tonic-clonic
seizures alone (TABLE 1-1). Controversy surrounds the use of the term idiopathic,
and removing it from epilepsy classification has been advocated. Idiopathic
was meant to refer to self or genetic. There is, however, concern that use of the
word genetic infers inherited, and many patients with epilepsy have de novo
mutations or have complex genetic syndromes that occur with or without
environmental factors. Many in the epilepsy community want to continue using
the term idiopathic generalized epilepsy, and the ILAE task force decided to use
it to refer to the previously mentioned epilepsies. When the clinician determines
that a clear genetic etiology is present, the term genetic generalized epilepsy may
be used to refer to the epilepsy syndrome. EEGs of the idiopathic generalized
TABLE 1-1 Idiopathic or Genetic Epilepsy Syndromes
Self-
limiting
Epilepsy Syndrome Seizure Types Age of Onset (Yes or No) EEG Findings
Childhood absence Absence, 4 to 10 years Yes Normal background, occipital intermittent

epilepsy generalized tonic- rhythmic delta activity, 3–3.5 Hz
clonic (rare) generalized spike-wave discharges
Juvenile absence Absence, Adolescence to No Normal background, polyspikes may be

epilepsy generalized tonic- early adulthood present, 3–3.5 Hz generalized spike-wave
clonic, myoclonic discharges
(rare)
Juvenile myoclonic Myoclonic, 10 years to mid- No Normal background, 3–3.5 Hz generalized

epilepsy generalized tonic- twenties spike-wave discharges, >4 Hz generalized
clonic, absence (rare) spike-wave discharges, high-amplitude
polyspike-wave discharges with
myoclonic seizures, photoparoxysmal
response in up to 40% of patients
Epilepsy with Generalized tonic- Childhood to No Normal background, generalized spike/

generalized tonic- clonic mid-adulthood polyspike-wave discharges
clonic seizures alone
316 APRIL 2019

epilepsies reveal a normal electrographic background and characteristic KEY POINTS
generalized spike-wave findings (TABLE 1-1).18
● When the patient has
Childhood absence epilepsy affects neurologically normal girls more than epilepsy as defined by the
boys and is typically self-limiting. Onset is usually between 4 and 10 years of age, International League Against
with remission usually occurring in adolescence. Patients present with absence Epilepsy, but it remains
seizures and occasionally with generalized tonic-clonic seizures. Early undetermined whether the
patient has focal or
occurrence of generalized tonic-clonic seizures is associated with a poorer
generalized epilepsy, the
prognosis. classification of unknown
Juvenile absence epilepsy has onset in adolescence and early adulthood, with epilepsy type is used.
peak onset between 10 and 13 years of age. Girls and boys are affected equally.
Absence seizures occur less frequently than in childhood absence epilepsy. ● The epilepsy syndrome is
a new addition to the current
Generalized tonic-clonic seizures occur early in the presentation, and myoclonic classification system and is
seizures, although rare, may also occur. In contrast to childhood absence defined as “a cluster of
epilepsy, this syndrome is not self-limiting. features incorporating
Juvenile myoclonic epilepsy is one of the most typical epilepsy syndromes. seizure types, EEG, and
imaging features that tend to
Onset ranges from before age 10 through the mid-twenties and later in some occur together.”
cases. Juvenile myoclonic epilepsy occurs more commonly in women. All
patients have myoclonic seizures and commonly have generalized tonic-clonic ● Idiopathic generalized
seizures. Absence seizures rarely occur. Most patients do not have spontaneous epilepsies include
childhood absence
remission and require lifelong treatment with antiepileptic medication.
epilepsy, juvenile
Epilepsy with generalized tonic-clonic seizures alone is characterized by absence epilepsy, juvenile
presentation of generalized tonic-clonic seizures with an age range of childhood myoclonic epilepsy, and
through mid-adulthood with peak onset in the second decade of life. Previously it generalized tonic-clonic
seizures alone.
was referred to as generalized tonic-clonic seizures upon awakening but was
changed after recognition that seizures can occur at any time of day. Similar to ● Reflex epilepsy
juvenile absence epilepsy and juvenile myoclonic epilepsy, epilepsy with syndromes are epilepsies in
generalized tonic-clonic seizures alone is not self-limiting, and lifelong which seizures are provoked
antiepileptic drug treatment is typically required. by a specific stimulus.
● Well-described focal
Reflex Epilepsy Syndromes epilepsy syndromes include
Reflex epilepsy syndromes are epilepsies in which seizures are provoked by a childhood epilepsy with
specific stimulus. Seizures are typically generalized tonic-clonic seizures, but centrotemporal spikes and
Panayiotopoulos syndrome.
other generalized seizure types may also occur. Rarely, focal seizures may
present as a reflex epilepsy. The most common reflex epilepsy syndrome is
photosensitive epilepsy. Other reflex epilepsy syndromes include reading
epilepsy and startle epilepsy.17,18
Focal Epilepsy Syndromes

Well-described focal epilepsy syndromes include childhood epilepsy with
centrotemporal spikes and Panayiotopoulos syndrome. Previously, childhood
epilepsy with centrotemporal spikes was referred to as benign epilepsy with
centrotemporal spikes. Childhood epilepsy with centrotemporal spikes is a
self-limited epilepsy that presents in the school years with brief focal motor
hemifacial seizures and nocturnal focal motor seizures evolving to bilateral
tonic-clonic seizures. EEG background is normal with sleep-activated
centrotemporal spikes. Panayiotopoulos syndrome is also a self-limited epilepsy
characterized by having focal autonomic seizures, often prolonged, and focal
occipital high-amplitude sleep-activated spikes seen on EEG. Possible autonomic
symptoms include vomiting, pallor, mydriasis, cardiorespiratory, gastrointestinal,
and thermoregulatory symptoms, incontinence, and hypersalivation.19

EPILEPSY ETIOLOGY
The etiology of seizures and epilepsies is emphasized in the new classification
system. In the prior classification system of the 1980s, etiology was inferred
when classifying the epilepsy. Idiopathic primarily referred to genetic causes,
symptomatic referred to the presence of a known disorder or lesion, and
cryptogenic referred to a presumed but unknown symptomatic cause. As discussed,
the term idiopathic is now used to refer to four well-described epilepsy syndromes.
The terms symptomatic and cryptogenic are no longer used.
Six etiologic categories (structural, genetic, infectious, metabolic, immune,
unknown) have been defined. When multiple potential etiologies are present,
priority should be given to the etiology with more relevant management issues.
Ongoing consideration of the etiology has clear implications for patient
management as well as research efforts as we continue to study why patients
develop seizures and we determine optimal treatments.
A structural etiology is determined when a structural abnormality is seen on
neuroimaging and when the signs and symptoms of seizures, in combination
with EEG data, suggest this abnormality is the probable cause of the seizures. If
the clinical and EEG data are discordant with localization of the visible structural
abnormality, then the imaging abnormality is not relevant to the patient’s
epilepsy. Structural abnormalities may be genetic, acquired, or both. Possible
structural abnormalities include stroke, trauma, tumor, malformations of
cortical development, and infection.
Genetic etiologies are determined if there is a known or presumed genetic
mutation in which seizures are a core symptom of the disorder.3 Genetic
epilepsies are diverse, and the list grows each year. Importantly, genetic does not
always mean inherited. Although some epilepsies are inherited, many occur
secondary to a de novo (new) mutation in the affected individual. In some cases,
the genetic mutation is not identified, but the clinical presentation, EEG findings,
and family history suggest a genetic etiology. In addition, the genetic etiology
for some epilepsy syndromes such as juvenile myoclonic epilepsy is inferred from
research studies including twin and familial aggregation studies. Overall, genetic
etiology is defined by having a known mutation, clinical presentation with
supportive data and family history, or a syndrome with evidence from research
studies to suggest a genetic etiology.
Infectious etiologies are the most common worldwide etiology. An important
distinguishing point is that the patient has epilepsy secondary to an infectious
etiology and not seizures in the setting of an acute infectious illness. Prototype
infectious etiologies include neurocysticercosis, HIV, cytomegalovirus, and
cerebral toxoplasmosis. Epilepsy onset secondary to a prior infectious insult such
as meningitis or encephalitis is also considered an infectious etiology.
Epilepsies with a metabolic etiology occur secondary to a known or presumed
metabolic disorder in which seizures are a core symptom of the disorder.3
Overlap with a genetic etiology may occur as many metabolic disorders have
known genetic mutations. Of course, identifying a genetic etiology early in
presentation is important because management interventions such as a change in
diet or supplementation can affect its natural course.
Immune etiologies are increasingly recognized as potential causes of
epilepsy. As with the other etiologies, seizures are a core symptom of the
immune disorder. In patients with identified immune etiologies, immunotherapy
should be considered. Examples of immune etiologies for seizures include
318 APRIL 2019

anti–N-methyl-D-aspartate (anti-NMDA) receptor encephalitis and anti–leucine- KEY POINTS
rich, glioma inactivated 1 (anti-LGI1) encephalitis.
● The etiology of seizures
The last etiologic category is unknown; up to one-third of patients with and epilepsies is
epilepsy have no clear etiology.20 A cause likely exists, but its identification may emphasized in the 2017
be limited by inadequate resources such as poor access to up-to-date brain International League Against
imaging, immune antibody testing, or genetic testing. It is hoped that with Epilepsy classification
system.
improved access to care as well as continued research into understanding the
epilepsies the percentage of patients who have an unknown etiology will ● A structural etiology is
diminish and that the structure of the new classification system will provide a determined when a
framework to better understand the epilepsies. structural abnormality is
seen on neuroimaging and
when the signs and
INCIDENCE OF EPILEPSIES AND ITS RELATION TO SEIZURE AND symptoms of seizures, in
EPILEPSY CLASSIFICATION AND RISK FACTORS combination with EEG data,
The incidence of epilepsy is defined as the number of new epilepsy cases over a suggest this abnormality
specified period of time. Incidence represents the number of new cases among is the probable cause of
the seizures.
susceptible persons in a given location and over a particular time span. Incidence
studies, in contrast to prevalence studies, provide a better understanding of ● Genetic etiologies are
etiology and the natural history of epilepsy.21 Epilepsy incidence studies are, determined if there is a
however, lacking and heterogeneous. Heterogeneity among reported incidence known or presumed genetic
mutation in which seizures
population studies may be addressed by use of universally adopted seizure and
are a core symptom of
epilepsy classification systems. A well-constructed classification is needed for the disorder.
comparison among studies.22 The newly proposed 2017 ILAE classification
system of seizures and the epilepsies provide a system that promotes standardization ● Infectious etiologies
of terminology, which may lead to reduced variability among incidence studies are the most common
worldwide etiology
and may, therefore, improve our understanding of seizure and epilepsy of epilepsy.
incidence. It has, however, also been argued that the system allows flexibility
as cases may be classified in different categories depending on workup, creating ● Epilepsies with a
a potential obstacle for epidemiologic studies.23 metabolic etiology occur
secondary to a known or
The proposed 2017 epilepsy classification creates a framework to more presumed metabolic
specifically define the epilepsy, epilepsy syndrome, and etiology. The emphasis disorder in which seizures
on etiology may lead to a better understanding and determination of epilepsy are a core symptom of
incidence, which, of course, depends on classification agreement among the disorder.
physicians. Universal implementation of this classification system and further
● Immune etiologies are
study will hopefully clarify the utility of the newly proposed system and whether increasingly recognized
its use will reduce heterogeneity among incidence studies and allow us to better as potential causes of
define significant factors that contribute to epilepsy incidence in all regions of epilepsy.
the world.
CONCLUSION
Seizure and epilepsy classifications have evolved since a system was first
introduced by Gastaut in the late 1960s. The 2017 ILAE classification of seizures,
epilepsies, and epilepsy syndromes aims to group seizures according to clinical
presentation and brain region onset and epilepsies according to seizure type, age
of onset, probability of remission, EEG findings, radiologic findings, and
genetics. An emphasis is now placed on etiology. Universal adoption and use of
this classification system have direct implications on our understanding of
epilepsy incidence. Multiple worldwide studies have found variable results with
marked heterogeneity. In addition, incidence studies are limited in number and
scope and rarely consider seizure type. Future studies using this standardized

classification system will potentially clarify epilepsy incidence and hopefully

reduce heterogeneity. In addition, further study is needed to evaluate the utility
of the 2017 classification system. Although it establishes standard terminology,
potential variability in coding and poor agreement among physicians may limit
its use.
USEFUL WEBSITE
INTERNATIONAL LEAGUE AGAINST EPILEPSY
The International League Against Epilepsy (ILAE)
is an international organization whose goals
include advancement and dissemination of
knowledge about epilepsy. The organization
promotes research, education, and training and
improves services and care for patients with
epilepsy. Links to published articles on seizure
and epilepsy classification are available on
this website.
epilepsydiagnosis.org
REFERENCES
1 Fisher RS, Cross JH, French JA, et al. Operational 9 Berg AT, Berkovic SF, Brodie MJ, et al. Revised
classification of seizure types by the terminology and concepts for organization of
International League Against Epilepsy: position seizures and epilepsies: report of the ILAE
paper of the ILAE Commission for Classification Commission on Classification and Terminology,
and Terminology. Epilepsia 2017;58(4):522–530. 2005-2009. Epilepsia 2010;51(4):676–685.
doi:10.1111/epi.13670. doi:10.1111/j.1528-1167.2010.02522.x.
2 Fisher RS, Cross JH, D’Souza C, et al. Instruction 10 Fisher RS, van Emde Boas W, Blume W, et al.
manual for the ILAE 2017 operational Epileptic seizures and epilepsy: definitions
classification of seizure types. Epilepsia 2017; proposed by the International League Against
58(4):531–542. doi:10.1111/epi.13671. Epilepsy (ILAE) and the International Bureau for
Epilepsy (IBE). Epilepsia 2005;46(4):470–472.
3 Scheffer IE, Berkovic S, Capovilla G, et al. ILAE
doi:10.1111/j.0013-9580.2005.66104.x.
classification of the epilepsies: position paper of
the ILAE Commission for Classification and 11 Fisher RS, Acevedo C, Arzimanoglou A, et al. ILAE
Terminology. Epilepsia 2017:58(4):512–521. official report: a practical clinical definition of
doi:10.1111/epi.13709. epilepsy. Epilepsia 2014;55(4):475–482.
doi:10.1111/epi.12550.
4 Gastaut H. Classification of the epilepsies.
Proposal for an international classification. 12 Hauser WA, Anderson VE, Loewenson RB,
Epilepsia 1969;10(suppl):14–21. doi:10.1111/ McRoberts SM. Seizure recurrence after first
j.1528-1157.1970.tb03873.x. unprovoked seizure. N Engl J Med 1982;307(9):
522–528. doi:10.1056/NEJM198208263070903.
5 Gastaut H. Clinical and electroencephalographical
classification of epileptic seizures. Epilepsia 1970; 13 Beretta S, Carone D, Zanchi C, et al. Long-term
11(1):102–113. doi:10.1111/j.1528-1157.1970.tb03871.x. applicability of the new ILAE definition of
epilepsy. Results from the PRO-LONG study.
6 Proposal for revised clinical and
Epilepsia 2017;58(9):1518–1523. doi:10.1111/
electroencephalographic classification of
epi.13854.
epileptic seizures. From the Commission on
Classification and Terminology of the 14 Falco-Walter JJ, Scheffer IE, Fisher R. The new
International League Against Epilepsy. definition and classification of seizures and
Epilepsia 1981;22(4):489–501. epilepsy. Epilepsy Res 2018;139:73–79.
doi:10.1016/j.eplepsyres.2017.11.015.
7 Proposal for classification of epilepsies and
epileptic syndromes. Commission on 15 Institute of Medicine. Epilepsy across the spectrum.
Classification and Terminology of the ilae.org/about-ilae/public-policy-and-advocacy/
International League Against Epilepsy. national-public-policy-activities/institute-of-
Epilepsia 1985;26(3):268–278. medicine-report. Accessed January 31, 2019.
8 Proposal for revised classification of epilepsies 16 Blumenfeld H. Impaired consciousness in
and epileptic syndromes. Commission on epilepsy. Lancet Neurol 2012;11(9):814–826.
Classification and Terminology of the doi:10.1016/S1474-4422(12)70188-6.
International League Against Epilepsy.
Epilepsia 1989;30(4):389–399.
320 APRIL 2019

17 International League Against Epilepsy. 20 Thomas RH, Berkovic SF. The hidden genetics of
epilepsydiagnosis.org. Accessed January 31, 2019. epilepsy-a clinically important new paradigm.
Nat Rev Neurol 2014;10(5):283–292. doi:10.1038/
18 Koutroumanidis M, Arzimanoglou A, Carabello R,
nrneurol.2014.62.
et al. The role of EEG in the diagnosis and
classification of the epilepsy syndromes: a tool 21 Abramovici S, Bagić A. Epidemiology of epilepsy.
for clinical practice by the ILAE Neurophysiology Handb Clin Neurol 2016;138:159–171. doi:10.1016/
Task Force (part 1). Epileptic Disord 2017;19(3): B978-0-12-802973-2.00010-0.
233–298. doi:10.1684/epd.2017.0935.
22 Camfield P. Issues in epilepsy classification for
19 Koutroumanidis M. Panayiotopoulos syndrome: population studies. Epilepsia 2012;53(suppl 2):
a common benign but underdiagnosed and 10–13. doi:10.1111/j.1528-1167.2012.03552.x.
explored epilepsy syndrome. Br Med J
23 Beghi E. Epilepsy: new classification of seizures
2002;324(7348):1228–1229. doi:10.1136/
and epilepsies - an advance? Nat Rev Neurol
bmj.324.7348.1228.
2017;13(6):324–325. doi:10.1038/nrneurol.2017.70.

REVIEW ARTICLE
Evaluation of Seizure
Etiology From Routine

C O N T I N UU M AUDIO
INTERVIEW AVAILABLE
Testing to Genetic
ONLINE
Evaluation
By Stephan U. Schuele, MD, MPH, FAAN
ABSTRACT
PURPOSE OF REVIEW: Recognizing the cause of a first seizure and identifying
the etiology of epilepsy are essential for management. A systematic
approach to patients who present with a first seizure helps distinguish
between an acute symptomatic seizure, a provoked or unprovoked
CITE AS:
seizure, and potential mimickers. Routine testing with EEG and MRI may
2019;25(2, EPILEPSY):322–342. reveal a predisposition for further seizures and help to establish the
underlying epilepsy syndrome. An acquired etiology can be identified in
Address correspondence to 30% of patients with established epilepsy. The remaining 70% of patients
Dr Stephan U. Schuele,
Northwestern University,
have a presumably genetic etiology. Particularly in patients with specific
Department of Neurology, epilepsy syndromes or suspicion for an autosomal dominant inheritance,
Abbott Hall #1425, 710 N Lake genetic testing and counseling should be considered.
Shore Dr, Chicago, IL 60611,
s-schuele@northwestern.edu.
RECENT FINDINGS: Neuroimaging, autoimmune antibodies, and genetic testing
RELATIONSHIP DISCLOSURE:
have revolutionized our ability to investigate the etiology of many
Dr Schuele has received personal
compensation for serving on the epilepsies. The new epilepsy classification distinguishes structural,
speaker’s bureau of Eisai Co, Ltd metabolic, genetic, infectious, and immune-mediated etiologies, which
and Sunovion Pharmaceuticals
Inc. Dr Schuele serves on the
often help determine prognosis and treatment.
board of directors of the
American Clinical SUMMARY: There is growing acceptance and demystification of the term
Neurophysiology Society and the
Epilepsy Foundation Greater
epilepsy as the most common cause for recurrent seizures. The new
Chicago, and on the editorial classification of epilepsy does not stop with the recognition of particular
board of Journal of Clinical epilepsy syndromes but aims to determine the underlying etiology.
Neurophysiology. Dr Schuele has
received research/grant support This can lead to earlier recognition of surgical candidates, a better
from the Danny Did Foundation understanding of many of the genetic epilepsies, and medical treatments
and the National Institutes of
aimed at the underlying mechanism causing the disease.
Health and has been asked to
provide expert medical
testimony in cases involving
sudden unexpected death in
INTRODUCTION
epilepsy.
E
valuating the cause of a seizure is essential to inform rational treatment
UNLABELED USE OF decisions and counseling,1 beginning with the question of whether a
PRODUCTS/INVESTIGATIONAL
USE DISCLOSURE:
spell is truly epileptic or instead may be related to some other
Dr Schuele reports no disclosure. physiologic or psychogenic paroxysmal, transient phenomenon.2 At
the next level, we need to decide whether an epileptic seizure was
© 2019 American Academy caused by an acute insult, provoked by other triggers, or related to a
of Neurology. predisposition for recurrent, unprovoked seizures, defined as epilepsy.
322 APRIL 2019

The term etiology is typically reserved for the cause of the underlying epilepsy KEY POINT
itself. Before deciding which tests are indicated in the evaluation of a patient’s
● Syncope is the most
epilepsy etiology, it is useful to classify the suspected seizure type(s) and to common mimicker of a
determine the type of epilepsy, either generalized or focal. Particularly when new-onset seizure. A
considering a genetic evaluation, it is helpful to recognize if the patient is situational trigger, an aura
presenting with a distinct epilepsy syndrome. of “going to faint,” and a
quick recovery often
The most recent classification of the epilepsies emphasizes the importance of
support a diagnosis of
searching for the underlying etiology which can be divided into structural, vasovagal syncope.
metabolic, genetic, infectious, immune, or unknown categories (FIGURE 2-1). For
a comprehensive evaluation of epilepsy, a number of comorbidities and
complications should be addressed (TABLE 2-1). Unfortunately, in many patients,
the steps to understand the cause of the seizure and the etiology of the epilepsy
are insufficient for specific categorization of the etiology, and patients end up
as yet another “seizure disorder.”
This article discusses the basic workup of patients presenting with a new-onset
seizure to determine the cause of the first seizure and whether the patient has
epilepsy. In patients with epilepsy, a comprehensive epilepsy evaluation is
needed to determine the underlying etiology. We will then discuss which
patients may benefit from advanced testing, including a genetic evaluation.
NEW-ONSET SEIZURE
The initial question when a patient presents with a first transient paroxysmal
episode is whether the event was likely epileptic or not, based on the patient’s
recollection and witnesses’ reports (FIGURE 2-2). Risk factors for provoked
seizures, such as sleep deprivation and alcohol or illicit drug exposure, need to be
FIGURE 2-1
Classifications of the epilepsies.
a
Denotes onset of seizure.
Reprinted with permission from Scheffer IE, et al. Epilepsia 2017.1 © 2017 John Wiley and Sons.

EVALUATION OF SEIZURE ETIOLOGY
explored. In approximately 20% of cases, patients may already reveal a history

of absence, myoclonic, or subtle focal aware seizures that were previously
unrecognized as signifying a new-onset epilepsy syndrome. CASE 2-1 exemplifies
the different aspects of a seizure history and the common pitfalls, such as missing
a history of prior auras or overlooking a subtle lesion on brain MRI.
The extended history and examination should determine whether the patient
has evidence of an acute symptomatic seizure (eg, a persistent focal deficit,
prolonged altered consciousness or fever) requiring urgent workup including
brain imaging, blood work, and lumbar puncture to look for infectious causes.
The examination should assess potential injuries from a convulsive seizure, such
as tongue bites, lacerations, back pain from a compression fracture, or shoulder
pain from a subluxation.
Recent developments in the definition and classification of epilepsy align the
diagnosis of epilepsy with the indication for treatment.3,4 Patients with a first
seizure, who have an epileptogenic structural abnormality on brain imaging
and/or interictal epileptiform activity on EEG, have a greater than 60% risk of
recurrence after a first unprovoked seizure and are considered as having
TABLE 2-1 Epilepsy Evaluation
Epilepsy Type: Focal or Generalized

◆ Presence of a specific epilepsy syndrome
◆ Focal: determine if frontal, temporal, or parietooccipital onset; laterality
Seizure Type
◆ Seizure description (patient and witness reports; descriptive seizure classification)
◆ EEG findings (interictal, ictal)
◆ International League Against Epilepsy seizure type (basic, advanced classification)
◆ Frequency, triggers, timing, awareness of events, risk factors
Etiology
◆ Structural
◆ Metabolic
◆ Genetic
◆ Infectious
◆ Immune
◆ Unknown
Comorbidities
◆ Nonepileptic seizures
◆ Depression, anxiety, attention deficit hyperactivity disorder
◆ Migraine
◆ Cognitive impairment
◆ Mortality, sudden unexpected death in epilepsy (SUDEP)
EEG = electroencephalogram.
324 APRIL 2019

KEY POINTS
● Laboratory testing after

a first seizure should include
a complete blood cell
count; blood chemistry,
including calcium,
magnesium, and phosphate;
thyroid-stimulating
hormone; and urine
toxicology. A 12-lead ECG
should also be considered.
● Acute brain insults can

have a risk of leading to late
seizures in up to 20% of
patients and do not warrant
prolonged seizure
prophylaxis, even if the
injury was associated with
early seizures or status.
Exceptions are possibly a
penetrating head trauma
and herpes encephalitis,
which are associated
with a 50% risk of
developing epilepsy.
FIGURE 2-2
Systematic approach to patients with new-onset seizure.
CT = computed tomography; EEG = electroencephalogram; MRI = magnetic resonance imaging.
Modified with permission from Gavvala JR, Schuele SU, JAMA.2 © 2016 American Medical Association.
epilepsy. Those patients typically benefit from antiepileptic medication, which

is also considered for patients after a first nocturnal convulsion.5
Patients with a provoked seizure are typically not treated but are counseled
about lifestyle modifications and, if necessary, referred for drug and/or alcohol
rehabilitation. Acute symptomatic seizures are often treated with temporary
antiepileptic coverage for a variable amount of time, depending on how quickly
the underlying process resolves or the acute trigger can be corrected. Acute brain
insults can have a risk of leading to late seizures in up to 20% of patients and do
not warrant prolonged long-term seizure prophylaxis, even if the injury was
associated with early seizures or status. Exceptions are possibly a penetrating
head trauma and herpes encephalitis, which are associated with a 50% risk of
Brain Imaging
Patients with a first seizure should undergo neuroimaging. Urgent imaging should
be performed in any patients with a new neurologic deficit, persistent altered
mental status, recent trauma, or prolonged headache. CT scan is often used as the
first imaging modality because of its ease of access and should be considered for

CASE 2-1 A 41-year-old right-handed woman presented for evaluation at an

epilepsy center. Three weeks earlier she had collapsed and convulsed on
a soccer field during a hot summer day. She remembered being at the
field watching her son play, walking toward the other side of the field,
and then losing consciousness. She was unaware of what happened when
she arrived in the emergency department. There was no witness
report, and she was thought to have had a syncopal event given the
circumstances as well as the results of the normal EEG and brain CT in the
emergency department.
During follow-up in an epilepsy center, it was revealed that the patient
had asked a witness who had seen the event, who described how the
patient collapsed to the ground, became rigid, and then started
convulsing for more than a minute. Afterward, the patient appeared
confused and incoherent.
The patient remembered on further questioning that she had a really
weird feeling of sudden, strong fear for about 10 seconds right before she
passed out. The patient had this sudden, distinct fear sensation a few
times before, 1 to 2 times per year, over the last 4 years. She was typically
able to calm herself down with a few deep breaths, usually lasting for a
minute or less. During some of these episodes, she had told her husband
that she felt weird, and when he checked on her, she seemed to always
be able to respond. This had happened while running but also when
resting.
Brain MRI had been done in the interim and was initially interpreted as
normal. On further review, her brain MRI was found to demonstrate a
6-mm ovoid signal abnormality in the right middle temporal gyrus
(FIGURE 2-3). Within the subcortical white matter of the right middle
temporal gyrus there was a region of T2/fluid-attenuated inversion
recovery (FLAIR) hyperintense signal that had a smaller cystic component
with a hypointense rim. This likely represented a focal cortical dysplasia
or glioneuronal tumor. Subsequent 36-hour ambulatory EEG was read
as normal.
The patient’s epilepsy was classified as right temporal lobe epilepsy
with the symptomatology of psychic aura preceding a bilateral tonic-
clonic seizure. According to the International League Against Epilepsy,
this would be classified as a focal aware emotional seizure progressing to
a bilateral tonic-clonic seizure. The frequency was one to two auras per
year with an isolated convulsion. The etiology was identified as right
middle temporal focal cortical dysplasia versus low-grade tumor, and
there were no comorbidities.
There were no further events, and the seizures were controlled with
levetiracetam. The patient underwent resective surgery to confirm the
etiology and ensure the underlying lesion was not a malignant tumor.
Intraoperative electrocorticography found spikes anterior to the lesion.
Surgical pathology demonstrated a ganglioglioma with surrounding tissue
showing focal cortical dysplasia type IIIb. The patient remained
seizure free on medication.
326 APRIL 2019

FIGURE 2-3
Right middle temporal gyrus lesion in the patient in CASE 2-1. A, Coronal T2-weighted turbo
spin echo. B, Coronal fluid-attenuated inversion recovery (FLAIR). C, Axial T2-weighted
turbo spin echo. D, Axial FLAIR. See case text for details.
This case illustrates the different aspects of a seizure history and common COMMENT
pitfalls, such as missing a history of prior auras or overlooking a subtle lesion
on brain MRI and the option for early surgery to understand the etiology and
distinguish benign versus potentially malignant underlying pathology in
selected candidates.

patients with new-onset seizures seen in the emergency department to assess

for an acute brain insult, such as a stroke, bleed, or traumatic injury, often
followed by a more accurate brain MRI performed with and without contrast.6,7
CT scans and standard MRI protocols often fail to detect the lesions associated
with chronic epilepsy; for this reason, an MRI should be performed using an
epilepsy-specific protocol. These lesions typically do not require acute
interventions and are arguably better evaluated during subsequent follow-up.
The use of an epilepsy-protocol MRI, with communication to the reader of the
suspected seizure focus before reading of the study, and review of the imaging
results by an expert neuroradiologist doubles the sensitivity in detecting the
often subtle abnormalities that may be amenable to epilepsy surgery.8,9
Electroencephalogram
Emergency EEG is indicated for patients who do not have a timely recovery after
a seizure, have fluctuating mental status changes, or show a neurologic deficit
that is not explained by the imaging findings. A routine EEG is insufficient to
CASE 2-2 A 36-year-old right-handed woman found to have an incidental

hypophyseal brain aneurysm underwent elective clipping via left pterional
craniotomy. She was awake and alert postoperatively, and the next day,
she was witnessed to have staring episodes and confusion. Her brain CT
demonstrated a small parenchymal hemorrhage in the posterior inferior
left frontal lobe and a small amount of subarachnoid blood. The patient
was started on levetiracetam. Video-EEG was initiated, which showed
evidence of left hemispheric dysfunction and moderate encephalopathy
during the first 2 recording days. On the third day, four electrographic
seizures arising from the left temporal area were seen without a clear
clinical correlate. Lacosamide was added, the seizure activity subsided,
and the patient improved quickly. Left temporal sharp waves were noted,
often in a semiperiodic pattern (FIGURE 2-4). The left temporal sharp wave
activity persisted during the subsequent 3 recording days, although with
diminishing frequency, and the patient was discharged.
The patient was seen in the epilepsy clinic 4 weeks later; she was
seizure free and still taking both medications. An EEG was performed at
that time, which showed resolution of the epileptiform activity. Both
seizure medications were sequentially tapered over a period of 2 months,
first by withdrawing levetiracetam, then lacosamide. The patient was
seen in follow-up 7 months postoperatively and has remained
seizure free.
The patient’s condition was classified as acute symptomatic seizures
with the symptomatology of an altered awareness/responsiveness
seizure. According to the International League Against Epilepsy, this would
be classified as a focal impaired awareness seizure. The frequency was
several per day, but after postoperative day 3, the patient was seizure
free. The etiology was identified as post–left pterional craniotomy for
aneurysm clipping and a small left inferior frontal hemorrhage with
subarachnoid blood, and there were no comorbidities.
328 APRIL 2019

detect subclinical seizure activity, which is captured within the first 30 minutes
in only one-fourth of patients found to have subclinical seizures on long-term
monitoring and in the first 2 hours in one-half of those patients. The yield
increases to greater than 90% with 24- to 36-hour continuous recordings.
CASE 2-2 illustrates the need for prolonged monitoring to capture subclinical
seizures in a patient with an acute symptomatic seizure after brain surgery,
but also emphasizes that the initial EEG may not be appropriate to assess the
long-term risk for subsequent development of epilepsy.
EEG recording after a new-onset seizure can show interictal epileptiform
activity and help determine the potential seizure type and recurrence risk after
a first event. Of patients with new-onset seizures, 29% will have epileptiform
abnormalities on EEG.6 An initial EEG is particularly valuable for patients with
a predisposition for generalized seizures. Those patients tend to have a higher
frequency of epileptiform discharges compared with focal epilepsies and a
typical response to activation with hyperventilation and photic stimulation. The
EEG in generalized epilepsies may normalize with appropriate treatment, and it
FIGURE 2-4
EEGs of the patient in CASE 2-2. A, EEG on postoperative day 4 with prominent left temporal
sharp waves and spikes, often in semiperiodic runs, during stage II sleep. B, EEG 1 month
after surgery. Asymmetry and continuous slow over the left temporal region as sequelae of
the surgery during stage I sleep; no epileptiform abnormalities are seen.
This case illustrates the need for prolonged monitoring to capture COMMENT
subclinical seizures in a patient with an acute symptomatic seizure after
brain surgery. It also emphasizes that EEGs in the early phase may not be
appropriate to assess the long-term risk of epilepsy. The patient remained
seizure free after discontinuing medication despite periodic focal sharp
waves in the acute seizure setting.

is important to obtain a pretreatment study. There does appear to be a slightly

higher yield of epileptiform abnormalities on EEGs performed in patients within
24 to 48 hours of a new-onset seizure. However, acute EEG changes may reflect
the sequelae of the acute insult to the brain or aftermath of the seizure and may
not necessarily be predictive of the risk of late seizure recurrence (CASE 2-2).
Patients presenting with a provoked seizure should get at least an EEG and a
brain imaging study because not infrequently they are found to have a remote
structural abnormality or an EEG indicating a predisposition for generalized
seizures.
EPILEPSY EVALUATION
The evaluation in a patient with established epilepsy differs in many aspects from
that of a first seizure. In particular, chances to have a reliable witness report for
the events are higher, and the yield of EEG to capture interictal epileptiform
abnormalities increases. In patients with frequent events, video-EEG recording
can be considered, and brain imaging needs to focus on structural abnormalities
amenable to surgery.
Seizure Evaluation
Patients with established epilepsy may have several seizure types. The interview
should provide guidance for the patient to describe the seizures in his or her
own words, to characterize different seizure types associated with specific
epilepsy syndromes, or recognize that focal seizures often evolve from an aura to
loss of awareness and sometimes convulsions. A variety of sources for witness
reports may be available, and the patient or family may have already recorded an
event on video. It may also become apparent when speaking to family members
and other witnesses that the patient is not aware of some seizures. For patients
with epilepsy who continue to have seizures, we want to know if the seizure was
caused by some trigger or nonadherence to medication or is indicative of an
incomplete medication response requiring medication adjustment. Seizures may
be seen mostly during certain times of the day or triggered by specific situations,
suggesting a form of rare reflex epilepsy.
The physician should use a systematic descriptive vocabulary to capture the
essential components of the clinical seizure and its evolution (FIGURE 2-5).10
Many epilepsy centers use this vocabulary to create a seizure classification based
purely on the clinical symptoms and observed behavioral signs, which is helpful
to localize the seizure onset and to capture all essential components of a seizure
and its evolution. This independence from other tests allows a seizure description
without having to know the EEG onset pattern, which is often more practical.11
CASES 2-1 and 2-2 demonstrate the use of a systematic descriptive classification in
parallel with the International League Against Epilepsy (ILAE) classification.
The ILAE seizure classification was revised in 2017.12,13 The terms for each
seizure type indicate the suspected or confirmed EEG onset (focal versus
generalized) followed by a brief descriptor of the patient’s symptoms, creating
an electroclinical entity. The basic version divides focal aware seizures and
focal impaired awareness seizures (FIGURE 2-6). Focal and generalized seizures
can be classified as motor and nonmotor. A more extended version allows
additional descriptors.
Approximately 70% to 80% of patients with chronic focal epilepsies
eventually develop interictal epileptiform discharges. The yield is noticeably
330 APRIL 2019

KEY POINTS
● Typical MRI lesions

associated with surgically
amenable epilepsy, such as
glioneuronal tumors,
low-grade gliomas,
hippocampal sclerosis,
small cavernous
malformations, and subtle
malformations of cortical
development (eg, cortical
dysplasia or isolated
periventricular nodular
heterotopia), are not seen
on brain CT scans and often
missed on standard MRIs.
● An epilepsy-protocol
FIGURE 2-5
brain MRI differs from a
Glossary of descriptive terminology for ictal symptomatology.
a
Autonomic symptoms can be verifiable for the observer (eg, tachycardia, goosebumps).
typical MRI in that it includes
Data from Blume WT, et al, Epilepsia.10 thin 1- to 3-mm slices
without interslice gap and
coronal fluid–attenuated
inversion recovery
higher in long-standing focal epilepsies compared with first-time seizures. The sequences, which offer
yield to capture the presence of discharges depends on the duration of the additional sensitivity over
recording. Today, most patients and physicians prefer the convenience of an standard-protocol MRIs.
ambulatory EEG and the ability to request interpretation by an expert reader
● A brain CT and an EEG
over repeat routine EEGs. In a study by Faulkner and colleagues,14 in patients
should be considered even
with epilepsy found to have interictal epileptiform abnormalities during 4 days in the presence of an
of EEG recordings, interictal epileptiform discharges were recorded in 44% alternative etiology for a
of patients within 4 hours, 58% within 8 hours, 85% within 24 hours, and 95% provoked seizure.
within 48 hours of recording. Recording for the full 96-hour period revealed
only 5% new interictal epileptiform discharges. The median latency to the
first discharge was significantly shorter in patients with generalized epilepsies
(43 minutes) compared with focal epilepsies (512 minutes for extratemporal
and 590 minutes for temporal lobe epilepsies), indicating a 2-day recording
length as most adequate for looking for interictal epileptiform activity
and a potentially shorter recording length in suspected generalized epilepsies.
FIGURE 2-6
International League Against Epilepsy classification of seizure types.

TABLE 2-2 Electroclinical Syndromes Arranged by Age at Onseta
Neonatal Period
◆ Benign familial neonatal epilepsy
◆ Early myoclonic encephalopathy
◆ Ohtahara syndrome
Infancy
◆ Epilepsy of infancy with migrating focal seizures
◆ West syndrome
◆ Myoclonic epilepsy in infancy
◆ Benign infantile epilepsy
◆ Benign familial infantile epilepsy
◆ Dravet syndrome
◆ Myoclonic encephalopathy in nonprogressive disorders
Childhood
◆ Febrile seizures plus (can start in infancy)
◆ Panayiotopoulos syndrome
◆ Epilepsy with myoclonic atonic (previously “astatic”) seizures
◆ Self-limiting epilepsy with centrotemporal spikes
◆ Autosomal dominant nocturnal frontal lobe epilepsy
◆ Late-onset childhood occipital epilepsy (Gastaut type)
◆ Epilepsy with myoclonic absences
◆ Lennox-Gastaut syndrome
◆ Epileptic encephalopathy with continuous spike-and-wave during sleep
◆ Landau-Kleffner syndrome
◆ Childhood absence epilepsy
Adolescence to Adulthood
◆ Juvenile absence epilepsy
◆ Juvenile myoclonic epilepsy
◆ Epilepsy with generalized tonic-clonic seizures alone
◆ Progressive myoclonic epilepsies
◆ Autosomal dominant epilepsy with auditory features
◆ Other familial temporal lobe epilepsies
Less Specific Age Relationship
◆ Familial focal epilepsy with variable foci (childhood to adult)
◆ Reflex epilepsies
a
Modified with permission from Berg AT, et al, Epilepsia.16 © 2010 John Wiley and Sons.
332 APRIL 2019

In most patients, particularly if they respond to medical therapy and KEY POINT
remain seizure free, a confirmation of the seizure and epilepsy type through
● According to national
symptomatology and interictal EEG is sufficient for management. However, guidelines, patients who
in patients whose seizures remain resistant to medication after 1 year, an continue to have seizures
evaluation in the epilepsy monitoring unit is recommended to understand the after 1 year on medical
actual cause of the seizures. Approximately 25% of patients considered therapy should be referred
to an epilepsy center.
intractable are found to have nonepileptic seizures, and 10% to 20% of patients
with epilepsy also have nonepileptic events. Some patients with generalized
epilepsies will not be recognized unless they are monitored off medication for
several days. Last but not least, there is the considerable number of patients
with focal epilepsy who are candidates for epilepsy surgery who benefit from
a presurgical evaluation in the epilepsy-monitoring unit.15
Epilepsy Type
The next step in the evaluation is to determine the type of epilepsy: focal,
generalized, or unknown. In the ILAE classification, the seizure type and
epilepsy type largely overlap with each other; however, there is recognition of a
subgroup of patients (eg, those with Lennox-Gastaut syndrome or Dravet
syndrome) who can have focal and generalized onset seizures and present
with a combined generalized and focal epilepsy. In focal epilepsies, the epilepsy
type should be further divided into the anatomic region of suspected seizure
origin based on seizure description, imaging, and interictal EEG investigations or
confirmed by ictal video-EEG–recorded seizures and accordingly classified as left
or right, temporal, frontal, or parietooccipital lobe epilepsy.
Epilepsy Syndrome
A variety of distinct epilepsy syndromes have been described, many with onset
in childhood or adolescence (TABLE 2-2).16 The recognition of these syndromes
is not only important for treatment and prognosis, it also guides genetic
evaluations because many of the established epilepsy syndromes have a likely
genetic etiology.
Etiology of Epilepsy
The etiology of epilepsy can be divided into six different categories: structural,
metabolic, genetic, infectious, immune, and unknown.1,16 The etiologies are not
mutually exclusive, allowing a structural-genetic cause for patients with tuberous
sclerosis or a metabolic-genetic cause for patients with epilepsy associated with
inborn errors of metabolism. Genetic generalized epilepsies include four epilepsy
syndromes still referred to as idiopathic generalized epilepsies: childhood
absence epilepsy, juvenile absence epilepsy, juvenile myoclonic epilepsy, and
sporadic generalized tonic-clonic seizures alone. Genetic generalized epilepsies
also include the group of patients with epileptic encephalopathies, which are
often due to a specific genetic cause.
STRUCTURAL. Most patients with epilepsy require a high-resolution epilepsy-

protocol brain MRI to assess the possibility of a structural etiology. An exception
can be made in patients with a convincing history and classical EEG findings
of an idiopathic generalized epilepsy. The presence of a resectable, epileptogenic
lesion should trigger a surgical evaluation in patients with medication-resistant

epilepsy. In patients with intractable focal epilepsy and a normal MRI, a

fludeoxyglucose positron emission tomography (FDG-PET) scan is useful in
looking for a hypometabolic region.
METABOLIC. Epilepsy associated with inborn errors of metabolism usually has

the features of an epileptic encephalopathy with seizures arising at an early age,
often refractory to antiepileptic drugs, and associated with severe cognitive,
sensory, and/or motor impairment.17 Most metabolic epilepsies have a presumed
genetic basis.1 They are typically categorized according to the biochemical
pathway involved and the age of onset. The presence of other neurologic
abnormalities, such as movement disorders and ataxia, systemic involvement,
parental consanguinity, and a positive family history of a similar illness, can be
clues.18 The general examination should look for dysmorphic features and an
abnormal head circumference. Skin and hair abnormalities can point toward
Menkes syndrome or a biotinidase deficiency. A formal ophthalmologic
evaluation is essential in identifying signs of pigmentary retinopathy (neuronal
ceroid lipofuscinosis, mitochondrial disorders), macular cherry-red spots
(gangliosidosis, Niemann-Pick disease), lens dislocation (sulfite oxidase
deficiency), and cataracts (serine biosynthesis defects). A selected number of
inborn errors of metabolism can present in adolescence and adulthood
(TABLE 2-3) and require a systematic evaluation for biochemical abnormalities
and genetic causes (FIGURE 2-7).
TABLE 2-3 Metabolic Epilepsies Presenting in Adolescence or Adulthood
Progressive Myoclonic Epilepsies

◆ Ceroid lipofuscinosis
◆ Sialidosis type 1
◆ Gaucher disease type III
◆ Mitochondrial cytopathies
◆ Lafora disease
◆ Unverricht-Lundborg disease
With Intellectual Disability
◆ Disorders of creatine metabolism
◆ Mitochondrial disorders
◆ Glucose transporter type 1 deficiency
◆ Urea cycle defects
◆ Organic acidemias
◆ Succinic semialdehyde dehydrogenase deficiency
◆ Lysosomal storage disorders (juvenile Niemann-Pick disease type C)
Without Intellectual Impairment
◆ Wilson disease
◆ Porphyria
334 APRIL 2019

FIGURE 2-7
Epilepsy associated with inborn errors of metabolism.
EEG = electroencephalogram; MRI = magnetic resonance imaging; MRS = magnetic resonance
spectroscopy; 5-MTHF = L-methyltetrahydrofolate.
Reprinted with permission from Sharma S, Prasad AN, Int J Mol Sci.18 © 2017 MDPI AG, Basel, Switzerland.
INFECTIOUS. Infections trigger not only acute symptomatic seizures during the
initial infection but may also lead to late seizures remote from the initial infection
onset and its resolution that are instead consistent with epilepsy. Infectious
etiologies include neurocysticercosis, viral encephalitis, bacterial meningitis,
fungal infection, tuberculosis, toxoplasmosis, malaria, and HIV.
Neurocysticercosis is more prevalent in developing countries and is often the
leading cause of epilepsy in those countries. Seizures can occur in any stage of
the cyst formation. Seizures in the acute phase of a bacterial meningitis are an
ominous sign and associated with increased mortality. Late epilepsy is seen
in approximately 5% to 10% of survivors after bacterial meningitis.19 Viral
encephalitis–related epilepsy is most commonly seen with herpes simplex type 1,
which develops in 50% of patients. Acute seizures are a presenting symptom
of herpes simplex type 1 in a similar proportion. Cytomegalovirus is the most
common fetal viral infection and can cause malformations of cortical
development and calcifications, with seizures typically presenting in the first
month of life. Subacute sclerosing panencephalitis is a chronic, progressive
disorder associated with an initial measles infection before the age of 2 years.
Acute symptomatic seizures are seen in influenza B, varicella, measles, mumps,
rubella, and West Nile virus, but there is limited information about how often the
acute viral encephalitis leads to late seizures.20 Approximately 50% of congenital
Zika virus cases are associated with epilepsy, typically presenting in the first

months of life.21 Tuberculous vasculitis and also cortical tuberculomas can lead to
epilepsy. Seizures in cerebral toxoplasmosis are typically associated with the
reactivation of a latent infection in immunocompromised individuals.
INFLAMMATORY. Increasingly, autoimmune antibody syndromes are recognized

as a cause of epilepsy.22 The majority of these antibodies are directed against
neuronal cell surface antigens, including synaptic neurotransmitter receptors,
ion channels, or related proteins (TABLE 2-4). A second group has antibodies
specific for intraneuronal nuclear or cytoplasmic antigens. Both groups have
an association with cancer risk, which is as high as 90% in the patients with
nuclear or cytoplasmic antibodies. Thyroid antibodies seen in patients with
steroid-responsive encephalopathy and patients with high levels of glutamic
acid decarboxylase 65 antibodies can present with immune-mediated epilepsy.
Patients with autoimmune epilepsy often present with a high frequency of
seizures and/or treatment-refractory status epilepticus, which is discussed in
“Epilepsy Emergencies: Status Epilepticus, Acute Repetitive Seizures, and
Autoimmune Encephalitis” by Stephen VanHaerents, MD, and Elizabeth E.
Gerard, MD, in this issue of Continuum.23 Aside from high seizure frequency and
treatment resistance, patients often show progressive mental status change,
neuropsychiatric symptoms, autonomic dysfunction, a viral prodrome, and
TABLE 2-4 Autoimmune Antibodies Associated With Epilepsy
Antinuclear and Cytoplasmic Antibodies

◆ Antineuronal nuclear antibody type 1 (ANNA-1) (anti-Hu)
◆ Antineuronal nuclear antibody type 2 (ANNA-2) (anti-Ri)
◆ Antineuronal nuclear antibody type 3 (ANNA-3)
◆ Antiglial neuronal antibody type 1 (AGNA) (SOX1)
◆ Purkinje cell antigen type 1 (PCA-1) (anti-Yo)
◆ Purkinje cell antigen type 2 (PCA-2)
◆ Purkinje cell antigen type Tr (PCA-Tr)
◆ Collapsin response mediator protein-5 (CRMP-5) (anti-CV2)
◆ Amphiphysin
◆ Glutamic acid decarboxylase 65 (GAD65)
Plasma Membrane Antibodies
◆ N-methyl-D-aspartate (NMDA) receptor
◆ Voltage-gated potassium channel (VGKC) complex
◆ Leucine-rich glioma inactivated 1 (LGI1)
◆ Contactin-associated protein-like 2 (Caspr2)
◆ α-Amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptor
◆ γ-Aminobutyric acid (GABA)-B receptor
◆ P/Q and N-type calcium channel
◆ Neuronal ganglionic AChR
◆ IgLON5 neuronal cell surface protein
336 APRIL 2019

sometimes facial or faciobrachial dyskinesias. Testing may show CSF findings KEY POINT
consistent with inflammation, increased signal in the mesial temporal lobe on
● Genetic testing should be
T2-weighted images, and associated malignancies on screening.24 In patients considered in all patients
with longer-standing epilepsy who may not have initially presented as an with an autosomal dominant
emergency, the yield of a positive antibody test result can be predicted through familial epilepsy, epileptic
an Antibody Prevalence in Epilepsy (APE) score of greater than 4, which has a encephalopathies, and
progressive myoclonic
sensitivity of 82.6% and a specificity of 82.0%.24 The score was validated in a
epilepsy and patients with
large cohort of patients with epilepsy and had a sensitivity of 97.7% in patients clinical findings suggestive
with epilepsy of unknown etiology to predict the presence of an autoimmune of a genetic etiology.
antibody, which was seen in 43 of 87 patients.25
GENETIC. Only approximately 30% of epilepsies are acquired, and the remaining
70% are likely caused by one or more genetic factors.26 The genetics of epilepsies
follow a complex pattern, and most genetically determined epilepsies have a
polygenic basis with several susceptibility genes contributing to the disease.27
Next-generation sequencing technology has revolutionized gene discovery in
epilepsy and many other disorders.28
Comparative genomic hybridization microarray (FIGURE 2-8) is typically
the first test for patients presenting with epilepsy and developmental delay,
intellectual impairment, and/or dysmorphic features. In patients with genetic
generalized epilepsy and intellectual disability, genomic hybridization
microarray is able to detect a copy number variant in 28%. Copy number variants
have been identified in approximately 1% of the idiopathic generalized epilepsies.
Phenotypic and genetic variability is seen in most genetic epilepsies, and
single-gene testing has been largely replaced by gene panels and whole exome
FIGURE 2-8
Genetic testing. Comparative genomic hybridization microarray is typically the first test for
patients presenting with epilepsy and developmental delay. Single-gene testing has been
largely replaced by gene panels and whole exome sequencing. Gene panels are either
targeted for a specific phenotype, eg, a progressive myoclonic epilepsy, or offered as a
comprehensive epilepsy gene panel. Whole exome sequencing and whole genome
sequencing are usually performed within the setting of a genetic clinic.
Courtesy Gemma Carvill, PhD, Northwestern University.

sequencing. Gene panels are either targeted for a specific phenotype, eg, a
progressive myoclonic epilepsy, or offered as a comprehensive epilepsy gene
panel. Whole exome sequencing and whole genome sequencing are usually
performed within the setting of a genetic clinic and research program with access
to genetic counseling and the ability to interpret variants of unknown significance.
Dravet syndrome is the best-known epileptic encephalopathy with a distinct
clinical presentation associated with a pathogenic variant in the SCN1A gene in
80% of patients. Dravet syndrome typically presents with early, prolonged
febrile seizures, which are often provoked by modest hyperthermia. Generalized
tonic-clonic, tonic, and hemiconvulsive seizures are most common. Myoclonic
seizures develop later in the course. Patients often show cognitive dysfunction,
ataxia, and psychomotor regression and demonstrate attention deficit
hyperactivity disorder–like behavior. However, variable penetrance and
phenotypic severity make clinical characterization challenging. Interestingly,
10% of families with generalized epilepsies with febrile seizures plus (GEFS+) are
positive for SCN1A, and family members are at risk of presenting with more
severe epilepsies, such as myoclonic-astatic seizures or Dravet syndrome. A
variety of other genes have been reported to cause Dravet syndrome–like
disorders, including SCN2A, SCN8A, SCN9A, SCN1B, PCDH19, GABRA1,
GABRG2, STXBP1, HCN1, CHD2, and KCNA2, requiring gene panel testing or
whole exome sequencing. Epileptic encephalopathies other than Dravet
syndrome have been associated with mutations in CDKL5, STXBP1, SCN2A,
SCN8A, KCNQ2 genes; most are de novo mutations. Recessive or mitochondrial
or more complex inheritance is rare and includes X-linked PCDH19
encephalopathy in girls with normal male carriers.
The clinical neurologist should be familiar with a small group of genetic
epilepsies that present with a rather distinct clinical phenotype and if clinically
indicated can be evaluated with commercially available gene panels. Notable
genetic epilepsies include the autosomal dominant focal epilepsies:
u Autosomal dominant temporal lobe epilepsy

u Autosomal dominant nocturnal frontal lobe epilepsy
u Familial focal epilepsy with variable foci
Autosomal dominant temporal lobe epilepsy is associated with mutations in

the LGI1 (30%) and RELN gene and has a 55% to 78% penetrance. Patients
present with distinct auditory auras or receptive aphasia that may progress to
tonic-clonic seizures. Onset is typically in adolescence or early adulthood, and
patients typically have normal intellect and good antiepileptic drug responses.
Autosomal dominant nocturnal frontal lobe epilepsy is associated with neuronal
nicotinic acetylcholine receptor mutations, which are seen in 20% of patients
with this disorder. A more severe form with intellectual disability is associated
with KCNT1 mutation. Penetrance is approximately 70%, and seizures typically
present in childhood to adolescence, characterized by clusters of nocturnal
seizures, which can be tonic, dystonic, or hypermotor, with 70% preceded by
nonspecific aura. The diagnosis is clinical, and the EEG often obscured or
normal. Patients respond well to carbamazepine and zonisamide, but 30% are
drug resistant. Familial focal epilepsy with variable foci is associated with
mutations in the DEPDC5, NPRL2, and NPRL3 genes, all part of the guanosine
triphosphatase (GTPase)–activating protein activity toward Rag (GATOR)
338 APRIL 2019

complex, accounting for at least 10% of familial frontal lobe epilepsies. Patients KEY POINTS
show variable foci and a 50% to 80% penetrance. Onset is variable from 1 month
● Whole exome sequencing
to 51 years of age with an average of 12 years. Long periods of remission in the and whole genome
teen/adult years are noted. sequencing are usually
It is worthwhile to be familiar with glucose transporter type 1 (GLUT1) performed within the setting
deficiency syndromes associated with SLC2A1 gene mutation given the excellent of a genetic clinic and
research program with
response to dietary treatment and the recent discovery of milder, later-onset
access to genetic counseling
variants.29 The typical presentation is in the first months of life. Children often and the ability to interpret
present with seizure clusters and status epilepticus during fasting (eg, before variants of unknown
breakfast). Associated features are often severe psychomotor retardation, dystonia, significance.
ataxia, and acquired microcephaly. The syndrome can present as early-onset
● It is worthwhile to be
absence epilepsy and myoclonic-astatic epilepsy. A diagnostic clue is a low fasting familiar with glucose
CSF glucose level. Excellent responses to ketogenic diet are seen. transporter type 1 (GLUT1)
Periventricular nodular heterotopia associated with a filamin A mutation is deficiency syndromes
important to recognize given the implication for prenatal counseling in this associated with SLC2A1 gene
mutation given the excellent
X-linked dominant disorder. It affects female patients who typically show response to dietary
normal or borderline intelligence and presents with seizures in close to 90%. treatment and the recent
Forty-nine percent of classic bilateral periventricular nodular heterotopia cases discovery of milder,
are associated with X-linked filamen A mutations, which are perinatally lethal in later-onset variants.
affected males. As a consequence, half of male offspring perish, and half of
● Genetic counseling for a
female offspring are affected. specific epilepsy syndrome
Understanding the genetic etiology of an individual with epilepsy has a can be straightforward in
number of clinical implications. A genetic diagnosis provides closure and can patients with a 100%
penetrant syndrome but
avoid unnecessary testing or lead to important screening for known complications
complicated in situations of
or malignancies. There is the hereditary aspect for some genetic epilepsies, incomplete penetrance and
which affects prognosis and counseling. Mitochondrial disorders presenting de novo mutations (eg,
with epilepsy may not be recognized until adulthood and the diagnosis has Dravet syndrome) and
a direct impact on treatment and counseling of the patient and family members.30 should be performed
with the help of a
Discovering a specific gene abnormality can also open the door for precision genetic counselor.
medicine. A gene abnormality may not only cause the disease but also be
associated with a specific mechanism leading to seizures amenable to specific ● The high prevalence of
treatment. SCN1A-associated Dravet syndrome responds well to medications comorbidities associated
with epilepsy is best
such as clobazam and valproate and can worsen with sodium channel blockers, addressed within a
whereas SCN8A-associated Dravet syndrome may benefit from sodium channel comprehensive care
blockers and PCDH19-related Dravet syndrome may respond well to steroids.31 setting including a social
The mammalian target of rapamycin (mTOR) signaling pathways affect worker, psychologist,
psychiatrist,
regulation of cell growth and proliferation. Mutations in the mTOR pathway can
neuropsychologist,
lead to a spectrum of malformations of cortical development ranging from focal dietician, and
cortical dysplasia to hemimegalencephaly and megalencephaly, or can cause genetic counselor.
multisystem disorders such as tuberous sclerosis complex or hamartoma
syndromes including Cowden and Lhermitte-Duclos disease.32 Several mTOR
regulatory genes, DEPDC5, MTOR, NPRL3, PI3KCA, and PTEN, are associated
with hemimegalencephaly and lesional and nonlesional focal cortical dysplasia
and can present as germline or somatic mutations in the brain. In
polyhydramnios, megalencephaly, and symptomatic epilepsy syndrome
(PMSE), an autosomal-recessive disorder associated with mutations in the
STRADA gene, treatment with the mTOR inhibitor rapamycin prevents the
occurrence of seizures. In tuberous sclerosis complex, seizure control with
everolimus, a synthetic mTOR inhibitor, has been more variable, suggesting that
treatment before seizure onset may be crucial for prevention of the epilepsy.32

Physicians treating patients with epilepsy are frequently asked about the
inheritance risk of epilepsy. In the absence of a family history or evidence of a
specific genetic syndrome, the epilepsy risk for the general population is estimated
at approximately 1%; for a child of a mother with epilepsy, 2.8% to 8.7%; for a
child of a father with epilepsy, 1.0% to 3.6%. It increases if the parent’s epilepsy
started before the age of 20 years.33 Genetic counseling for a specific epilepsy
syndrome can be straightforward in patients with a 100% penetrant syndrome but
complicated in situations of incomplete penetrance and de novo mutations (eg,
Dravet syndrome) and should be performed with the help of a genetic counselor.
COMORBIDITIES. Epilepsy presents with a bidirectional relationship with other
neurologic (eg, stroke, migraine, dementia, traumatic brain injury) and
psychiatric (eg, depression and anxiety) comorbidities.34 Several diseases,
including depression, anxiety, dementia, migraine, heart disease, peptic ulcers,
and arthritis, are up to 8 times more common in people with epilepsy than in the
general population.35 The prevalence of comorbidities persists even in patients in
seizure remission, and patients with inactive epilepsy remain at risk of
premature mortality, suggesting a systemic component involved in the etiology
of epilepsy. This component may be mediated by basic pathophysiologic
processes and/or genetic factors.36 To have a comprehensive understanding of
the etiology of epilepsy, these comorbidities and systemic vulnerability should be
taken into account and adequately treated.
CONCLUSION
Evaluating the etiology of seizures has noticeably changed over the past decade.
There is a greater emphasis to distinguish acute symptomatic and provoked
seizures (requiring treatment of the underlying condition) from an unprovoked
seizure, which may already represent epilepsy. There is growing acceptance and
demystification of the term epilepsy as the most common cause of recurrent
seizures. The new classification of epilepsy does not stop with the recognition of
particular epilepsy syndromes but aims to recognize the underlying etiology.
This can lead to earlier recognition of surgical candidates, a better understanding
of many of the epileptic encephalopathies, and with time, hopefully medical
treatments aimed at the underlying mechanism causing the disease.37
USEFUL WEBSITES
INTERNATIONAL LEAGUE AGAINST EPILEPSY GENE REVIEW
DIAGNOSTIC MANUAL
The website provides an excellent diagnostic Peer-reviewed, regularly updated chapters
manual for epilepsy diagnosis in clinical practice. providing clinically relevant information on inherited
Epilepsydiagnosis.org conditions, covering diagnosis, management, and
genetic counseling.
www.ncbi.nlm.nih.gov/books/NBK1116/
REFERENCES
1 Scheffer IE, Berkovic S, Capovilla G, et al. ILAE 2 Gavvala JR, Schuele SU. New-onset seizure in
classification of the epilepsies: Position paper of adults and adolescents: a review. JAMA 2016;
the ILAE Commission for Classification and 316(24):2657–2668. doi:10.1001/jama.2016.18625.
Terminology. Epilepsia 2017;58(4):512–521.
doi:10.1111/epi.13709.
340 APRIL 2019

3 Hesdorffer DC, Benn EK, Cascino GD, Hauser WA. 16 Berg AT, Berkovic SF, Brodie MJ, et al. Revised
Is a first acute symptomatic seizure epilepsy? terminology and concepts for organization of
Mortality and risk for recurrent seizure. Epilepsia seizures and epilepsies: report of the ILAE
2009;50(5):1102–1108. doi:10.1111/j.1528-1167. Commission on Classification and Terminology,
2008.01945.x. 2005–2009. Epilepsia 2010;51(4):676–685.
doi:10.1111/j.1528-1167.2010.02522.x.
4 Fisher RS, Acevedo C, Arzimanoglou A, et al. ILAE
official report: a practical clinical definition of 17 Papetti L, Parisi P, Leuzzi V, et al. Metabolic
epilepsy. Epilepsia 2014;55(4):475–482. doi:10.1111/ epilepsy: an update. Brain Dev 2013;35(9):
epi.12550. 827–841. doi:10.1016/j.braindev.2012.11.010.
5 Krumholz A, Wiebe S, Gronseth GS, et al. 18 Sharma S, Prasad AN. Inborn errors of metabolism
Evidence-based guideline: management of an and epilepsy: current understanding, diagnosis,
unprovoked first seizure in adults: report of the and treatment approaches. Int J Mol Sci 2017;
Guideline Development Subcommittee of the 18(7):pii:E1384. doi:10.3390/ijms18071384.
American Academy of Neurology and the
19 Ramantani G, Holthausen H. Epilepsy after
American Epilepsy Society. Neurology 2015;
cerebral infection: review of the literature and
84(16):1705–1713. doi:10.1212/WNL.
the potential for surgery. Epileptic Disord 2017;
0000000000001487.
19(2):117–136. doi:10.1684/epd.2017.0916.
6 Jetté N, Wiebe S. Initial evaluation of the patient
20 Misra UK, Tan CT, Kalita J. Viral encephalitis and
with suspected epilepsy. Neurol Clin 2016;34(2):
epilepsy. Epilepsia 2008;49(suppl 6):13–18.
339–350. doi:10.1016/j.ncl.2015.11.008.
doi:10.1111/j.1528-1167.2008.01751.x.
7 Ho K, Lawn N, Bynevelt M, et al. Neuroimaging of
21 Pastula DM, Yeargin-Allsopp M, Kobau R.
first-ever seizure: contribution of MRI if CT is
Enhanced epilepsy surveillance and awareness
normal. Neurol Clin Pract 2013;3(5):398–403.
in the age of Zika. JAMA Neurol 2017;74(6):
doi:10.1212/CPJ.0b013e3182a78f25.
631–632. doi:10.1001/jamaneurol.2017.0215.
8 Von Oertzen J, Urbach H, Jungbluth S, et al.
22 Brenner T, Sills GJ, Hart Y, et al. Prevalence of
Standard magnetic resonance imaging is
neurologic autoantibodies in cohorts of patients
inadequate for patients with refractory focal
with new and established epilepsy. Epilepsia
epilepsy. J Neurol Neurosurg Psychiatry 2002;
2013;54(6):1028–1035. doi:10.1111/epi.12127.
73(6):643–647.
23 VanHaerents S, Gerard EE. Epilepsy Emergencies:
9 Bernasconi A, Bernasconi N, Bernhardt BC,
status epilepticus, acute repetitive seizures, and
Schrader D. Advances in MRI for ‘cryptogenic’
autoimmune encephalitis. Continuum 2019;
epilepsies. Nat Rev Neurol 2011;7(2):99–108.
25(2, Epilepsy):454–476. doi:10.1212/
doi:10.1038/nrneurol.2010.199.
CON.0000000000000716.
10 Blume WT, Lüders HO, Mizrahi E, et al. Glossary
24 Dubey D, Alqallaf A, Hays R, et al. Neurological
of descriptive terminology for ictal semiology:
autoantibody prevalence in epilepsy of unknown
report of the ILAE task force on classification
etiology. JAMA Neurol 2017;74(4):397–402.
and terminology. Epilepsia 2001;42(9):1212–1218.
doi:10.1001/jamaneurol.2016.5429.
doi:10.1046/j.1528-1157.2001.22001.x.
25 Dubey D, Singh J, Britton JW, et al. Predictive
11 Kellinghaus C, Loddenkemper T, Najm IM, et al.
models in the diagnosis and treatment of
Specific epileptic syndromes are rare even in
autoimmune epilepsy. Epilepsia 2017;58(7):
tertiary epilepsy centers: a patient-oriented
1181–1189. doi:10.1111/epi.13797.
approach to epilepsy classification. Epilepsia
2004;45(3):268–275. doi:10.1111/j.0013-9580. 26 Myer CT, Mefford HC. Advancing epilepsy
2004.36703.x. genetics in the genomic era. Genome Med 2015;
7:91. doi:10.1186/s13073-015-0214-7.
12 Fisher RS, Cross JH, French JA, et al.
Operational classification of seizure types by the 27 Scheffer IE. Epilepsy genetic revolutionizes
International League Against Epilepsy: position clinical practice. Neuropediatrics 2014;45(2):
paper of the ILAE Commission for Classification 70–74. doi:10.1055/s-0034-1371508.
and Terminology. Epilepsia 2017;58(4):522–530.
28 Myers KA, Johnstone DL, Dyment DA. Epilepsy
doi:10.1111/epi.13670.
genetics: current knowledge, applications, and
13 Fisher RS, Cross JH, D'Souza C, et al. Instruction future directions. Clin Genet 2018. doi:10.1111/
manual for the ILAE 2017 operational classification cge.13414.
of seizure types. Epilepsia 2017;58(4):531–542.
29 Koch H, Weber YG. The glucose transporter type
doi:10.1111/epi.13671.
1 (Glut1) syndromes. Epilepsy Behav 2018;pii:
14 Faulkner HJ, Arima H, Mohamed A. The utility of S1525–5050(18)30348–2. doi:10.1016/j.yebeh.
prolonged outpatient ambulatory EEG. Seizure 2018.06.010.
2012;21(7):491–495. doi:10.1016/j.seizure.2012.
30 Steele HE, Chinnery PF. Mitochondrial causes of
04.015.
epilepsy: evaluation, diagnosis, and treatment.
15 Schuele SU, Lüders HO. Intractable epilepsy: Semin Neurol 2015;35(3):300–309. doi:10.1055/
management and therapeutic alternatives. s-0035-1552624.
Lancet Neurol 2008;7(6):514–524. doi:10.1016/
S1474-4422(08)70108-X.

31 Steel D, Symonds JD, Zuberi SM, et al. Dravet 35 Yuen AWC, Keezer MR, Sander JW. Epilepsy is a
syndrome and its mimics: beyond SCN1A. Epilepsia neurological and a systemic disorder. Epilepsy
2017;58(11):1807–1816. doi:10.1111/epi.13889. Behav 2018;78:57–61. doi:10.1016/j.yebeh.
2017.10.010.
32 Crino PB. The mTOR signalling cascade: paving
new roads to cure neurological disease. Nat Rev 36 Keezer MR, Sisodiya SM, Sander JW. Comorbidities
Neurol 2016;12(7):379–392. doi:10.1038/nrneurol. of epilepsy: current concepts and future
2016.81. perspectives. Lancet Neurol 2016;15(1):106–115.
doi:10.1016/S1474-4422(15)00225-2.
33 Winawer MR, Shinnar S. Genetic epidemiology of
epilepsy or what do we tell families? Epilepsia 37 Scheffer IE. A new classification and class
2005;46(suppl 10):24–30. doi:10.1111/j.1528-1167. 1 evidence transform clinical practice in epilepsy.
2005.00354.x. Lancet Neurol 2018;17(1):7–8. doi:10.1016/S1474-
4422(17)30432-5.
34 Kanner AM. Management of psychiatric and
neurological comorbidities in epilepsy. Nat Rev
Neurol 2016;12(2):106–116. doi:10.1038/nrneurol.
2015.243.
342 APRIL 2019

Optimizing Management REVIEW ARTICLE

of Medically Responsive C O N T I N U UM A U D I O
I NT E R V I E W A V A I L AB L E
ONLINE
Epilepsy
By Derek Bauer, MD; Mark Quigg, MD, MSc, FANA, FAES
ABSTRACT
PURPOSE OF REVIEW: This article reviews the management of patients with
medically responsive epilepsy, including discussion of factors that may
lead to transient breakthrough seizures and patient and physician
strategies to maintain freedom from seizures.
RECENT FINDINGS:Imperfect adherence, unanticipated changes in ongoing

medical therapy, inadvertent use of proconvulsants or concurrent CITE AS:
medications that alter epilepsy medication kinetics, and a variety of CONTINUUM (MINNEAP MINN)
2019;25(2, EPILEPSY):343–361.
seizure precipitants such as stress or sleep deprivation may alter long-term
seizure control. Address correspondence to
Dr Mark Quigg, Department of
Neurology, University of Virginia,
SUMMARY: The majority of patients with epilepsy are medically responsive.
PO Box 800394, Charlottesville,
Many potential factors may lead to breakthrough seizures in these VA 22908, quigg@virginia.edu.
patients. Identification of these factors, patient education, and use of
self-management techniques including mindfulness therapy and Dr Bauer is a site subinvestigator
cognitive-behavioral therapy may play a role in protecting patients with for GWEP-1521, a double-blind,
epilepsy against breakthrough seizures. randomized, placebo-
controlled study to investigate
the efficacy and safety of
cannabidiol as add-on therapy
in patients with tuberous
INTRODUCTION sclerosis complex who
M
experience inadequately
uch of the neurologist’s attention in the evaluation and care of controlled seizures. Dr Quigg
the patient with epilepsy is devoted to the goal of seizure has received research/grant
remission. Continuing seizures confer higher risks of health support as principal investigator
of studies from the National
care use, underemployment or unemployment, or sudden Institutes of Health/National
death. Once the hard work of seizure remission has succeeded, Institute of Neurological
however, less attention is sometimes paid to staying seizure free. This article Disorders and Stroke, the
University of Virginia Brain
reviews the maintenance of patients with medically responsive epilepsy, Institute, and ZETO Inc. Dr Quigg
including discussion of factors that may lead to transient breakthrough seizures has received publishing
royalties from Elsevier and has
and patient and physician strategies to maintain freedom from seizures. given expert medical testimony.
UNLABELED USE OF
DEFINITIONS AND INCIDENCE PRODUCTS/INVESTIGATIONAL
USE DISCLOSURE:
Of the 50 million people worldwide and the 3.4 million people in the Drs Bauer and Quigg report no
United States with epilepsy (roughly 1.2% of the population), two-thirds will disclosures.
become seizure free with appropriate pharmacotherapy.1 Many of these patients
will remain reliably seizure free, but some will have their daily routine © 2019 American Academy
interrupted by breakthrough seizures. of Neurology.

MEDICALLY RESPONSIVE EPILEPSY
KEY POINT Breakthrough seizures—the primary topic of this article—must be

distinguished from two other possibly confusing entities. First, reflex epilepsies
● Approximately two-thirds
of patients with epilepsy will
are those in which seizures are provoked objectively and consistently by specific
become seizure free with stimuli or activities in patients who otherwise lack spontaneous seizures.2 For
pharmacotherapy. example, reading epilepsy does not fall into the scope of this review. Second,
acute symptomatic seizures occur in patients with or without epilepsy and are
provoked at the time of an acute, and usually severe, epileptic insult.3 Examples
of acute symptomatic seizures include those provoked by proconvulsant
medications, illicit drug or alcohol withdrawal, traumatic head injury, or renal
dialysis. Although the measures reviewed in this article may help susceptible
individuals, the focus of this article is on maintaining seizure freedom (ie,
avoidance of the breakthrough seizure) in a patient with known epilepsy.
A breakthrough seizure is defined as one that occurs despite the use of
antiseizure medications that have otherwise successfully prevented seizures in
the patient in the past. The duration of the seizure remission period (the
minimum interseizure interval) needed to be considered seizure free varies with
inherent seizure frequency. The International League Against Epilepsy (ILAE)
defines seizure free or medically responsive epilepsy as seizure freedom for
12 months or 3 times the longest previous interseizure interval, whichever is
longer.4 This definition is handily referred to as “the rule of three.”
Other studies define the durations of seizure remission differently. Later
analyses with the use of Bayesian techniques suggests the “rule of three to six,”
whereby those with a high preintervention probability of seizure freedom should
be without seizures for 3 times the previous interseizure interval prior to being
considered seizure free, while those that have low preintervention probability of
seizure freedom should wait up to 6 times the previous interseizure interval.5
Clinical research has varied from the ILAE definition of seizure freedom. For
example, in a study of patients with seizures presenting to the emergency
department, Divino and colleagues6 defined breakthrough seizures as those that
occurred more than 6 months from the previous seizure that were not clearly
associated with antiseizure medication nonadherence. Alternatively, in a
secondary analysis of a large comparison of new versus standard antiseizure
medications, Bonnett and colleagues7 required at least 12 months of seizure
freedom for a recurrent seizure to be considered breakthrough. For the purposes
of this review, seizure freedom follows the standard ILAE definition.
The estimated prevalence or incidence of breakthrough seizures varies widely
by definition and sample. A Ugandan study estimated prevalence in a sample of
267 patients, of which 74% experienced breakthrough seizures; interseizure
interval was not defined.8 In an English sample of more than 2000 patients in a
prospective, observational trial of standard versus newer antiepileptic drugs
(the SANAD [Standard and New Antiepileptic Drugs] trial9), the investigators
defined breakthrough seizures as those that recurred after at least 12 months of
remission regardless of previous seizure frequency.7 The incidence of
breakthrough seizures was 37%.
CONSEQUENCES AND COSTS

As CASE 3-1 illustrates, breakthrough seizures can result in interruptions of
schedules, potential injuries, or loss of work or driving privileges. Breakthrough
seizures may place patients at worse risk because circumstances and activities are
less restrained than those with continuing seizures. Seizures may result in motor
344 APRIL 2019

A 19-year-old woman presented to the emergency department for a CASE 3-1
generalized tonic-clonic seizure witnessed by her college roommate. She
had a history of juvenile myoclonic epilepsy and had been seizure free on
lamotrigine 100 mg 2 times a day from her junior year in high school (near
the time of onset) to her second year in college (the time of her
presentation to the emergency department).
She admitted to occasionally missing her lamotrigine morning dose,
but she wasn’t sure whether she had missed a dose on the day of this
seizure.
Her medication had been recently refilled, and the generic pill she
was accustomed to had changed manufacturers (and thus possible
pharmacokinetic properties). She had also recently started oral
contraceptives. Upon further questioning, she noted that she had just
finished a term paper that had required relatively sleepless nights.
Because she presented with a head contusion from her seizure and was
initially sleepy and disoriented after her seizure, she underwent a head
CT, which was normal. Further laboratory testing in the emergency
department showed that her lamotrigine level was 2 mcg/mL (lower limit
of reference range >2.5 mcg/mL). During the neurology consultation in
the emergency department, her gynecologist was contacted, and it was
determined that she had been placed on a progesterone-only oral
contraceptive, which was thought unlikely to be a contributing factor to
low lamotrigine levels.
Given the patient’s self-reported issues with adherence, her pharmacy
was contacted. The patient filled her monthly lamotrigine prescription 9
times out of the prescribed 12 months. The medication possession ratio of
0.75 would suggest that she is missing the medication at least 25%
of the time. Based upon this evidence, it was determined her seizure was
most likely due to nonadherence. To address this, her lamotrigine was
changed from 100 mg 2 times a day to 200 mg 1 time a day of the extended
release formulation. In addition, she was counseled to maintain a regular
sleep schedule.
Following the emergency department visit, she missed a day of classes
and was prohibited from driving for 6 months.
This case illustrates an example of a breakthrough seizure. Breakthrough COMMENT

seizures interrupt a period of apparent seizure remission in a patient with
known epilepsy. The costs in terms of injury, time, health care utilization,
and employment can be considerable.
This case also displays the many factors that need to be considered with
breakthrough seizures. Medication interactions, adherence, as well as the
presence of possible seizure precipitants are all impediments to sustained
seizure freedom. In this patient’s case, the significance of her intermittent
nonadherence was only made clear through discussion with her pharmacy.
With adjustment of her antiseizure medication regimen to once-daily dosing,
the patient was able to more reliably take her medication and once again
achieved long-term seizure freedom.

vehicle injuries or other trauma. Breakthrough seizures may present as status

epilepticus,10 and higher rates of seizures and antiseizure medication
nonadherence are associated with sudden unexpected death in epilepsy
(SUDEP).11 Economic costs follow trauma and morbidity. In a case-control study
of those with breakthrough seizures, cases had significantly higher rates of
hospitalization and emergency department visits compared with controls and
higher total all-cause direct health care costs (median $12,714 versus $5095). All
told, patients with breakthrough seizures had 2.3 times higher adjusted health
care costs and 8.1 times higher adjusted epilepsy-related costs than controls.6
In the longitudinal management of patients with medically responsive
epilepsy, the primary goal should be to maintain seizure freedom without side
effects from antiseizure medication. The neurologist should be aware of the
significant nonadherence issues inherent with chronic medication management,
be vigilant for potential medication factors or interactions that can affect stable
antiseizure medication kinetics, and identify common seizure precipitants.
Management of these issues requires both physician expertise and buy-in from
the patient.
An increasingly expanding body of literature exists regarding patient
participation and empowerment in the long-term care of epilepsy, known as
self-management, which is defined as “activities or steps that an individual or
family can perform that are known to either control the frequency of seizures or
promote the well-being of the person with seizures.”12 In other words, the
maintenance of seizure freedom is best achieved when the physician and patient
are reading from the same playbook.
RISK FACTORS FOR BREAKTHROUGH SEIZURES

Risks for breakthrough seizures have not been uniformly studied. In multivariate
analyses of patients in the SANAD trial, three factors were found to be robust
predictors of seizure recurrence: a history of neurologic insult defined as learning
disability or a focal neurologic deficit resulting in functional impairment, the
total number of tonic-clonic seizures recorded before achieving 12-month
remission, and time taken to achieve 12-month remission.7
Although valuable for general prognosis of the potential and incidence of
breakthrough seizures, the SANAD study, however, did not evaluate variables
that could be modified by the physician or patient. Only two studies consisting of
surveys of patients with breakthrough seizures have enumerated these factors
in comparative fashion.8,13 Both studies found that antiseizure medication
adherence was the leading factor associated with breakthrough seizures (27% to
56%). Other factors fell into the broad category of seizure precipitants and will
be discussed in the appropriate section below.
ADHERENCE
Treatment adherence is an important factor in maintenance of seizure control,
and a gentle skepticism while taking a history of adherence helps both the
physician and patient.
Prevalence and Measurements of Antiseizure Medication Adherence

Nonadherence is not confined to epilepsy. The World Health Organization
estimates that medication nonadherence across all diseases is 50% worldwide.14
Patients in US clinical drug trials for a variety of diseases demonstrate adherence
346 APRIL 2019

at rates ranging from 43% to 78%.15 Despite being an endemic issue, the KEY POINTS
characterization of medication adherence, including in patients with epilepsy, is
● Comparatively little
limited by lack of standardization in definitions and measurements.15,16 For evidence exists for
example, adherence may be characterized by dose adherence (ie, taking the evaluating risk factors for
proper number of pills of a medication in a given time interval) or by interval breakthrough seizures. The
adherence (ie, taking medication in the prescribed intervals between doses). The available data suggest that
nonadherence is the most
most tangible example of dose adherence is pill count, which provides an
important factor leading to
objective measure of the number of pills taken over a period of time. Interval breakthrough seizures.
adherence may be assessed through patient diary or electronic monitoring
technology. A practical adherence measurement in this area is the medication ● Antiseizure medication
possession ratio, the ratio of total days with available medication to total days nonadherence is common
and should be screened for
prescribed. A medication possession ratio of less than 0.8 is generally considered when a patient’s seizures are
nonadherent.16 This threshold translates to missing 3 doses a week in a uncontrolled.
twice-daily dosing regimen, which is an eye-opening rate to most physicians.
Given this threshold, the proportion of patients who miss 3 or more doses per ● Increasing complexity of
the antiseizure medication
week ranges between 26% and 79%, with a suggested overall rate of 40%.16 regimen, including
Patients who are nonadherent may change patterns of nonadherence over increased dosing frequency
time. Modi and colleagues17 found that rather than fitting into a binary adherent or increased numbers of
or nonadherent paradigm, patients fell into four different patterns of adherence medications, is associated
with an increased risk of
during long-term antiseizure medication therapy: high adherence, moderate
medication nonadherence.
adherence, severe early nonadherence, and variable nonadherence. In that study,
15% of patients displayed shifts among groups over the course of follow-up.
Another clinically important example of variable adherence is so-called “white
coat compliance,” which is marked by improving patient adherence in the days
around a physician office visit. This has been found in both pediatric and adult
epilepsy populations, with a reported increase in adherence around the time of
office visits ranging from 5% to 15%.18,19
Risk Factors for Antiseizure Medication Nonadherence

Frequent daily dosing is the traditional explanation for medication nonadherence.
With the use of pill bottles fitted with an electronic dose counter, Cramer
and colleagues20 found that antiseizure medication adherence dropped with
frequency: 87% adherence with 1 time a day, 81% with 2 times a day, 77% with
3 times a day, and 39% with 4 times a day. This study also compared other
measures of nonadherence and found that neither drug levels nor pill counts
correlated accurately with the electronic dose counter, finding that the first
two overestimated adherence. In line with this reasoning were later studies
using pill count techniques that failed to confirm frequency-adherence
relationships.21
Antiseizure medication polytherapy may also be an important barrier to
adherence.13,21–23 Ferrari and colleagues23 found nonadherence rates of 55% for
those receiving monotherapy compared to 71% for those receiving polytherapy.
Demographics such as age and sex have no consistent effects on adherence.13,21–23
In the authors’ experience in a tertiary epilepsy clinic, the span between late
adolescence and early adulthood appears to be an especially challenging time for
adherence as patients begin to act independently. The few studies that have
specifically examined this age group tend to agree, with adolescents or young
adults demonstrating worse adherence than younger children.8,24,25
Socioeconomic factors play a role in antiseizure medication adherence. For
example, in a study centered specifically on adherence in pediatric patients with

epilepsy, lower socioeconomic status was the sole factor associated with
nonadherence.17 At the other end of the age spectrum, adherence in elderly
adults measured from prescription refill counts (proportion of days covered, a
measurement related to medication possession ratio) in Medicare Part D found
that zip codes in high poverty areas were more likely to be nonadherent than
those from zip codes in wealthier areas.26
Race has also been shown to correlate with antiseizure medication
nonadherence, with minority patients, especially those who identified as African
American, at higher risk of nonadherence. In an indigent care clinic (a sample
considered as a leveling factor for wealth), compared with white patients,
African American patients had lower adherence equivalent to 2 days of missed
antiseizure medications per month in a twice daily regimen.27 In the
population-based study by Piper and colleagues,26 adherence was worse among
African American patients (40%) compared with white patients (29%).
Lastly, psychiatric factors such as mood, anxiety, or cognition may play roles
in adherence. A 2017 report from Wang and colleagues28 suggested that
moderate to severe anxiety was associated with nearly a threefold risk of
nonadherence while social support may offer some degree of protection.
Similarly, Ettinger and colleagues29 found that depression was associated with
medication nonadherence as well as poorer quality of life, although this was not
consistent across all metrics studied.
Improving Antiseizure Medication Adherence

Methods to improve antiseizure medication adherence have not been
systemically compared, but many commonly used techniques are available to the
experienced physician (TABLE 3-1). Traditionally, clinicians have depended on
what some could dub the “doomsayer approach,” pulling out verbal tarot cards
predicting, for the nonadherent, woes ranging from possible antiseizure
medication adverse effects, trauma, loss of employment, loss of driving
privileges, and the ultimate trump card, SUDEP. Although the last gets the
attention of parents and loved ones, mortality can be an abstract concept to
younger people.
Reminders may aid in adherence; those patients with smartphones enjoy a
host of medication reminder–specific apps as well as a sophisticated calendar
system that can be programmed for regular medication reminders. Less
TABLE 3-1 Methods to Aid in Antiseizure Medication Adherence
Type of Method Intervention

Reminders Smartphone alarm, smartphone app, pill box, premeasured pill packets
Supervision Recruitment of family or friends, home health care, pill dispensers/

compliance auditors, school administration
Medication Change from antiseizure medication with short half-life to one with
long half-life, fill every 3 months instead of every month, alter
scheduling to coincide with other cues
Programmatic Behavioral counseling, educational counseling

intervention
348 APRIL 2019

technical, common techniques include explicit medication lists, pill boxes, and KEY POINTS
for those patients or families who are less literate, prepackaged “blister packs”
● Many demographic
can be available that contain fixed and clearly labeled antiseizure factors may convey an
medication doses. increased risk of
Antiseizure medication supervision can aid patients who are nonadherent. nonadherence, including
Recruiting family members or friends is the most common approach, but some socioeconomic status and
race, although other factors
patients may require, depending on availability, scheduled compliance checks by
such as gender and age are
home health care providers. Pediatric patients whose parents may have not clearly associated with
cognitive, social, or schedule problems can be helped by altering antiseizure increased risk.
medication scheduling to guarantee that at least a morning or afternoon dose is
dispensed at school rather than at home. ● Current data suggest that
adherence may be improved
The wily physician can sometimes aid adherence by altering antiseizure by increasing patient
medication scheduling, eg, advising that doses can be taken with meals or with education and providing
other cues rather than adhering to more “perfect” on-the-clock equal intervals. feedback to address
Formulations can be manipulated; patients with trouble remembering their twice concerns that may be
obstacles to adherence.
daily antiseizure medication may improve with an extended release once daily
formulation. Families who may have problems with monthly paychecks not
extending to the end of the month or who have travel problems sometimes
can benefit from antiseizure medications dispensed in 3-month rather than
1-month refills.
Finally, patients with persistent nonadherence may benefit from a more
systematic approach derived from cognitive-behavioral techniques.30 A few
recent randomized controlled trials with small populations have evaluated
various self-management interventions to improve adherence. Dash and
colleagues31 investigated the effect of educational intervention upon patient
adherence. This study compared 90 participants in the active intervention group
who received guided content including basic information about epilepsy and
antiseizure medications, rationales for treatment, and techniques for coping with
epilepsy, which was randomized against control participants who received
standard care. Participants in the intervention arm demonstrated a statistically
significant improvement in adherence questionnaire scores as well as decreased
seizure frequency.
Pakpour and colleagues32 compared 137 participants who were randomly
assigned to behavioral therapy that targeted antiseizure medication adherence
via counseling sessions that occurred 3 times a week that reinforced the need for
adherence, explored patient concerns that may cause nonadherence, and
reviewed methods to overcome these concerns against 138 control participants
and found that 60% of the participants in the intervention group compared with
38% of controls had antiseizure medication levels within reference range
6 months postintervention.
Tang and colleagues33 evaluated two types of intervention: 59 participants
were randomly assigned to receive medication education alone (that included
monthly calls from a pharmacist) versus 65 participants who received a
combination of this educational intervention plus behavioral intervention that
targeted systematic medication scheduling. Although there was no significant
difference between the two groups in follow-up analysis, both groups
experienced improved adherence and reduction in seizure frequency.
In summary, current data suggest that adherence may be improved by
increasing patient education and providing feedback to address concerns that
may be obstacles to adherence. Overall, relatively few randomized controlled

studies have evaluated methods to improve adherence in patients with epilepsy,

which was recently highlighted in a systematic review by da Mota Gomes and
colleagues.30 These studies demonstrate that short-term antiseizure medication
adherence can improve with active intervention.
Antiseizure Medication Monitoring

Expert guidelines suggest a limited but important role for antiseizure
medication monitoring via serum concentrations.34 Once a treatment regimen
is deemed effective, a trough antiseizure medication concentration can
establish a proper baseline by which to judge future problems such as toxicities
or changes in effectiveness, such as breakthrough seizures or suspected
nonadherence. Antiseizure medication levels can be obtained either in
anticipation of or in the face of changes in health or development that may
perturb steady-state levels, such as in children when they are rapidly growing,
changes in clearance as the result of aging or hepatic-renal disease, or
pregnancy. Drug levels may be helpful when encountering possible formulation
variability or pharmacokinetic interactions when coinciding medications change.
However, regular, routine blood level monitoring of antiseizure medications
is generally not needed and can even be harmful if an effective regimen is
altered because the result falls outside the “normal” range. Note that nearly every
indication for monitoring provided in expert guidelines depends on changes
of levels within individuals, not on comparison of an individual to grouped
ranges. The interpretation of normal should be done cautiously because there
can be considerable variability of reference ranges across different laboratories.35
Most importantly, what is normal for one patient may be toxic or ineffective
for another. In short, serum concentrations are only part of the story; dose
adjustments should not be made on the basis of levels alone but should be
performed when guided by the patient’s clinical state.
Antiseizure Medication Instability, Interactions, and Proconvulsants

Some patients maintain rigorous adherence but experience breakthrough
seizures arising from unanticipated changes in previously stable and effective
antiseizure medication regimens.
MEDICATION OR OTHER INTERACTIONS THAT MAY LOWER ANTISEIZURE MEDICATION

LEVELS. Antiseizure medication levels may drop unexpectedly in a previously
stable antiseizure medication regimen. Interactions between antiseizure
medications, between other medications and antiseizure medications, or from
dietary intake may affect antiseizure medication levels.
While a comprehensive listing of all interactions that result in the lowering
of antiseizure medication levels is too extensive for this review, examples of
significant interactions are discussed. Common in the authors’ clinical
experience is the interaction considered in the patient in CASE 3-1: the induced
metabolism of lamotrigine by estrogen. Estrogen-containing compounds (as well
as the menstrual elevation of estrogen in the menstrual phase of the menstrual
cycle) act as inducers of the uridine 5'-diphospho-glucuronosyltransferase
enzyme, which, in turn, metabolizes lamotrigine, resulting in an approximately
50% decrease in lamotrigine levels.36,37 Progesterone, on the other hand, has
little effect on lamotrigine.32 Although interactions with enzyme-inducing
antiseizure medications such as phenytoin and carbamazepine also feature
350 APRIL 2019

similar interactions with lamotrigine, since antiseizure medications are typically KEY POINT
prescribed by neurologists and estrogens are typically prescribed by non-
● Medications and
neurologists, dips in effective lamotrigine levels are more apt to arise unexpectedly supplements may cause
because of the introduction of drugs that are not antiseizure medications. seizures, either through
A similar underappreciated but apparent interaction in the authors’ interactions with antiseizure
clinical practice is between phenytoin and tobacco. Tobacco is an established medication or through
proconvulsant properties.
inducer of multiple cytochrome P450 isoenzymes.38 The clinical significance of
this is poorly documented, but there is at least one case report of phenytoin
toxicity reported with smoking cessation.39 In the authors’ experience, tobacco
smoking may result in lowered phenytoin levels that persist despite attempts at
increasing doses, leading to the perception of nonadherence.
Medication interactions are not limited to prescribed medications; over-the-
counter medications or herbal preparations should also be routinely considered
in patients with medication-responsive epilepsy. More than 50% of patients
with epilepsy use herbs and dietary supplements, and 29% of those do not report
these exposures to their physicians.40 TABLE 3-2 contains a listing of commonly
used herbal supplements that either decrease antiseizure medication levels or
decrease seizure threshold.41,42
An example of an herb-drug interaction in epilepsy is ginkgo biloba, which is a
known hepatic enzyme inducer and has been previously shown to decrease levels
of phenytoin and valproate, which, in turn, has been documented to lead to
breakthrough seizures.41 The repercussions from such interactions can be
substantial and even deadly as at least one case report exists of seizure-related
death (a drowning) occurring because of the antiseizure medication–lowering
effects of gingko use.43
PROCONVULSANTS. As noted above, many common supplements

( TABLE 3-2) and prescription medications (TABLE 3-3)44 may lower seizure
threshold. In the authors’ experience, the ubiquitous beta-lactam antibiotics
(eg, ampicillin), which are frequently used in children who are febrile, ill, or
sleep deprived, can, in the setting of epilepsy, lower seizure threshold enough
to facilitate subsequent seizures. Certain psychotropic medications are also
associated with an increased risk of breakthrough seizures. For example,
bupropion is a commonly prescribed medication for either depression or
smoking cessation. Bupropion is thought to confer a dose-dependent lowering
of seizure threshold or emergence of epileptiform EEG abnormalities.45,46 A
limitation of this literature is that reports have focused on the emergence of
seizures or EEG findings in the general population; studies have not focused on
the specific risk to patients with epilepsy.
SWITCHING ANTIEPILEPTIC DRUGS. Breakthrough seizures may be precipitated

by well-intended iatrogenesis. Changing of antiseizure medications may occur
in a medically responsive patient when faced with a host of antiseizure
medication–related problems. For example, the child with unexpectedly poor
school performance or the adult with severe osteoporosis may require antiseizure
medication switching; or an anticipated medical condition such as planned
pregnancy may require replacement of teratogenic valproate with an antiseizure
medication with less fetal adverse effects, and an upcoming organ transplant
may require replacement of an antiseizure medication with one without
interactions with immunosuppressant drugs.

Switching antiseizure medications in patients with medically responsive

epilepsy is not without risk. A 2016 prospective study of patients with epilepsy
undergoing an antiseizure medication switch versus controls who were
maintained on current therapy suggested an approximate 15% increased risk of
breakthrough seizure activity over the next 6 months.47
Sometimes the prescribing hand of the physician is not forced by medical
concerns but by patients’ financial limitations. Much controversy surrounds
denials by third-party payers for coverage of brand name antiseizure
medications and subsequent “forced” replacement with theoretically equivalent
generic antiseizure medications. The data regarding brand name to generic
TABLE 3-2 Common Medicinal Herbs, Primary Uses, and Effects on Lowering of
Seizure Threshold or Antiseizure Medication Levelsa
Herb Typical Use Proconvulsant Effect
Asafoetida Memory, mood, sedative Lowering of seizure threshold
Borage Fever, diuretic Lowering of seizure threshold,

lowering of antiseizure medication levels
Ephedra Cold, flu, respiratory Lowering of seizure threshold
Ergot Migraine Lowering of seizure threshold
Eucalyptus Cold, flu, respiratory Lowering of seizure threshold
Evening Weight loss, menstrual Lowering of seizure threshold,

primrose pain lowering of antiseizure medication effects
Ginkgo biloba Memory, mood Lowering of seizure threshold,

lowering of antiseizure medication levels
Ginseng Mood, erectile Lowering of seizure threshold

dysfunction
Pennyroyal Abortion, menses Lowering of seizure threshold

induction
Sage Hyperlipidemia, Lowering of seizure threshold

gastrointestinal
symptoms
Shankhpushpi Memory Lowering of antiseizure medication levels

(Ayurveda)
Star anise Colic, gastrointestinal Lowering of seizure threshold

symptoms
Star fruit Diuretic Lowering of seizure threshold
Wormwood Gastrointestinal Lowering of seizure threshold

symptoms
Yohimbine Erectile dysfunction Lowering of seizure threshold, lowering of

antiseizure medication levels
a
Data from Samuels N, et al, Epilepsia,41 and Tyagi A, et al, Epilepsia.42
352 APRIL 2019

interchange are mixed, although the data increasingly support efficacy of
generics, as recent large-scale prospective and retrospective investigations
support bioequivalence between brand name antiseizure medications and
generic substitutions approved by the US Food and Drug Administration (FDA),
with overall no statistical difference in seizure frequency.48,49 However, older
single-site studies have suggested that brand name to generic antiseizure
medication exchanges are poorly tolerated.50 Current data also suggest that
switching between different generic formulations of the same antiseizure
medication is not associated with increased seizures.51 Physicians should be
vigilant about patient adherence during the initial interchange between brand
and generic formulations, as evidence suggests that changes in pill appearance
may increase the risk of nonadherence.52 These data led the American Epilepsy
Society to recommend routine counseling about possible changes in color and
shape in generic antiseizure medication substitutions in its 2016 position
statement on the topic of generic pharmacotherapy.53
Seizure Precipitants
Seizure precipitants are “any endogenous or exogenous factor that promotes the
occurrence of epileptic seizures.”54 Seizure precipitants are different from acute
symptomatic seizures in that healthy individuals will not seize when exposed to a
Drug-Drug Interactions That Depress Antiseizure Medication Levelsa TABLE 3-3
Medications That Decrease

Antiseizure Medication Antiseizure Medication Levels
Brivaracetam Rifampin
Clonazepam, other benzodiazepines Rifampin, enzyme-inducing antiseizure

medications
Enzyme-inducing antiseizure medications Enzyme-inducing antiseizure medications

(eg, phenytoin, carbamazepine, primidone)
Eslicarbazepine acetate, oxcarbazepine Enzyme-inducing antiseizure medications
Felbamate Enzyme-inducing antiseizure medications
Lamotrigine Rifampin, estrogen, lopinavir/ritonavir,

enzyme-inducing antiseizure medications
Primidone Diuretics
Perampanel Enzyme-inducing antiseizure medications

(not primidone)
Rufinamide Enzyme-inducing antiseizure medications
Tiagabine Enzyme-inducing antiseizure medications
Topiramate Phenytoin, carbamazepine
Valproate/valproic acid Carbapenem antibiotics
Zonisamide Enzyme-inducing antiseizure medications
a
Modified with permission from Vossler DG, et al, Epilepsy Curr.44 © 2018 American Epilepsy Society.

seizure precipitant, while patients with and without epilepsy may seize in the
face of severe brain or systemic insults, which typify acute symptomatic
seizures.3 The reported prevalence of seizure precipitants varies widely based
on practice setting. A study of a community-based practice found the prevalence
of all precipitants to be 47%, while rates range higher, between 62% and 97%,
at tertiary epilepsy centers54–56 Despite a significantly wide range of estimates
of the overall prevalence of seizure precipitants, data regarding specific
patient-reported seizure precipitants are remarkably consistent, with stress,
sleep deprivation, and fatigue being the most commonly documented
(FIGURE 3-1).54–56 Patients with different epilepsy syndromes probably have
different susceptibilities among precipitants. For example, in the study by Frucht
and colleagues,54 patients with temporal lobe epilepsy cited sleep (as opposed to
sleep deprivation) as a precipitant disproportionately less frequently than
syndromes such as nontemporal focal epilepsy or idiopathic generalized epilepsy.
STRESS AND FATIGUE. The most commonly reported seizure precipitant is

54–56
stress. The original study of precipitant prevalence by Frucht and
colleagues54 documented that 30% of adult patients surveyed cited stress as a
clear precipitant. In this study, more women (36%) than men (24%) cited stress,
and the syndrome most susceptible to stress was temporal lobe epilepsy. Fatigue
is a reported seizure precipitant in 13% to 15% of patients.54–56 Isolating fatigue
as a seizure precipitant is difficult, as evidence suggests that emotional stress,
fatigue, and sleep deprivation are correlated with one another.54 Stress, sleep
deprivation, and fatigue are likely interrelated and may share a common
pathophysiology.54 Experimental models of epilepsy support the hypothesis
that stress, especially early exposure during brain development, facilitates
epileptogenic changes in stress-dependent neurotransmitters and hormone
levels within the brain.57 Observations of mass stress in humans echo animal
findings; for example, seizure control worsened shortly after the terror attacks in
the United States on September 11, 2001.58
FIGURE 3-1
Distribution of seizure precipitants for 400 adult patients surveyed at a tertiary care
epilepsy clinic.
Reprinted with permission from Frucht MM, et al, Epilepsia.54 © 2000 John Wiley and Sons.
354 APRIL 2019

SLEEP DEPRIVATION AND INSOMNIA. Sleep deprivation is also a frequently KEY POINTS
documented seizure precipitant, estimated to occur in 18% to 25% of patients.54,56,59
● Switching antiseizure
Although sleep deprivation correlates highly with stress and fatigue, sleep medications may result in
deprivation and insomnia require separate discussions because of the history seizure recurrence, even in
of the use of sleep deprivation as an interictal spike– or seizure-activating the setting of prolonged
procedure in routine and epilepsy-monitoring EEG.60 seizure freedom.
Insomnia and sleep disturbances are a common problem in patients with
● Current data from large
epilepsy, affecting 24% to 55% of patients.61–65 Quigg and colleagues65 population studies suggest
documented that more than 50% of patients with epilepsy experienced some that generic antiepileptic
degree of insomnia, with 43% having clinically significant insomnia. Most studies drugs are bioequivalent to
correlate sleep deprivation or insomnia with worse seizure control. For example, name brand medication.
previous diary studies66 and surveys8,13,54 found that patients reported that ● Stress and sleep
sleep deprivation provoked seizures. deprivation are the most
Sleep deprivation and insomnia, through the enhancement of the stress common seizure
response, may worsen seizure control because of dysregulation of hypothalamic precipitants in patients
with epilepsy.
pituitary function. Insufficient sleep and the “hyperarousal” of insomnia causes
compensatory changes in homeostatic processes; stress hormones such as ● Insomnia is common in
noradrenaline and corticosteroids are dysregulated in primary insomnia.67 patients with epilepsy and
Altered corticosteroid responses to stress have been observed in children is correlated with poor
seizure control.
susceptible to stress-precipitated seizures.68 The interactions between sleep
disruptions and epilepsy may not be one way; seizures and the epileptic state may
alter circadian regulation and affect sleep distribution (among other
circadian rhythms).69,70
FLASHING LIGHTS. Flashing or flickering lights are reported as a precipitant in 4%

and 17% of surveyed patients.54,56 Flashing lights can be the cause of a reflex
photosensitive epilepsy, but the discussion here will center on flashing lights as
a more generic precipitant in patients with otherwise well-controlled epilepsy.
Flashing lights can cause mass seizure precipitation; in 1997, television airing of
a Pokémon cartoon that featured a sequence of flashing lights caused seizures
in a total of 685 children, among whom were children who likely had latent
epilepsy and had yet to have a seizure, and those with known epilepsy
with susceptibility to photic stimulation.71 Susceptibility to flashing lights is
not distributed equally among syndromes; patients with idiopathic generalized
epilepsies have 3 to 4 times the prevalence of abnormal responses to photic
stimulation during EEG as those with symptomatic focal epilepsies.72
EXERCISE. Exercise appears to have disparate effects on groups of patients.

Exercise in the form of “physical exertion” was reported by Frucht and
colleagues54 as a seizure precipitant in 9% of patients; a similar proportion also
reported that becoming overheated or exposed to heat and humidity was a
precipitant. Nakken and colleagues73 similarly found 10% of patients with
epilepsy reported exercise as a seizure precipitant; however, in contrast, the
study also found that 36% of patients reported improvement in seizure control
with regular exercise. A small randomized controlled study of exercise in
epilepsy found that regular physical activity not only improved seizure
frequency but also improved self-perceived “physical self-concept and vigor.”74
ALCOHOL. Alcohol use has also been described as a seizure precipitant54; higher
rates (15%) may be seen where alcohol use is more culturally accepted.75 These
studies did not pin down whether it was alcohol ingestion or the phase of

withdrawal that patients reported as the susceptibility. A review of alcohol use in

patients with epilepsy concluded that regular use in “small amounts” (one to two
drinks) neither increases seizure frequency nor affects antiseizure medication
levels.76 Greater use, however, places patients at risk of withdrawal symptoms,
and lowered seizure thresholds may occur despite the absence of symptoms of
frank delirium tremens, with the greatest risk of seizures occurring about 7 to
48 hours after the last drink.77 Furthermore, the use of alcohol may combine with
other precipitants such as nonadherence or sleep deprivation. Samsonsen and
colleagues,78 in a survey of patients hospitalized for seizures, found that alcohol
intake was associated with decreased sleep in patients with epilepsy. Alcohol
avoidance has also been reported as a self-management intervention that may
improve seizure frequency.79
MANAGEMENT OF SEIZURES DUE TO PRECIPITANTS. Breakthrough seizures due to

a precipitating factor should be considered evidence of suboptimal seizure
control and predictive of more seizures.4,7 The authors recommend the following
process in evaluating breakthrough seizures. First, acute symptomatic seizures
(the result of a new acute epileptogenic injury or progression in a patient
with a progressive epileptic disorder) should be considered. Second, either
history, detective work such as pill counts, or drug levels should establish
adherence. Third, if adherent, antiseizure medication therapies should be
optimized or changed, since breakthrough seizures can be considered as jumping
over an inadequate seizure threshold. Fourth, the physician should attempt
to identify the seizure precipitants and help strategize their avoidance. If
precipitants clearly appear to be present, avoidance strategies may be sufficient
without changes in the antiseizure medication regimen. Fifth, the clinician
should inform the patient that adequate time be designated to consider the
patient seizure free again (ie, using the rule of three to estimate how long the
patient is required to be seizure free before being considered back in “safe territory”).
Avoidance of endogenous precipitants such as stress and fatigue are nuanced
issues. Baldin and colleagues80 highlighted that the presence of anxiety and
mood disorder were associated with stress and increased risk of breakthrough
seizures. The interconnection between anxiety, depression, and stress is complex.
Privitera and colleagues81 have shown that patients with a history of depression
were more likely to have stress as a precipitating factor for seizures. Evidence
also suggests that stress and anxiety may be mediated by depression in patients
with epilepsy.82 These findings underpin the need for physician screening of
depression and mood disorders in patients with epilepsy, not only as a means
of improving mood but also to potentially optimize and maintain seizure control.
Self-management techniques have also been found to be beneficial for
stress and psychiatric comorbidities associated with epilepsy.83 In one study,
mindfulness therapy, a form of meditation, was shown to decrease seizure
frequency and reduce symptoms of anxiety and depression as compared to
control participants, who received social support only.33 In a small randomized
controlled trial, yoga was also associated with improved quality of life and seizure
frequency, although this study was limited by lack of adequate controls as both
randomly assigned groups received active interventions (yoga versus behavioral
therapy).84 Cognitive-behavioral therapy for anxiety and depression has also
been shown to improve mood and seizure frequency.85,86 A recent randomized
controlled trial investigated seizure and stress reduction through muscle
356 APRIL 2019

relaxation techniques. Patients in the treatment arm were instructed in a muscle KEY POINTS
relaxation protocol; controls were instructed in a “focused attention” protocol.
● Mood disorders are
Seizure frequency improved equally in both groups, but stress reduction common in patients with
improved significantly better in the treatment group.87 epilepsy. Screening should
Deleterious effects of sleep deprivation and insomnia on seizure control can, be performed to address
hopefully, be helped by improvements in sleep hygiene or treatment of both inherent mood
concerns but also to
insomnia, although hard data are largely lacking. The following is a short list of
potentially improve
simple sleep suggestions used in counseling patients in sleep hygiene: seizure control.
● Self-management
u Awaken at the same time every day (eg, 6:30 AM or 7:00 AM) and do something active and techniques may improve
in the light upon awakening psychiatric comorbidities
u No sleep “when the sun is up” (don’t steal from nighttime sleep) associated with epilepsy,
and some evidence suggests
u Give yourself at least an 8-hour window for sleep, and do something “boring” before going this may translate into
to bed improved seizure frequency.
u No caffeine past noon
● The first-line treatment
u No electronics in the bedroom (eg, smartphone, computer, television) for insomnia is cognitive-
behavioral therapy;
The authors of this article provide patients these suggestions because they sedative-hypnotic use
should be avoided as much
can be tackled by most patients and their families largely because of simplicity.
as possible in patients with
Note that these sleep suggestions mediate waking behavior; patients need to be epilepsy because of risks
counseled that sleep cannot be forced but slips into the space prepared for it by of polypharmacy and the
one’s daytime activities. A sleep history concentrating on time in bed, time fluctuation of seizure
thresholds, such as
arising, sleep fragmentation, in-bed and environmental distractions, caffeine
in the withdrawal from
use, and factors such as snoring and apneas can disclose sleep habits that may and habituation to
affect seizure control and daytime sleepiness. In the authors’ experience, a sleep benzodiazepines.
history is easily obtained but never given without explicit questioning. The
first-line treatment for insomnia is cognitive-behavioral therapy88; sedative-
hypnotic use should be avoided as much as possible in patients with epilepsy
because of risks of polypharmacy and the fluctuation of seizure thresholds,
such as in the withdrawal from and habituation to benzodiazepines. The
evidence for the proconvulsant properties of medications used for their soporific
effects due to activities at histamine receptors—tricyclic antidepressants, such
as amitriptyline or trazodone, or antihistamines, such as diphenhydramine—is
mixed.89,90 Nevertheless, in the authors’ practice, when faced with insomnia or
sleep deprivation, patient education or other behavioral methods are attempted
first, and sleep-aid agents with less of an epileptogenic burden such as melatonin
or gabapentin are preferred.
Avoidance of flashing light exposure can be surprisingly difficult. For
example, police and ambulance strobes or the stark shadows seen on bright
winter days while driving through the woods can serve as unexpected sources of
strong photic stimulation to drivers and passengers alike. Electronic devices
feature flashing lights and rapid screen redraws, and video games still feature
scenes that can provoke seizures (although guidelines for filtering especially
potent light wavelengths have been legislated).91 Special sunglasses can be tinted
with colors specifically to block particularly ictogenic light spectra, although
plain gray glasses may have equal effects.92,93 Patients can be quite clever in
self-treatment; one patient of the authors whose seizures were triggered by the
flashing “hold” button on her office phone masked most of the light with finger
nail polish, fixing the problem.

CONCLUSION
Medically responsive epilepsy is more common than medically intractable
epilepsy. Despite having a relatively straightforward prognosis, the longitudinal
treatment of the population who are seizure free requires vigilance and
education. The fundamental basics of care in this population center on a working
knowledge of antiseizure medications, patient adherence, and factors that
could precipitate seizures.
REFERENCES
1 Brodie MJ, Barry SJ, Bamagous GA, et al. Patterns 10 Boggs JG. Mortality associated with status
of treatment response in newly diagnosed epilepticus. Epilepsy Curr 2004;4(1):25–27.
epilepsy. Neurology 2012;78(20):1548–1554. doi:10.1111/j.1535-7597.2004.04110.x.
doi:10.1212/WNL.0b013e3182563b19.
11 Lathers CM, Koehler SA, Wecht CH, Schraeder PL.
2 Wieser HG, Blume WT, Fish D, et al. ILAE Forensic antiepileptic drug levels in autopsy cases
Commission Report. Proposal for a new of epilepsy. Epilepsy Behav 2011;22(4):778–785.
classification of outcome with respect to doi:10.1016/j.yebeh.2011.10.011.
epileptic seizures following epilepsy surgery.
12 Michaelis R, Tang V, Wagner JL, et al. Cochrane
Epilepsia 2001;42(2):282–286. doi:10.1046/
systematic review and meta-analysis of the
j.1528-1157.2001.35100.x.
impact of psychological treatments for people
3 Beghi E, Carpio A, Forsgren L, et al. with epilepsy on health-related quality of life.
Recommendation for a definition of acute Epilepsia 2018;59(2):315–332. doi:10.1111/epi.13989.
symptomatic seizure. Epilepsia 2010;51(4):
13 Al-Kattana M, Afifib L, Shamloula R, El Din
671–675. doi:10.1111/j.1528-1167.2009.02285.x.
Mostafaa E. Assessment of precipitating factors
4 Kwan P, Arzimanoglou A, Berg AT, et al. Definition of breakthrough seizures in epileptic patients.
of drug-resistant epilepsy: consensus proposal Egypt J Neurol Psychiatr Neurosurg 2015;52(3):
by the ad hoc Task Force of the ILAE Commission 165–171.
on Therapeutic Strategies. Epilepsia 2010;51(6):
14 Sabate E. Adherence to long-term therapies:
1069–1077. doi:10.1111/j.1528-1167.2009.02397.x.
evidence for action. World Health Organization
5 Westover MB, Cormier J, Bianchi MT, et al. website. who.int/chp/knowledge/publications/
Revising the “Rule of Three” for inferring seizure adherence_full_report.pdf?ua=1. Accessed
freedom. Epilepsia 2012;53(2):368–376. January 31, 2019.
doi:10.1111/j.1528-1167.2011.03355.x.
15 Osterberg L, Blaschke T. Adherence to medication.
6 Divino V, Petrilla AA, Bollu V, et al. Clinical and N Engl J Med 2005;353(5):487–497. doi:10.1056/
economic burden of breakthrough seizures. NEJMra050100.
Epilepsy Behav 2015;51:40–47. doi:10.1016/
16 Malek N, Heath CA, Greene J. A review of
j.yebeh.2015.06.013.
medication adherence in people with epilepsy.
7 Bonnett LJ, Powell GA, Tudur Smith C, Marson AG. Acta Neurol Scand 2017;135(5):507–515. doi:10.1111/
Risk of a seizure recurrence after a breakthrough ane.12703.
seizure and the implications for driving: further
17 Modi AC, Rausch JR, Glauser TA. Patterns of
analysis of the standard versus new antiepileptic
nonadherence to antiepileptic drug therapy in
drugs (SANAD) randomised controlled trial.
children with newly diagnosed epilepsy. JAMA
BMJ Open 2017;7(7):e015868. doi:10.1136/bmjopen-
2011;305(16):1669–1676. doi:10.1001/jama.2011.506.
2017-015868.
18 Modi AC, Ingerski LM, Rausch JR, et al. White coat
8 Kaddumukasa M, Kaddumukasa M, Matovu S,
adherence over the first year of therapy in
Katabira E. The frequency and precipitating
pediatric epilepsy. J Pediatr 2012;161(4):695.
factors for breakthrough seizures among
e1–699.e1. doi:10.1016/j.jpeds.2012.03.059.
patients with epilepsy in Uganda. BMC Neurol
2013;13:182. doi:10.1186/1471-2377-13-182. 19 Cramer JA, Scheyer RD, Mattson RH. Compliance
declines between clinic visits. Arch Intern Med
9 Marson AG, Al-Kharusi AM, Alwaidh M, et al. The
1990;150(7):1509–1510. doi:10.1001/archinte.1990.
SANAD study of effectiveness of carbamazepine,
00390190143023.
gabapentin, lamotrigine, oxcarbazepine, or
topiramate for treatment of partial epilepsy: an 20 Cramer JA, Mattson RH, Prevey ML, et al. How
unblinded randomised controlled trial. Lancet often is medication taken as prescribed? A
2007;369(9566):1000–1015. doi:10.1016/S0140- novel assessment technique. JAMA 1989;
6736(07)60460-7. 261(22):3273–3277. doi:10.1001/jama.1989.
03420220087032.
358 APRIL 2019

21 Bautista RE, Rundle-Gonzalez V. Effects of 34 Patsalos PN, Berry DJ, Bourgeois BF, et al.
antiepileptic drug characteristics on medication Antiepileptic drugs–best practice guidelines for
adherence. Epilepsy Behav 2012;23(4):437–441. therapeutic drug monitoring: a position paper by
doi:10.1016/j.yebeh.2012.02.002. the subcommission on therapeutic drug
monitoring, ILAE Commission on Therapeutic
22 Cramer J, Vachon L, Desforges C, Sussman NM.
Strategies. Epilepsia 2009:49(7):1239–1276.
Dose frequency and dose interval compliance
doi: 10.1111/j.1528-1167.2008.01561.x.
with multiple antiepileptic medications during a
controlled clinical trial. Epilepsia 1995;36(11): 35 Reimers A, Berg JA, Burns ML, et al. Reference
1111–1117. doi:10.1111/j.1528-1157.1995.tb00469.x. ranges for antiepileptic drugs revisited: a practical
approach to establish national guidelines. Drug
23 Ferrari CM, de Sousa RM, Castro LH. Factors
Des Devel Ther 2018;12:271–280. doi:10.2147/
associated with treatment non-adherence in
DDDT.S154388.
patients with epilepsy in Brazil. Seizure 2013;
22(5):384–389. doi:10.1016/j.seizure.2013.02.006. 36 Reimers A. Hormone replacement therapy with
estrogens may reduce lamotrigine serum
24 Shetty J, Greene SA, Mesalles-Naranjo O,
concentrations: a matched case-control study.
Kirkpatrick M. Adherence to antiepileptic drugs
Epilepsia 2017;58(1):e6–e9. doi:10.1111/epi.13597.
in children with epilepsy in a Scottish population
cohort. Dev Med Child Neurol 2016;58(5): 37 Reimers A, Helde G, Brodtkorb E. Ethinyl
469–474. doi:10.1111/dmcn.12942. estradiol, not progestogens, reduces lamotrigine
serum concentrations. Epilepsia 2005;46(9):
25 Jacob L, Hamer HM, Kostev K. Adherence to
1414–1417. doi:10.1111/j.1528-1167.2005.10105.x.
antiepileptic drugs in children and adolescents:
a retrospective study in primary care settings in 38 Anderson GD, Chan LN. Pharmacokinetic drug
Germany. Epilepsy Behav 2017;75:36–41. interactions with tobacco, cannabinoids, and
doi:10.1016/j.yebeh.2017.07.001. smoking cessation products. Clin Pharmacokin
2016:55(11):1353–1368. doi:10.1007/s40262-016-
26 Piper K, Richman J, Faught E, et al. Adherence to
0400-9.
antiepileptic drugs among diverse older
Americans on Part D Medicare. Epilepsy Behav 39 Tyrlikova M, Brazdil I, Rektor I. Influence of
2017;66:68–73. doi:10.1016/j.yebeh.2016.10.017. nicotine on fluctuation of phenytoin plasma
level. Epilepsy Behav 2010;17(4):607. doi:10.1016/
27 Bautista RE, Graham C, Mukardamwala S. Health
j.yebeh.2010.01.123.
disparities in medication adherence between
African-Americans and Caucasians with epilepsy. 40 Kaiboriboon K, Guevara M, Alldredge BK.
Epilepsy Behav 2011;22(3):495–498. doi:10.1016/ Understanding herb and dietary supplement use
j.yebeh.2011.07.032. in patients with epilepsy. Epilepsia 2009;50(8):
1927–1932. doi:10.1111/j.1528-1167.2009.02090.x.
28 Wang S, Chen C, Jin B, et al. The association of
psychosocial variables with adherence to 41 Samuels N, Finkelstein Y, Singer SR, Oberbaum M.
antiepileptic drugs in patients with temporal Herbal medicine and epilepsy: proconvulsive
lobe epilepsy. Epilepsy Behav 2017;77:39–43. effects and interactions with antiepileptic
doi:10.1016/j.yebeh.2017.09.024. drugs. Epilepsia 2008;49(3):373–380. doi:10.1111/
j.1528-1167.2007.01379.x.
29 Ettinger AB, Good MB, Manjunath R, et al. The
relationship of depression to antiepileptic drug 42 Tyagi A, Delanty N. Herbal remedies, dietary
adherence and quality of life in epilepsy. supplements, and seizures. Epilepsia 2003;44(2):
Epilepsy Behav 2014;36:138–143. doi:10.1016/ 228–235. doi:10.1046/j.1528-1157.2003.19902.x.
j.yebeh.2014.05.011.
43 Kupiec T, Raj V. Fatal seizures due to potential
30 da Mota Gomes M, Navarro T, Keepanasseril A, herb-drug interactions with Ginkgo biloba. J Anal
et al. Increasing adherence to treatment in Toxicol 2005;29(7):755–758. doi:10.1093/
epilepsy: what do the strongest trials show? Acta jat/29.7.755.
Neurol Scand 2017;135(3):266–272. doi:10.1111/
44 Vossler DG, Weingarten M, Gidal BE, American
ane.12668.
Epilepsy Society Treatments Committee.
31 Dash D, Sebastian TM, Aggarwal M, Tripathi M. Summary of antiepileptic drugs available in the
Impact of health education on drug adherence United States of America. Epilepsy Curr 2018;
and self-care in people with epilepsy with low 18(4 suppl 1):1–26. doi:10.5698/1535-7597.18.4s1.1.
education. Epilepsy Behav 2015;44:213–217.
45 Johannessen Landmark C, Henning O,
doi:10.1016/j.yebeh.2014.12.030.
Johannessen SI. Proconvulsant effects of
32 Pakpour AH, Gholami M, Esmaeili R, et al. antidepressants - what is the current evidence?
A randomized controlled multimodal behavioral Epilepsy Behav 2016;61:287–291. doi:10.1016/
intervention trial for improving antiepileptic drug j.yebeh.2016.01.029.
adherence. Epilepsy Behav 2015;52(pt A):133–142.
46 Alper K, Schwartz KA, Kolts RL, Khan A. Seizure
doi:10.1016/j.yebeh.2015.08.036.
incidence in psychopharmacological clinical
33 Tang F, Zhu G, Jiao Z, et al. The effects of trials: an analysis of Food and Drug Administration
medication education and behavioral intervention (FDA) summary basis of approval reports. Biol
on Chinese patients with epilepsy. Epilepsy Behav Psychiatry 2007;62(4):345–354. doi:10.1016/
2014;37:157–164. doi:10.1016/j.yebeh.2014.05.017. j.biopsych.2006.09.023.

47 Finamore JM, Sperling MR, Zhan T, et al. Seizure 60 Fountain NB, Kim JS, Lee SI. Sleep deprivation
outcome after switching antiepileptic drugs: a activates epileptiform discharges independent
matched, prospective study. Epilepsia 2016; of the activating effects of sleep. J Clin
57(8):1294–1300. doi:10.1111/epi.13435. Neurophysiol 1998;15(1):69–75.
48 Kesselheim AS, Stedman MR, Bubrick EJ, et al. 61 Xu X, Brandenburg NA, McDermott AM, Bazil CW.
Seizure outcomes following the use of generic Sleep disturbances reported by refractory
versus brand-name antiepileptic drugs: a partial-onset epilepsy patients receiving
systematic review and meta-analysis. Drugs 2010; polytherapy. Epilepsia 2006;47(7):1176–1183.
70(5):605–621. doi:10.2165/10898530- doi:10.1111/j.1528-1167.2006.00591.x.
000000000-00000.
62 Malow BA, Bowes RJ, Lin X. Predictors of
49 Berg M, Welty TE, Gidal BE, et al. Bioequivalence sleepiness in epilepsy patients. Sleep 1997;20(12):
between generic and branded lamotrigine in 1105–1110. doi:10.1093/sleep/20.12.1105.
people with epilepsy: the EQUIGEN randomized
63 de Weerd A, de Haas S, Otte A, et al. Subjective
clinical trial. JAMA Neurol 2017;74(8):919–926.
sleep disturbance in patients with partial
doi:10.1001/jamaneurol.2017.0497.
epilepsy: a questionnaire-based study on
50 Andermann F, Duh MS, Gosselin A, Paradis PE. prevalence and impact on quality of life.
Compulsory generic switching of antiepileptic Epilepsia 2004;45(11):1397–1404. doi:10.1111/j.0013-
drugs: high switchback rates to branded 9580.2004.46703.x.
compounds compared with other drug classes.
64 Vendrame M, Yang B, Jackson S, Auerbach SH.
Epilepsia 2007;48(3):464–469. doi:10.1111/j.1528-
Insomnia and epilepsy: a questionnaire-based
1167.2007.01007.x.
study. J Clin Sleep Med 2013;9(2):141–146.
51 Kesselheim AS, Bykov K, Gagne JJ, et al. Switching doi:10.5664/jcsm.2410.
generic antiepileptic drug manufacturer not
65 Quigg M, Gharai S, Ruland J, et al. Insomnia in
linked to seizures: a case-crossover study.
epilepsy is associated with continuing seizures
Neurology 2016;87(17):1796–1801. doi:10.1212/
and worse quality of life. Epilepsy Res 2016;122:
WNL.0000000000003259.
91–96. doi:10.1016/j.eplepsyres.2016.02.014.
52 Kesselheim AS, Misono AS, Shrank WH, et al.
66 Rajna P, Veres J. Correlations between night
Variations in pill appearance of antiepileptic
sleep duration and seizure frequency in temporal
drugs and the risk of nonadherence. JAMA Intern
lobe epilepsy. Epilepsia 1993;34(3):574–579.
Med 2013;173(3):202–208. doi:10.1001/2013.
doi:10.1111/j.1528-1157.1993.tb02598.x.
jamainternmed.997.
67 Rodenbeck A, Hajak G. Neuroendocrine
53 Vossler DG, Anderson GD, Bainbridge J. AES
dysregulation in primary insomnia. Rev Neurol
position statement on generic substitution of
(Paris) 2001;157(11 pt 2):S57–S61.
antiepileptic drugs. Epilepsy Curr 2016;16(3):
209–211. doi:10.5698/1535-7511-16.3.209. 68 van Campen JS, Jansen FE, Pet MA, et al. Relation
between stress-precipitated seizures and the
54 Frucht MM, Quigg M, Schwaner C, Fountain NB.
stress response in childhood epilepsy. Brain 2015;
Distribution of seizure precipitants among
138(pt 8):2234–2248. doi:10.1093/brain/awv157.
epilepsy syndromes. Epilepsia 2000;41(12):
1534–1539. 69 Quigg M, Straume M, Menaker M, Bertram EH.
Epilepsy effects circadian rhythm of body
55 Ferlisi M, Shorvon S. Seizure precipitants
temperature in an animal model of mesial
(triggering factors) in patients with epilepsy.
temporal lobe epilepsy [abstract/poster].
Society for Research in Biological Timing.
j.yebeh.2014.02.019.
Florida: J Biological Rhythms, 1998.
56 Wassenaar M, Kasteleijn-Nolst Trenité DG, de
70 Quigg M, Clayburn H, Straume M, Menaker M,
Haan GJ, et al. Seizure precipitants in a
Bertram EH. Acute seizures may induce phase
community-based epilepsy cohort. J Neurol 2014;
shifts in the circadian rhythm of body
261(4):717–724. doi:10.1007/s00415-014-7252-8.
temperature in kindled rats [abstract/platform].
57 Dubé CM, Molet J, Singh-Taylor A, et al. AES. San Diego: Epilepsia, 1998:136.
Hyper-excitability and epilepsy generated by
71 Ishida S, Yamashita Y, Matsuishi T, et al.
chronic early-life stress. Neurobiol Stress 2015;2:
Photosensitive seizures provoked while viewing
10–19. doi:10.1016/j.ynstr.2015.03.001.
“pocket monsters,” a made-for-television
58 Klein P, van Passel L. Effect of stress related to animation program in Japan. Epilepsia 1998;39(12):
the 9/11/2001 terror attack on seizures in patients 1340–1344. doi:10.1111/j.1528-1157.1998.tb01334.x.
with epilepsy. Neurology 2005;64(10):1815–1816.
72 Trenité DG. Photosensitivity, visually sensitive
doi:10.1212/01.WNL.0000162026.52300.3C.
seizures and epilepsies. Epilepsy Res 2006;
59 Balamurugan E, Aggarwal M, Lamba A, et al. 70(suppl 1):269–279. doi:10.1016/j.eplepsyres.
Perceived trigger factors of seizures in persons 2006.02.012.
with epilepsy. Seizure 2013;22(9):743–747.
doi:10.1016/j.seizure.2013.05.018.
360 APRIL 2019

73 Nakken KO, Bjørholt PG, Johannessen SI, et al. 84 Lundgren T, Dahl J, Yardi N, Melin L. Acceptance
Effect of physical training on aerobic capacity, and commitment therapy and yoga for
seizure occurrence, and serum level of drug-refractory epilepsy: a randomized
antiepileptic drugs in adults with epilepsy. controlled trial. Epilepsy Behav 2008;13(1):
Epilepsia 1990;31(1):88–94. doi:10.1111/j.1528- 102–108. doi:10.1016/j.yebeh.2008.02.009.
1157.1990.tb05365.x.
85 McLaughlin DP, McFarland K. A randomized trial
74 McAuley JW, Long L, Heise J, et al. A prospective of a group based cognitive behavior therapy
evaluation of the effects of a 12-week outpatient program for older adults with epilepsy: the
exercise program on clinical and behavioral impact on seizure frequency, depression and
outcomes in patients with epilepsy. Epilepsy psychosocial well-being. J Behav Med 2011;34(3):
Behav 2001;2(6):592–600. 201–207. doi:10.1007/s10865-010-9299-z.
75 Heckmatt J, Shaikh AA, Swash M, Scott DF. 86 Gandy M, Karin E, Fogliati VJ, et al. A feasibility
Seizure induction by alcohol in patients with trial of an Internet-delivered and transdiagnostic
epilepsy experience in two hospital clinics. cognitive behavioral therapy treatment program
J R Soc Med 1990;83(1):6–9. doi:10.1177/ for anxiety, depression, and disability among
014107689008300105. adults with epilepsy. Epilepsia 2016;57(11):
1887–1896. doi:10.1111/epi.13569.
76 Gordon E, Devinsky O. Alcohol and marijuana:
effects on epilepsy and use by patients with 87 Haut SR, Lipton RB, Cornes S, et al. Behavioral
epilepsy. Epilepsia 2001;42(10):1266–1272. interventions as a treatment for epilepsy: a
doi:10.1046/j.1528-1157.2001.19301.x. multicenter randomized controlled trial.
Neurology 2018;90:e963–e970. doi:10.1212/
77 Hauser WA, Ng SK, Brust JC. Alcohol, seizures,
WNL.0000000000005109.
and epilepsy. Epilepsia 1988;29(suppl 2):
S66–S78. doi:10.1111/j.1528-1157.1988.tb05800.x. 88 Qaseem A, Kansagara D, Forciea MA, et al.
Management of chronic insomnia disorder in
78 Samsonsen C, Sand T, Bråthen G, et al. The
adults: a clinical practice guideline from the
impact of sleep loss on the facilitation of
American College of Physicians. Ann Intern Med
seizures: a prospective case-crossover study.
2016;165(2):125–133. doi:10.7326/M15-2175.
Epilepsy Res 2016;127:260–266. doi:10.1016/
j.eplepsyres.2016.09.014. 89 Jang DH, Manini AF, Trueger NS, et al. Status
epilepticus and wide-complex tachycardia
79 Kobau R, DiIorio C. Epilepsy self-management:
secondary to diphenhydramine overdose. Clin
a comparison of self-efficacy and outcome
Toxicol (Phila) 2010;48(9):945–948. doi:10.3109/
expectancy for medication adherence and
15563650.2010.527850.
lifestyle behaviors among people with epilepsy.
Epilepsy Behav 2003;4(3):217–225. doi:10.1016/ 90 Hill T, Coupland C, Morriss R, et al. Antidepressant
S1525-5050(03)00057-X. use and risk of epilepsy and seizures in people
aged 20 to 64 years: cohort study using a primary
80 Baldin E, Hauser WA, Pack A, Hesdorffer DC.
care database. BMC Psychiatry 2015;15:315.
Stress is associated with an increased risk of
doi:10.1186/s12888-015-0701-9.
recurrent seizures in adults. Epilepsia 2017;58:
1037–1046. doi:10.1111/epi.13741. 91 Fisher RS, Harding G, Erba G, Barkley GL, Wilkins A.
Photic- and pattern-induced seizures: review for
81 Privitera M, Walters M, Lee I, et al. Characteristics
the Epilepsy Foundation of America Working
of people with self-reported stress-precipitated
Group. Epilepsia 2005;46(9):1426–1441. doi:10.1111/
seizures. Epilepsy Behav 2014;41:74–77.
j.1528-1167.2005.31305.x.
doi:10.1016/j.yebeh.2014.09.028.
92 Takahashi T, Tsukahara Y. Usefulness of blue
82 Thapar A, Kerr M, Harold G. Stress, anxiety,
sunglasses in photosensitive epilepsy. Epilepsia
depression, and epilepsy: investigating the
1992;33(3):517–521. doi:10.1111/j.1528-1157.1992.
relationship between psychological factors and
tb01702.x.
seizures. Epilepsy Behav 2009;14(1):134–140.
doi:10.1016/j.yebeh.2008.09.004. 93 Wilkins AJ, Baker A, Amin D, et al. Treatment of
photosensitive epilepsy using coloured glasses.
83 McKee HR, Privitera MD. Stress as a seizure
Seizure 1999;8(8):444–449. doi:10.1053/
precipitant: identification, associated factors,
seiz.1999.0337.
and treatment options. Seizure 2017;44:21–26.
doi:10.1016/j.seizure.2016.12.009.

REVIEW ARTICLE

Identification
ONLINE
and Treatment of

VIDEO CONTENT
Drug-Resistant Epilepsy
A VA I L A B L E O N L I N E By Ji Yeoun Yoo, MD; Fedor Panov, MD
ABSTRACT
PURPOSE OF REVIEW: Drug-resistant epilepsy is a potentially life-threatening
condition affecting one-third of people living with epilepsy. Despite
existing evidence of improved outcomes in patients who received surgical
treatment compared to continued medical treatment, epilepsy surgery
remains underused in patients with drug-resistant epilepsy. This article
discusses the gap between evidence and practice and common
misconceptions about epilepsy surgery and reviews the current diagnostic
and therapeutic surgical options.
CITE AS: RECENT FINDINGS: Three randomized controlled trials comparing the medical
versus surgical treatment for patients with drug-resistant epilepsy have
2019;25(2, EPILEPSY):362–380.
shown the superiority of surgery in controlling seizures and improving
Address correspondence to patients’ quality of life. In addition to resective surgery, neuromodulation
Dr Ji Yeoun Yoo, Mount Sinai through devices such as responsive neurostimulation and vagal nerve
Hospital, 1468 Madison Ave,
Annenberg Bldg 210, New York,
stimulation have also shown efficacy in seizure control that increases over
NY 10029, jiyeoun.yoo@mssm.edu. time. Diagnostic and therapeutic surgical tools are tailored to the needs of
each patient.
Dr Yoo has received publishing
royalties from Elsevier for the SUMMARY: Appropriate patients with drug-resistant epilepsy benefit more
book Rowan's Primer of EEG and from epilepsy surgery than from continuing medical treatment. These
has received personal
compensation as a lecturer for patients should be referred to comprehensive epilepsy centers where a
the Korean Epilepsy Society thorough presurgical workup and surgical options can be provided. The
and for serving on the advisory
board of Zimmer Biomet. Dr Yoo
gap between evidence and practice can be bridged by education,
has received research/grant community outreach, and providers’ earnest efforts to improve the quality
support from the NeuroNEXT of life of patients with epilepsy.
Program of the National
Institutes of Health/National
Institute of Neurological
Disorders and Stroke. Dr Panov INTRODUCTION
D
compensation as a consultant for rug-resistant epilepsy is defined as the “failure of adequate trials of
NeuroPace, Inc and Zimmer two tolerated, appropriately chosen and used antiepileptic drug
Biomet.
(AED) schedules (whether as monotherapies or in combination) to
UNLABELED USE OF achieve sustained seizure freedom,” which could be either 3 times
PRODUCTS/INVESTIGATIONAL the prior interseizure interval or 1 year, whichever is longer.1
USE DISCLOSURE:
Drs Yoo and Panov report no
Drug-resistant epilepsy affects about one-third of people living with epilepsy.2
disclosures. Patients with epilepsy have worse quality of life, family function, and social
support compared to other patients who are chronically ill.3 Furthermore,
© 2019 American Academy patients with drug-resistant epilepsy face an increased risk of sudden unexpected
of Neurology. death in epilepsy (SUDEP) (TABLE 4-1).4
362 APRIL 2019

Since the development of modern surgical approaches to epilepsy in the 1950s,
resection of a seizure focus has become an important treatment tool for
drug-resistant epilepsy. Introduction of modern nonresective techniques, such as
laser interstitial thermal therapy, gamma knife radiosurgery, and devices such
as vagal nerve stimulation (VNS) and responsive neurostimulation, has recently
expanded the surgical options.
The effectiveness of epilepsy surgery has been demonstrated in randomized
controlled trials, case series, and observational studies. Epilepsy surgery can
potentially convert drug-resistant epilepsy to drug-responsive epilepsy. It may
also decrease the burden of antiepileptic medication side effects, thus improving
Overview of Drug-Resistant Epilepsya TABLE 4-1
Characteristic Comments
Epidemiology Focal epilepsy and focal onset seizures, with prevalence of 50% each among those with
epilepsy, are the most common syndromes and seizure types.5
Natural history and intractability The course of epilepsy can be predicted with reasonable accuracy within the first
2 years of therapy, but early remission does not necessarily predict a favorable
long-term outcome.6
Predictors of drug resistance Predictors of drug resistance include high initial seizure frequency, underlying etiology,
and failure to respond to antiepileptic drugs early on.7
Surgical utilization and referrals Epilepsy surgery is underutilized worldwide.8,9
Cost of epilepsy Successful epilepsy surgery is more cost-effective than medical therapy, and greater
savings accrue with earlier surgery.10,11
Indications for surgery The most common criteria used to define surgical candidacy are seizure frequency
(>1 per month) and failure of ≥2 antiepileptic drugs.12 In cases of lesional epilepsy,
surgery can be considered even earlier.
Seizure outcome after surgery Approximately 60–65% of patients are seizure free after temporal lobe resection,
compared with 40% of patients after extratemporal resection.13
Use of antiepileptic drugs after There is uncertainty about the proportion of patients who discontinue antiepileptic
surgery drugs after surgery.
Neuropsychological outcomes Epilepsy surgery is associated with specific cognitive decline (most often involving
after surgery verbal memory and naming after dominant lobe resections), but cognition may also
improve in some patients.14
Psychiatric outcomes after surgery There is either improvement or no change in psychiatric outcomes after surgery.15
Quality of life after surgery Quality of life improves after surgery and is strongly influenced by seizure freedom.16
Social outcome after surgery After surgery, improvement is seen in patients’ employment, driving, and relationships.17
Complications after surgery Surgical complications are usually minor or transient; major and minor neurologic
complications were reported in 4.7% and 10.9% of patients, respectively, with
resective surgery, with the most common being minor visual field deficits (affecting one
quadrant or less).18
Mortality after surgery Mortality appears to be lower if patients are rendered seizure free after epilepsy
surgery.19
a
Modified with permission from Jette N, et al, Neurology.12 © 2012 American Academy of Neurology.

DRUG-RESISTANT EPILEPSY
quality of life. However, epilepsy surgery currently remains grossly underused.8

This article addresses why this occurs and what clinicians can do for people living
with drug-resistant epilepsy.
SURGICAL VERSUS MEDICAL TREATMENT FOR DRUG-RESISTANT EPILEPSY

Numerous case studies and observational studies about the efficacy of epilepsy
surgery have been conducted. Because of the difficulty of study design and
ethical implications in delaying surgical therapy, randomized controlled trials
comparing medical versus surgical treatment for refractory epilepsy were lacking
until 2001. Currently, three randomized controlled trials have shown the
superiority of surgery compared to continued medical treatment in patients with
drug-resistant epilepsy, not only for seizure control but also for quality of life.20
The first randomized controlled trial for adults with refractory temporal lobe
epilepsy, comparing patients who underwent immediate epilepsy surgery to
those randomly assigned to a 1-year waiting period of continued medical
treatment (this arm was usual care at that center), demonstrated the superiority
of surgical outcome. At 1 year, 58% of the surgery group were free of disabling
seizures compared to 8% in the medical group (P<.001). Improved quality of life
as well as a strong trend toward better social functioning (higher rates of
employment and school attendance) were present in the surgical versus the
medical arm.20 This pivotal study was followed by a multicenter randomized
controlled trial that compared early surgery to continued medical therapy in
patients with temporal lobe epilepsy.21 Although this study was terminated early
because of slow accrual, 11 of 15 patients in the surgery group were seizure free at
2 years, while none of the 23 patients in the medical group achieved seizure
freedom (odds ratio = ∞; 95% confidence interval, 11.8 to ∞; P<.001). Recently,
the superiority of surgery was also demonstrated in children with refractory
epilepsy (including, but not limited to temporal lobe epilepsy) in a single-center
randomized controlled trial showing 77% (44 of 57 patients) of participants in
the surgery group free of seizures at 1 year, versus 7% (4 of 59 patients) of
participants in the medical group (P<.001), as well as improved behavior and
quality of life in the surgical group.22
Beyond seizure frequency and quality of life, a significant decrease in
mortality was demonstrated in a large clinical cohort study (n = 1110, total
follow-up of 8126.62 person-years from 1986 to 2013) and was most pronounced
in patients who became free of generalized tonic-clonic seizures after surgery.23
Epilepsy surgery is also cost-effective when compared to costs of direct medical
care and long-term disability.10,11
A GAP BETWEEN EVIDENCE AND PRACTICE

Based on the Class I study in favor of epilepsy surgery20 and other literature, a
formal practice guideline in epilepsy surgery was published in 2003 jointly by the
American Academy of Neurology (AAN), American Epilepsy Society, and
American Association of Neurological Surgeons.24 It recommended that patients
for whom appropriate trials of first-line AEDs have failed should be considered
for referral to an epilepsy surgery center. Despite this practice parameter, referral
patterns for patients with drug-resistant epilepsy have not changed significantly
in the following years.8 According to one estimate, only 1% of patients with
drug-resistant epilepsy are referred to epilepsy centers,25 and this usually occurs
after an average of more than 20 years after the onset of their epilepsy.26
364 APRIL 2019

The reasons for the delays or lack of referral for surgery range from patients’ KEY POINTS
or families’ fears, health care providers’ lack of knowledge about epilepsy
● Drug-resistant epilepsy is
surgery, society or community health care access issues, or social and cultural defined as the “failure of
issues (eg, stigma).27 adequate trials of two
tolerated, appropriately
POSSIBLE BARRIERS TOWARD TIMELY REFERRAL FOR PATIENTS WITH chosen and used
antiepileptic drug schedules
(whether as monotherapies
Several common misconceptions about epilepsy surgery continue to be held by or in combination) to
health care providers.25 This section discusses issues of importance in removing achieve sustained seizure
some barriers to timely referral of patients. freedom.”
● Drug-resistant epilepsy
Safety Profile of Epilepsy Surgery affects about one-third of
Epilepsy surgery is often considered dangerous and thus is thought of as a last people living with epilepsy.
resort for patients with drug-resistant epilepsy. In fact, epilepsy surgery is
generally safe; most complications associated with epilepsy surgery are minor ● Currently, three
randomized controlled
(defined as resolving completely within 3 months of surgery, whereas major trials have shown superiority
complications persist beyond that time frame).18 According to a systematic of surgery compared
review, major and minor medical complications were reported in 1.5% and 5.1% to continued medical
of patients, respectively, following resective surgery, the most common being treatment in patients with
drug-resistant epilepsy, not
CSF leak. Major and minor neurologic complications were reported in 4.7% and
only for seizure control but
10.9% of patients, respectively, with resective surgery, with the most common also for quality of life.
being minor visual field deficits (affecting one quadrant or less). Perioperative
mortality was rare (0.4% in temporal lobe cases and 1.2% in extratemporal ● A formal practice
cases); these numbers are skewed upward by large pediatric resections in guideline by the American
Academy of Neurology,
multilobar cases and are lower for the adult population.18 American Epilepsy Society,
Cognition and memory deficits are not contraindications to surgery. Epilepsy and American Association of
surgery is associated with specific cognitive changes, most notably verbal Neurological Surgeons
memory decline (44% decline in left-sided surgery and 20% decline in right) and recommends that the
patients for whom
naming (34% decline in left-sided surgery) in anterior temporal lobectomy.14 appropriate trials of first-line
Decline in visuospatial memory was also reported in about 20% of patients antiepileptic drugs have
irrespective of the side of the surgery.14 However, paradoxical gains in verbal failed should be considered
memory (gains in 7% to 14% of patients) and visuospatial memory (gains in 10% for referral to an epilepsy
surgery center.
to 15% of patients) and an increase in verbal fluency in left-sided surgery
(increase in 27% of patients) were also reported.14 The ability of the healthy ● A gap between evidence
circuits to recover and even improve in function after the seizure focus has been and practice exists in regard
removed is postulated as the reason for the postoperative gains mentioned to treatment for patients
with drug-resistant epilepsy.
above. Neuromodulation and selective laser ablation have been shown to avoid
Although evidence clearly
some of the above cognitive problems by minimizing the effect of surgery on the dictates referral of patients
surrounding brain tissue.28,29 It is important to note that the risk of cognitive with drug-resistant epilepsy
and memory decline is less when an abnormality is seen on imaging such as to a comprehensive epilepsy
mesial temporal sclerosis, or with earlier age of seizure onset, or with preexisting center, such referrals are
still not completed due to a
memory and language deficits. Preoperative neuropsychological assessment is variety of reasons.
useful for creating a risk-benefit profile, and presurgical counseling as well as
postsurgical rehabilitation referrals could diminish such deficits that may occur.30 ● Epilepsy surgery is
generally safe; most
complications are minor
Futility of Future Sequential Antiepileptic Drug Trials After Initial Drug and transient.
Resistance Is Clearly Established
The definition of drug-resistant epilepsy was driven in part from a prospective
study that showed that only 11% of patients with epilepsy became seizure free
after failure of the first AED and only 3% of patients with epilepsy became

seizure free after failure of the second AED.2 Despite the development of newer
AEDs, the rate of drug-resistant epilepsy does not seem to have significantly
changed.31 In a prospective observational study, the likelihood of seizure
freedom declined with successive drug regimens, most markedly from the first
to the third.32 Given the high likelihood of seizures remaining uncontrolled after
failure of two AEDs (ie, having recurrent seizure[s] after adequate trials) and
multiple disadvantages from living with epilepsy, any patient for whom two or
more AEDs have failed should be referred to a comprehensive epilepsy center
for multidisciplinary evaluation and surgical consideration, as consistent with
current guidelines.33
Patients With Drug-Resistant Epilepsy and Nonlesional Brain Magnetic

Resonance Imaging May Still Be Candidates for Surgery
Patients who have epilepsy due to a structural abnormality seen on brain MRI
have better surgical outcomes than those without an MRI abnormality. However,
not all radiographically “normal” MRIs are indeed normal. The most common
culprit is focal cortical dysplasia.34 In one study, 60% of patients with histologically
confirmed focal cortical dysplasia had “normal” MRIs when performed and
reported outside a major epilepsy center, but only 37% of these were still
“normal” when reviewed at the epilepsy center.35 Detailed EEG analysis by
epileptologists in combination with MRI interpretation by experienced
neuroradiologists with attention to the patient’s seizure symptomatology can
help identify subtle findings. Milder forms of dysplasia may still be invisible in
high-resolution MRIs and, in these cases, other imaging studies such as positron
emission tomography (PET) or ictal single-photon emission computed
tomography (SPECT) can be helpful.
Localization of the Epileptogenic Zone Requires Multimodality Evaluation

The theoretical concept of an epileptogenic zone is defined as “the minimum
amount of cortex that must be resected (inactivated or completely disconnected)
to produce seizure freedom.”36 There is not one ideal tool for epileptogenic zone
delineation; it is rather an approximation through multiple modalities, including
MRI, EEG, and neuropsychological tests. A focal lesion seen on a patient’s MRI
may or may not be responsible for the person’s epilepsy, and careful review
through multiple modalities are still necessary.
Eloquent Cortical Epileptogenic Foci May Still Be Amenable to

Epilepsy Surgery
Careful mapping of brain function is performed before resective surgery,
especially when the epileptogenic zone is in or near eloquent cortex. The regions
with essential function can be delineated and protected during resection.
Involvement of eloquent cortex is not a contraindication to epilepsy surgery.
When the resection of the epileptogenic zone is not possible because of a likely
functional deficit, neuromodulation such as responsive neurostimulation (RNS)
or VNS can be an effective treatment option. In a recent open-label observational
study of patients with medically intractable seizures arising from eloquent and
other neocortical areas who were implanted with RNS, median seizure reduction
was 70% in patients with frontal and parietal seizure onsets, 58% in patients
with temporal neocortical onset, and 51% in those with multilobar onsets, without
significant acute or chronic neurologic deficits.37
366 APRIL 2019

Patients With Drug-Resistant Epilepsy With Multifocal Brain Lesions or KEY POINTS
Generalized Interictal Spiking May Still Be Candidates for Epilepsy Surgery
● Cognition and memory
A treatment-viable seizure onset zone can still be identified even if multiple or deficits are not
diffuse lesions are seen on MRI or if multifocal spikes are seen on EEG. The contraindications to
patient’s habitual seizure symptomatology along with EEG and imaging findings epilepsy surgery.
can successfully delineate the epileptogenic zones that may involve only a part
● The risk of cognitive and
of a lesion, leading to successful resection, laser ablation, or neuromodulation.
memory decline is less when
Bilateral spikes can be misleading, especially when the epileptiform activity an abnormality is seen in
comes from a frontal or occipital region because of rapid synchronization. It can imaging such as mesial
be especially tricky when encephalomalacia is present in the epileptogenic zone, temporal sclerosis, or with
as the epileptiform discharges can be falsely lateralizing. Careful review of the earlier age of seizure onset,
or with preexisting memory
imaging along with the patient’s habitual seizure symptomatology and EEG and language deficits.
analysis (eg, phase reversal, lateralizing higher amplitude) can help identify a
unilateral onset. What seems to be generalized epilepsy or bilateral spikes on EEG ● After failure (having
may still have a unilateral seizure onset zone. In cases of drug-resistant truly recurrent seizures despite
adequate trials) of two
generalized epilepsy, VNS can be considered, which has a 60% reported antiepileptic drugs, the
responder rate (>50% seizure reduction).38 chance of a third antiepileptic
drug effectively controlling
Patients With Drug-Resistant Epilepsy and Psychiatric Comorbidity May a patient’s seizures is
very low.
Still Be Candidates for Epilepsy Surgery
Psychiatric comorbidity is not a contraindication to surgery. In a systematic ● “Normal” brain MRIs are
review of literature regarding psychiatric outcomes after epilepsy surgery, not always normal. A careful
generally no changes or some improvements in psychiatric condition were review by dedicated
seen.15 De novo psychiatric issues postsurgery were relatively rare, ranging from neuroradiologists, with
attention to the patient’s
milder syndromes such as adjustment disorder (18%) to more severe conditions seizure symptomatology,
such as psychosis (1%). In a recent prospective case-control study, the patients and correlation with EEG
who received epilepsy surgery experienced a significant decrease in findings, can help identity
psychopathologic alterations, distress perception, anxiety, and depression when subtle findings.
compared to those who continued with medical treatment.39 Current evidence ● A focal lesion is not
also supports that cognition and mood remain stable with all neurostimulation always the culprit for the
therapies, with some evidence that there may be improvement.40 It should be patient’s seizures, and
understood that, in the setting of psychiatric illness, seizures will worsen a careful review through
multiple modalities is
patient’s quality of life and increase his or her day-to-day medical risk.
still necessary.
Epilepsy Center Referral Facilitates Comprehensive Epilepsy Care ● Involvement of

A referral to a comprehensive epilepsy center is not only for possible epilepsy eloquent cortex is not a
surgery but is also to identify a “pseudoresistance” and to adjust treatments contraindication to
epilepsy surgery. When
accordingly. This “pseudoresistance” results from medication noncompliance the resection of the
(which could be due to sensitivity to side effects), seizures that are nonepileptic epileptogenic zone is not
(including psychogenic attacks), failure to identify a treatable cause, wrong possible, a device can be an
drug for the patient’s type of epilepsy or wrong dosage or time of medication, effective treatment option.
and lifestyle issues such as substance abuse or sleep deprivation.25 In a ● A treatment-viable
comprehensive epilepsy center, these issues can be elucidated and treated by seizure onset zone can still
a multidisciplinary team approach with an improved outcome and without be identified even if multiple
surgical intervention. As an example, the often-overlooked dietary adjustment to or diffuse lesions are seen
on MRI or if multifocal
a ketogenic diet or modified Atkins diet can be very effective in seizure control.41
spikes are seen on EEG.
PRESURGICAL WORKUP
Workup with the goal of achieving seizure freedom by surgical means begins in
earnest with a referral to a comprehensive epilepsy center (FIGURE 4-1). A cohesive

FIGURE 4-1
Decision tree for patients with epilepsy.
AED = antiepileptic drug; EEG = electroencephalography; MRI = magnetic resonance imaging;
PET = positron emission tomography; SPECT = single-photon emission computed tomography.
368 APRIL 2019

team crossing medical specialty boundaries is required to design an KEY POINTS
appropriate pathway for each patient. Such teams usually include
● What seems to be
epileptologists, neuropsychologists, neurosurgeons, and neuroradiologists. Yet, generalized epilepsy or
it is the invaluable support team that may include administrators, schedulers, bilateral spikes on EEG may
nurses, and advanced practice professionals that really allows for the complex still have a unilateral seizure
process to move forward smoothly, maintaining constant communication with onset zone, and in cases of
truly generalized epilepsy,
the patient and his or her family. Such teamwork builds confidence and reduces
a device can be an
anxiety for the patient. effective option.
The basic tests required include admission for video-EEG monitoring and
brain imaging, as well as a neuropsychology evaluation. Additional tests may ● A referral to an epilepsy
include magnetoencephalography (MEG), SPECT, functional MRI (fMRI), center is not only for
possible epilepsy surgery
Wada test (intracranial sodium amobarbital procedure that establishes language but is also to identify
and memory representation of each hemisphere), PET scans, and repetition of a “pseudoresistance” and
any tests deemed outdated from previous workups. Setting appropriate to adjust treatments
expectations for patients and learning the goals of them and their families are accordingly.
critical. At times, seizure freedom may not be achievable, but significant ● Invasive diagnostic tools
medication reduction or decrease in the severity of the episodes may be a realistic are used for delineating the
target. When noninvasive data are insufficient to delineate the epileptogenic intracranial seizure onset
zone, invasive monitoring using intracranial electrodes is indicated to zone and for functional brain
mapping. The diagnostic
anatomically define the epileptogenic zone as well as the nearby functional areas.
tools include grids, strips,
The surgical treatment of most patients is divided into the diagnostic phase depth electrodes, or stereo-
and the therapeutic phase; albeit, some patients will only go as far as the EEG, and these tools can be
diagnostic phase, while, in a select group, a single therapeutic stage may used as dictated by the
needs of the specific
be appropriate.
patient.
DIAGNOSTIC EPILEPSY SURGERY

Invasive diagnostic tools are used for delineating the intracranial seizure onset
zone and for functional brain mapping. The diagnostic tools include grids, strips,
depth electrodes, or stereo-EEGs, and these tools can be used as dictated by the
needs of the specific patient.
Grids and Strips

The classic surgical workup preferred for decades includes a craniotomy for
placement of subdural grids and strips over the suspicious areas of the brain to
elucidate the seizure focus during an inpatient monitoring phase that follows
(FIGURE 4-2). Grids and strip studies offer great coverage of the cortical surface
and allow the spatial contiguity in detecting the ictal onset and its immediate
spread.42 It is an efficacious option, especially when the hypothetical
epileptogenic zone is anatomically restricted to superficial cortical areas and is
in close relation with eloquent cortex.43 When necessary, it can be combined
with one or more depth electrodes to sample a deep focus (foci).
Stereo-Electroencephalography
Stereo-EEG uses three-dimensional analysis of many contacts placed into the
brain with stereotactic guidance with the aim of delineating the seizure
foci and the network of seizure propagation (FIGURE 4-3). Stereo-EEG stresses
the investigation of connectivity and spread with the goal of finding a node
where surgical intervention may be the most beneficial. It allows sampling of
deep cortical areas inaccessible with grids or strips. However, it lacks the spatial
continuity because the density of the electrodes cannot be as high as with grids. It

KEY POINT also relies heavily on the

implantation strategy with a
● Different surgical options
are available, including
preimplantation hypothesis.42,43
resection, laser ablation,
and neuromodulatory Trends
devices, with therapeutic
surgery tailored to the The grids and strips approach is
specific patient. favored when the epilepsy has
been lateralized and a high
likelihood of eloquent cortex
involvement exists.
The stereo-EEG approach is
favored when both hemispheres
need to be investigated; when the
likelihood of the patient needing
or agreeing to a craniotomy is low;
and when a deep structure such
as the mesial temporal lobe,
periventricular nodular
heterotopia, insula, or the
cingulate gyrus is hypothesized
as the culprit (TABLE 4-2).
A combination workup, while
technically demanding, is possible
with the placement of the
stereo-EEG electrodes first and
then proceeding to placement of
FIGURE 4-2
strips and/or grids. A combination
Grids and strips. A, Photograph of a craniotomy of the grid/strips and depth
and grid. B, X-ray of grid and strips. electrodes can also be used.
FIGURE 4-3
Stereo-EEG. A, Photograph of stereo-EEG electrode placement. B, X-ray of stereo-EEG
electrodes.
370 APRIL 2019

THERAPEUTIC EPILEPSY SURGERY
Different surgical options are available, including resection, laser ablation, and
neuromodulatory devices (RNS, VNS, deep brain stimulation), with therapeutic
surgery tailored to the specific patient.
This section reviews case examples of each of the three surgical strategies, in
turn highlighting the advantages of each.
Resection
Resective surgery remains the gold standard against which all other procedures
are judged, as the majority of the randomized controlled trials were based on
this surgical method (CASE 4-1). The 2001 pivotal article by Wiebe and
colleagues20 outlined the benefits of surgical treatment and set up the number
needed to treat at two (only two patients need to undergo the procedure to make
a drastic difference in the life of one). This number compares favorably to many
other surgical procedures (eg, the number needed to treat for carotid
endarterectomy varies from 6 to 19.)44 Variations on the theme of resection
include callosotomy, hemispherectomy, and functional hemispherotomy, as
well as numerous disconnections designed specifically for the pathology
encountered. While beyond the scope of this article, the unifying theme of
these operations is localization of the seizure focus and its anatomic
(resection) or functional (disconnection, callosotomy) removal from the rest
of the functioning brain.
Thermal Ablation
If the seizure focus is deemed resectable without a significant decrease in
neuropsychological function or damage to another eloquent area such as
speech or motor areas, yet is difficult to get to with an open surgery, an
ablation may be performed. Such a procedure retains the minimally invasive
nature of the treatment while achieving comparable results to open resection
in well-selected patients.45 Data support improved neuropsychological function
after laser treatment compared to the open resection.29 Transcranial focused
Comparison of Stereo-Electroencephalography Versus Grids and Strips TABLE 4-2

Monitoring in the Evaluation of Patients with Epilepsy
Stereo-Electroencephalography in Preference to Grids or Strips

◆ Sulcal cortical areas including focal cortical dysplasia
◆ Deeply located cortical areas (eg, insulo-opercular complex, limbic system)
◆ Deep-seated or periventricular lesions such as hypothalamic hamartoma and
periventricular heterotopia
◆ Bilateral exploration
◆ MRI-negative patients
Grids or Strips in Preference to Stereo-Electroencephalography
◆ An extensive unilateral epileptogenic zone related to a lesion, requiring surface as well as
selected deep sampling
◆ Mapping the speech area when the epileptogenic zone is located within

ultrasound, an ultrasound-induced thermal ablation, may potentially be used

for treatment of epilepsy as well.46 While the success rate long-term may be
slightly lower than open resection, the option of trying this method first and
turning to more invasive surgery later is attractive to both patients and
practitioners. Such an approach may go a long way in breaking down the
barriers to the surgical treatment of epilepsy (CASE 4-2).
Neuromodulation
Patients with drug-resistant epilepsy with an epileptogenic focus located in
eloquent cortex are the most difficult to treat. As resection or ablation are not
valid options because of the potential damage to the patients’ function, these
patients have previously been relegated to continuing futile medical
management or placement of a vagal nerve stimulator with moderate efficacy in
seizure reduction (CASE 4-3).
CASE 4-1 A 58-year-old woman presented with a history of childhood-onset,

frequent, highly stereotyped, and brief nocturnal events characterized
by sudden awakening from sleep with a feeling of being in a trance and
having involuntary movements such as body rocking and grimacing,
typically lasting 10 to 15 seconds and clustering over 1 to 2 hours nightly.
Her seizures were highly medically refractory to trials of 10 different
antiepileptic drugs.
Brain MRI showed an area of focal left cingulate cortical thickening
with gray-white junctional blurring suspicious for focal cortical
dysplasia (FIGURE 4-4). Routine EEG was normal, and scalp-recorded
ictal video-EEG monitoring was nonlocalizing. Ictal single-photon
emission computed tomography (SPECT) demonstrated anterior
superior bifrontal hyperperfusion consistent with a possible focal
epileptogenic zone.
Stereo-EEG was implanted to cover the hypothesized areas of focal
seizure onset from the left hemisphere as determined by neuroimaging,
with contacts sampling predominantly from the left cingulate, frontal,
and insular regions. Several of her characteristic focal seizures with
altered awareness were captured (VIDEO 4-1, links.lww.com/CONT/
A274), which demonstrated highly stereotyped clinical behavior of
facial grimacing and oral and axial body rocking and writhing
automatisms. Ictal onset was consistently localized at the left
cingulate cortex, with nearly instantaneous spread to the left insular
region. Stereo-EEG demonstrated rhythmic sharp waves and fast
activity in contacts LN1-LN8, LY1-LY3, LZ21-LZ23, and LT4-LT6,
confirming ictal onset in the left cingulate region. A left cingulate focal
cortical resection demonstrated cortical type 2b dysplasia and has
rendered the patient seizure free at 2-year follow-up.
372 APRIL 2019

In 2013, the US Food and Drug Administration (FDA) approved a closed loop
responsive neurostimulation device for detection and treatment of drug-resistant
epilepsy. The latest 7-year data show a 72% mean reduction in seizure frequency.37
VNS remains an important tool for patients with generalized epilepsy or for
those who are not candidates for the previously described approaches. A recent
promising feature of VNS allows for the detection of cardiac rhythm and
stimulation based on this marker of possible seizure activity. The efficacy of VNS
does improve over time, with 60% of patients experiencing a significant
response (>50% seizure reduction), but the goal of seizure freedom is unlikely to
be attained with this therapy (8% seizure freedom after 2 years).38
Deep brain stimulation has been recently approved by the FDA for the
treatment of epilepsy. SANTE (Stimulation of the Anterior Nuclei of Thalamus
for Epilepsy), a multicenter, double-blind, randomized trial for patients with
drug-resistant focal epilepsy (n = 110), showed significant long-term seizure
FIGURE 4-4
Brain MRI of the patient in CASE 4-1 with drug-resistant focal cingulate epilepsy. Axial (A),
sagittal (B), and coronal (C) double inversion recovery sequences demonstrate subtle focal
thickening and blurring of the gray/white matter junction (arrows).
Courtesy of David B. Burkholder, MD; Jeffrey W. Britton, MD, FAAN; Elson L. So, MD, FAAN; Cheolsu Shin,
MD; Sotiris Mitropanopoulos, MD; Lily Wong-Kisiel, MD; Gregory A. Worrell, MD; Jamie Van Gompel, MD;
and Erik K. St. Louis, MD, MS, FAAN.
This case illustrates the utility of stereo-EEG in clarifying the epileptogenic COMMENT
zone, establishing that the brain MRI lesion was directly associated with
ictal onset of this patient’s habitual clinical seizures, and enabling the
surgical epilepsy team to pinpoint the boundaries of the surgical epileptic
focus necessary for producing an excellent outcome. Previously, patients
with similar presentations may have required widespread implantation of
subdural strip and grid electrodes, which pose risks of greater
perioperative morbidity. The advent of stereo-EEG has improved access
for offering surgical resections to patients with drug-resistant epilepsy
that can increase patients’ chances of becoming seizure free.
Case courtesy of David B. Burkholder, MD; Jeffrey W. Britton, MD, FAAN; Elson L. So, MD, FAAN;
Cheolsu Shin, MD; Sotiris Mitropanopoulos, MD; Lily Wong-Kisiel, MD; Gregory A. Worrell, MD;
Jamie Van Gompel, MD; and Erik K. St. Louis, MD, MS, FAAN.

CASE 4-2 A 27-year-old right-handed woman was referred for an evaluation at an

epilepsy center. She had a history of focal epilepsy since age 24, and her
seizures were described as feeling “cold, chilly, and scary” followed by
staring with unresponsiveness and lip smacking. Despite being on three
antiepileptic drugs, she continued to have one to three focal seizures with
impaired awareness per month. She reported having a depressed mood.
Three years prior, she had undergone video-EEG monitoring, which
showed occasional left temporal slowing and frequent left temporal sharp
waves as well as occasional left temporal rhythmic delta activity. A
previous brain MRI had been unremarkable; however, the MRI was not
done with a “seizure protocol.”
Given the EEG findings, a repeat brain MRI with thin cuts through the
temporal lobe was done at her current presentation, which revealed a left
temporal heterotopia involving the left temporal parahippocampal and
fusiform gyri (FIGURE 4-5A). A positron emission tomography (PET) scan
showed decreased metabolism in the lateral aspect of the left temporal
lobe. Repeat video-EEG monitoring was performed to characterize her
seizures, and two typical seizures were captured with left temporal onset.
Neuropsychological testing found deficits in attention and confrontation
naming, suggesting frontal or frontotemporal dysfunction, but did not
strongly lateralize. Wada testing confirmed left-sided language and
demonstrated poor left memory function (1/12 recall) and good right
memory function (10/12 recall).
She subsequently underwent left-sided stereo-EEG implantation with
electrodes in the hippocampus; amygdala; and orbitofrontal, posterior
temporal (following the course of the heterotopia), insular, and parietal
lobes. Three typical seizures were captured with left hippocampal onset,
and the contacts within the heterotopia were involved within 2 seconds of
the seizure onset (FIGURE 4-5B). She subsequently underwent laser ablation
of the left hippocampus and heterotopia (along the inferior aspect of the
temporal horn of the left lateral ventricle) (FIGURE 4-5C).
She was seizure free at her 1-year postoperative follow-up. She was
able to get a job, and her depressive mood improved significantly. At her
1-year follow-up neuropsychological testing, she showed improvements in
both learning and memory (eg, verbal immediate recall improved from
fourth to 13th percentile, and delayed memory improved from fourth to
42nd percentile). Gains in attention and concentration were also seen.
COMMENT In this patient, the seizure symptomatology and EEG findings raised
suspicion for a medial temporal–onset epilepsy, which prompted
additional imaging with attention to the medial temporal lobe, revealing a
left temporal heterotopia. To investigate the seizure onset zone, stereo-
EEG monitoring was planned, with a hypothesis that the seizure onset was
likely from the left medial temporal lobe within or along the heterotopia.
After confirming the patient’s seizure onset zone, laser ablation was
performed to access the deep location of the heterotopia and mesial
temporal lobe, rendering this patient seizure free.
374 APRIL 2019

FIGURE 4-5
Imaging and EEG of the patient in CASE 4-2. A, Preablation coronal T1-weighted MRI
shows left temporal heterotopia involving the left temporal parahippocampal and
fusiform gyri (white arrows). B, Postablation coronal T2-weighted MRI after laser ablation
of the left hippocampus and left temporal heterotopia (white arrows). C, Stereo-EEG
showing seizure onset starting with left hippocampal fast rhythm followed by
hippocampal spike and low-voltage fast activity. Within 2 seconds, rhythmic beta
activity is seen in the heterotopia, lasting about 2 seconds (each location of the
contact is depicted). Seizures continue to evolve in the medial temporal region
(hippocampus and amygdala).
L HCP = left hippocampus.

CASE 4-3 A 37-year-old right-handed man

with a history of drug-resistant
epilepsy since the age of 2 was
referred to an epilepsy center.
His seizures started without
an aura, and witnesses had
described that he repeated
phrases like “go, go, go,”
became confused, and
sometimes walked into a
corner of the room. Despite
being on two antiepileptic
drugs, he had one bilateral
tonic-clonic seizure per month.
He was depressed.
His video-EEG showed
occasional left temporal sharp
waves, and two of his typical
seizures were captured with
left temporal onset. Brain
MRI with seizure protocol
was unremarkable. His
neuropsychological testing
showed frontotemporal
dysfunction, left worse than
the right. His Wada test
revealed left-sided language
and poor memory on both
sides, although memory on the
left (4/12) was better than the
right side (1.5/12). Given this
contradictory Wada test finding
and no obvious lesions seen on
FIGURE 4-6
MRI, he underwent bilateral Imaging and EEG of the patient in CASE 4-3.
stereo-EEG monitoring, Stereo-EEG electrode placements are shown (A).
including depth electrodes to Responsive neurostimulation device is placed with
both hippocampi and amygdala two leads into the bilateral hippocampi (B). Left
hippocampal–onset seizure (C, upper two lanes) is
(FIGURE 4-6A). Stereo-EEG detected, which triggered stimulation therapy with
showed frequent bilateral successful abolition of further seizure propagation.
hippocampal/amygdala Examples of left hippocampal–onset seizure (D)
epileptiform discharges, more and right hippocampal onset seizure (E) are shown.
376 APRIL 2019

frequent in the left than right, and two habitual seizures were captured
with the left hippocampal onset. A responsive neurostimulation device
was implanted to the bilateral hippocampi both as a long-term diagnostic
tool and as therapy (FIGURE 4-6B).
At his 1-year follow-up, his seizures had decreased from one
bilateral tonic-clonic seizure per month to three bilateral tonic-clonic
seizures in 1 year (FIGURE 4-6C); a left hippocampal onset seizure
was detected, which triggered stimulation therapy with successful
abolition of further seizure propagation. For 1 year, responsive
neurostimulation captured a total of 37 seizures, 26 from the left onset,
and two from the right (nine seizure onsets were not captured)
(FIGURE 4-6D and 4-6E). His responsive stimulation was recently revised to
stimulate one depth electrode in the left hippocampus and one strip in
the left lateral posterior temporal lobe.
This patient’s seizure symptomatology, EEG findings, and COMMENT

neuropsychological testing results indicated a left medial temporal lobe
epilepsy; however, brain MRI was nonlesional, and Wada testing showed
worse memory on the right side. Given the discordant data and long
duration of epilepsy, intracranial EEG monitoring was considered
necessary, and a responsive neurostimulation device was placed as both a
long-term diagnostic and therapeutic tool.

KEY POINT reduction; by 2 years of bilateral stimulation, seizures were reduced by a median
56%, a 50% responder rate improvement occurred in 54% of patients, seizures
● The efficacy of vagal
nerve stimulation improves
were less severe, and quality of life was improved.47
over time, with 60% of
patients experiencing a
significant response (>50% CONCLUSION
seizure reduction), but the
Drug-resistant epilepsy comprises one-third of all patients with epilepsy.
goal of seizure freedom is
unlikely to be attained with Drug-resistant epilepsy represents a life-threatening disorder. Epilepsy surgery is
this therapy (8% seizure an important and effective tool in patients with drug-resistant epilepsy, and any
freedom after 2 years). patient who has recurrent seizures after adequate trials of two appropriately
chosen AEDs should be considered. Early identification of these patients and
timely referral to a comprehensive epilepsy center are critical. A thorough
presurgical evaluation by a multidisciplinary team and modern surgical
diagnostic and therapeutic approaches, including minimally invasive techniques,
can decrease the existing gap between evidence and practice, thus improving
the care of patients with epilepsy.
VIDEO LEGEND
VIDEO 4-1
Stereo-EEG of a 58-year-old woman with
drug-resistant focal cingulate epilepsy. Video
demonstrates rhythmic sharp waves and fast activity
in contacts LN1-LN8, LY1-LY3, LZ21-LZ23, and LT4-LT6,
confirming ictal onset in the left cingulate region.
links.lww.com/CONT/A274
© 2019 American Academy of Neurology
REFERENCES
1 Kwan P, Arzimanoglou A, Berg AT, et al. Definition 7 Mohanraj R, Brodie MJ. Early predictors of outcome
of drug resistant epilepsy: consensus proposal in newly diagnosed epilepsy. Seizure 2013;22(5):
by the ad hoc Task Force of the ILAE Commission 333–344. doi:10.1016/j.seizure.2013.02.002.
on Therapeutic Strategies. Epilepsia 2010;51(6):
8 Haneef Z, Stern J, Dewar S, Engel J Jr. Referral
1069–1077. doi:10.1111/j.1528-1167.2009.02397.x.
pattern for epilepsy surgery after evidence-based
2 Kwan P, Brodie MJ. Early identification of refractory recommendations: a retrospective study. Neurology
epilepsy. N Engl J Med 2000;342(5):314–319. 2010;75(8):699–704. doi:10.1212/
doi:10.1056/NEJM200002033420503. WNL.0b013e3181eee457.
3 Wiebe S, Bellhouse DR, Fallahay C, Eliasziw M. 9 de Flon P, Kumlien E, Reuterwall C, Mattsson P.
Burden of epilepsy: the Ontario Health Survey. Empirical evidence of underutilization of referrals
Can J Neurol Sci 1999;26(4):263–270. for epilepsy surgery evaluation. Eur J Neurol 2010;
17(4):619–625. doi:10.1111/j.1468-1331.2009.02891.x.
4 Ficker DM, So EL, Shen WK, et al. Population-based
study of the incidence of sudden unexplained 10 Picot MC, Jaussent A, Neveu D, et al. Cost-
death in epilepsy. Neurology 1998;51(5): effectiveness analysis of epilepsy surgery in a
1270–1274. doi:10.1212/WNL.51.5.1270. controlled cohort of adult patients with intractable
partial epilepsy: a 5-year follow-up study. Epilepsia
5 Banerjee PN, Filippi D, Allen Hauser W. The
2016;57(10):1669–1679. doi:10.1111/epi.13492.
descriptive epidemiology of epilepsy-a review.
Epilepsy Res 2009;85(1):31–45. doi:10.1016/ 11 Langfitt JT, Holloway RG, McDermott MP, et al.
j.eplepsyres.2009.03.003. Health care costs decline after successful
epilepsy surgery. Neurology 2007;68(16):1290–1698.
6 Lamberink HJ, Otte WM, Geerts AT, et al.
doi:10.1212/01.wnl.0000259550.87773.3d.
Individualised prediction model of seizure
recurrence and long-term outcomes after 12 Jette N, Quan H, Tellez-Zenteno JF, et al.
withdrawal of antiepileptic drugs in seizure-free Development of an online tool to determine
patients: a systematic review and individual appropriateness for an epilepsy surgery evaluation.
participant data meta-analysis. Lancet Neurol 2017; Neurology 2012;79(11):1084–1093. doi:10.1212/
16(7):523–531. doi:10.1016/S1474-4422(17)30114-X. WNL.0b013e3182698c4c.
378 APRIL 2019

13 Téllez-Zenteno JF, Dhar R, Wiebe S. Long-term 27 Jetté N, Sander JW, Keezer MR. Surgical treatment
seizure outcomes following epilepsy surgery: a for epilepsy: the potential gap between evidence
systematic review and meta-analysis. Brain 2005; and practice. Lancet Neurol 2016;15(9):982–994.
128(pt 5):1188–1198. doi:10.1093/brain/awh449. doi:10.1016/S1474-4422(16)30127-2.
14 Sherman EM, Wiebe S, Fay-McClymont TB, et al. 28 Loring DW, Kapur R, Meador KJ, Morrell MJ.
Neuropsychological outcomes after epilepsy Differential neuropsychological outcomes
surgery: systematic review and pooled estimates. following targeted responsive neurostimulation
Epilepsia 2011;52(5):857–869. doi:10.1111/j.1528-1167. for partial-onset epilepsy. Epilepsia 2015;56(11):
2011.03022.x. 1836–1844. doi:10.1111/epi.13191.
15 Macrodimitris S, Sherman EM, Forde S, et al. 29 Drane DL, Loring DW, Voets NL, et al. Better object
Psychiatric outcomes of epilepsy surgery: a recognition and naming outcome with MRI-guided
systematic review. Epilepsia 2011;52(5):880–890. stereotactic laser amygdalohippocampotomy
doi:10.1111/j.1528-1167.2011.03014.x. for temporal lobe epilepsy. Epilepsia 2015;56(1):
101–113. doi:10.1111/epi.12860.
16 Spencer SS, Berg AT, Vickrey BG, et al. Health-
related quality of life over time since resective 30 Dulay MF, Busch RM. Prediction of
epilepsy surgery. Ann Neurol 2007;62(4):327–334. neuropsychological outcome after resection
doi:10.1002/ana.21131. of temporal and extratemporal seizure foci.
Neurosurg Focus 2012;32(3):E4. doi:10.3171/
17 Hamiwka L, Macrodimitris S, Tellez-Zenteno JF,
2012.1.FOCUS11340.
et al. Social outcomes after temporal or
extratemporal epilepsy surgery: a systematic 31 French JA, White HS, Klitgaard H, et al. Development
review. Epilepsia 2011;52(5):870–879. doi:10.1111/ of new treatment approaches for epilepsy:
j.1528-1167.2011.03074.x. unmet needs and opportunities. Epilepsia
2013;54(suppl 4):3–12. doi:10.1111/epi.12294.
18 Hader WJ, Tellez-Zenteno J, Metcalfe A, et al.
Complications of epilepsy surgery: a systematic 32 Brodie MJ. Road to refractory epilepsy: the
review of focal surgical resections and invasive Glasgow story. Epilepsia 2013;54(suppl 2):5–8.
EEG monitoring. Epilepsia 2013;54(5):840–847. doi:10.1111/epi.12175.
doi:10.1111/epi.12161.
33 Engel J Jr, Wiebe S, French J, et al. Practice
19 Sperling MR, Harris A, Nei M, et al. Mortality after parameter: temporal lobe and localized neocortical
epilepsy surgery. Epilepsia 2005;46(suppl 11): resections for epilepsy. Epilepsia 2003;44(6):
49–53. doi:10.1111/j.1528-1167.2005.00410.x. 741–751. doi:10.1046/j.1528-1157.2003.48202.x.
20 Wiebe S, Blume WT, Girvin JP, et al. A randomized, 34 Bingaman WE. Surgery for focal cortical dysplasia.
controlled trial of surgery for temporal-lobe Neurology 2004;62(6 suppl 3):S30–S34.
epilepsy. N Engl J Med 2001;345(5):311–318. doi:10.1212/01.WNL.0000114508.31261.E6.
doi:10.1056/NEJM200108023450501.
35 Salamon N, Kung J, Shaw SJ, et al. FDG-PET/MRI
21 Engel J Jr, McDermott MP, Wiebe S, et al. Early coregistration improves detection of cortical
surgical therapy for drug-resistant temporal lobe dysplasia in patients with epilepsy. Neurology
epilepsy: a randomized trial. JAMA 2012;307(9): 2008;71(20):1594–1601. doi:10.1212/01.wnl.
922–930. doi:10.1001/jama.2012.220. 0000334752.41807.2f.
22 Dwivedi R, Ramanujam B, Chandra PS, et al. Surgery 36 Lüders HO, Najm I, Nair D, et al. The epileptogenic
for drug-resistant epilepsy in children. N Engl J Med zone: general principles. Epileptic Disord 2006;
2017;377(17):1639–1647. doi:10.1056/NEJMoa1615335. 8(suppl 2):S1–S9.
23 Sperling MR, Barshow S, Nei M, Asadi-Pooya AA. 37 Jobst BC, Kapur R, Barkley GL, et al. Brain-
A reappraisal of mortality after epilepsy surgery. responsive neurostimulation in patients with
Neurology 2016;86(21):1938–1944. doi:10.1212/ medically intractable seizures arising from elo-
WNL.0000000000002700. quent and other neocortical areas. Epilepsia
2017;58(6):1005–1014. doi:10.1111/epi.13739.
24 Engel J Jr, Wiebe S, French J, et al. Practice
parameter: temporal lobe and localized neocortical 38 Englot DJ, Rolston JD, Wright CW, et al. Rates and
resections for epilepsy: report of the Quality predictors of seizure freedom with vagus nerve
Standards Subcommittee of the American Academy stimulation for intractable epilepsy. Neurosurgery
of Neurology, in association with the American 2016;79(3):345–353. doi:10.1227/NEU.
Epilepsy Society and the American Association 0000000000001165.
of Neurological Surgeons. Neurology 2003;60(4):
39 Ramos-Perdigués S, Baillés E, Mané A, et al. A
538–547. doi:10.1046/j.1528-1157.2003.48202.x.
prospective study contrasting the psychiatric
25 Engel J Jr. What can we do for people with drug- outcome in drug-resistant epilepsy between
resistant epilepsy? The 2016 Wartenberg Lecture. patients who underwent surgery and a control
Neurology 2016;87(23):2483–2489. group. Epilepsia 2016;57(10):1680–1690. doi:10.1111/
doi:10.1212/WNL.0000000000003407. epi.13497.
26 Berg AT, Langfitt J, Shinnar S, et al. How long 40 Chan AY, Rolston JD, Rao VR, Chang EF. Effect of
does it take for partial epilepsy to become neurostimulation on cognition and mood in
intractable? Neurology 2003;60(2):186–190. refractory epilepsy. Epilepsia Open 2018;3(1):18–29.
doi:10.1212/01.WNL.0000031792.89992.EC. doi:10.1002/epi4.12100.

41 McDonald TJW, Cervenka MC. Ketogenic diets for 45 Willie JT, Laxpati NG, Drane DL, et al. Real-time
adults with highly refractory epilepsy. Epilepsy magnetic resonance-guided stereotactic laser
Curr 2017;17(6):346–350. doi:10.5698/1535-7597. amygdalohippocampotomy for mesial temporal
17.6.346. lobe epilepsy. Neurosurgery 2014;74(6):569–84;
discussion 584–585. doi:10.1227/NEU.
42 Podkorytova I, Hoes K, Lega B. Stereo-
0000000000000343.
encephalography versus subdural electrodes for
seizure localization. Neurosurg Clin N Am 2016; 46 Hersh DS, Eisenberg HM. Current and future uses
27(1):97–109. doi:10.1016/j.nec.2015.08.008. of transcranial focused ultrasound in neurosurgery.
J Neurosurg Sci 2018;62(2):203–213. doi:10.23736/
43 Mullin JP, Sexton D, Al-Omar S, et al. Outcomes
S0390-5616.17.04230-8.
of subdural grid electrode monitoring in the
stereoelectroencephalography era. World 47 Fisher R, Salanova V, Witt T, et al. Electrical
Neurosurg 2016;89:255–258. doi:10.1016/j.wneu. stimulation of the anterior nucleus of thalamus
2016.02.034. for treatment of refractory epilepsy. Epilepsia
2010;51(5):899–908. doi:10.1111/j.1528-1167.
44 Barnett HJ, Meldrum HE, Eliasziw M, North
2010.02536.x.
American Symptomatic Carotid Endarterectomy
Trial (NASCET) collaborators. The appropriate
use of carotid endarterectomy. CMAJ 2002;
166(9):1169–1179.
380 APRIL 2019

Antiepileptic Drug REVIEW ARTICLE

Treatment of Epilepsy C O N T I N U UM A U D I O
ONLINE
in Children
By Ahsan N. V. Moosa, MD
ABSTRACT
PURPOSE OF REVIEW: The treatment of epilepsy in children is highly
individualized at each and every major step in the management. This review
examines various factors that modify the treatment from the point of
initiation of therapy to the decision to stop an antiepileptic drug (AED).
RECENT FINDINGS:AED therapy leads to seizure freedom in about 70% of all

children with epilepsy. AED initiation could be delayed until a second
seizure in most children and may be avoided altogether in many children
with self-limited childhood focal epilepsies. Three key factors influence
the choice of AED: seizure type(s), efficacy of the drug for the seizure type,
and the side effect profile of the drug(s). For epileptic spasms, steroids
and vigabatrin are the most effective treatment options. For absence
seizures, ethosuximide and valproic acid are superior to lamotrigine.
For focal seizures, many newer AEDs have favorable side effect profiles
with efficacy comparable to older-generation drugs. For generalized
epilepsies, valproic acid remains the most effective drug for a broad range
of seizure types. Genetic and metabolic etiologies may guide unique
treatment choices in some children. After 2 years or more of seizure
freedom, if the recurrence risk after AED withdrawal is acceptable, slow
weaning of AEDs should be done over the span of 6 weeks or longer.
After discontinuation, about 70% of patients remain seizure free, and of
those with recurrence, the majority achieve seizure control with restarting CITE AS:
an AED. When treatment with two or more AEDs fails, other treatment CONTINUUM (MINNEAP MINN)
opportunities for drug-resistant epilepsy, including epilepsy surgery, 2019;25(2, EPILEPSY):381–407.
vagal nerve stimulation, and dietary therapies should be considered.

Dr Ahsan N. V. Moosa, Epilepsy
SUMMARY: Carefully selected medical therapy guided by seizure type and Center, Cleveland Clinic, 9500
AED characteristics is effective in more than two-thirds of children Euclid Ave, Desk S-51, Cleveland,
OH 44195, naduvia@ccf.org.
with epilepsy.
Dr Moosa reports no disclosure.
INTRODUCTION UNLABELED USE OF
M
ore than two-thirds of children with newly diagnosed USE DISCLOSURE:
epilepsy respond to antiepileptic drug (AED) treatment and Dr Moosa reports no disclosure.
enter long-term remission.1,2 Lack of response to the initial
two appropriately selected AEDs predicts poor response © 2019 American Academy
to other drugs.3 Hence, choosing the initial AED is of of Neurology.

ANTIEPILEPTIC DRUG TREATMENT OF EPILEPSY IN CHILDREN
paramount importance. Seizure type and epilepsy phenotype are the

primary drivers of antiepileptic therapy selection. In adults, the dichotomous
classification of seizures into focal and generalized is valuable in guiding
AED choices. However, in young children, the expression of epilepsy has
certain age-dependent characteristics with unique seizure types and syndromes,
requiring therapies seldom used in adults. Additionally, the seizure burden
can be very high in many children with refractory epilepsies such as
Lennox-Gastaut syndrome in which multiple seizures per day are common.
Treatment of such children often requires polytherapy that may cause
cognitive, sedative, and behavioral side effects that adversely affect quality
of life even more than seizures. A carefully selected treatment regimen
implemented systematically may assist in minimizing seizures and, more
importantly, minimizing adverse effects of medication, improving the overall
quality of life. On the other end of the spectrum, certain childhood epilepsies
have a benign long-term prognosis and may not need long-term AEDs.
This article will address common scenarios in pediatric epilepsy practice,
including when to start an AED, which factors determine the AED selection, and
when to consider stopping AED therapy.
INITIATING ANTIEPILEPTIC DRUG THERAPY

The decision to initiate an AED primarily depends on the risk of seizure
recurrence. A recurrence risk of 60% or more is often considered the cutoff
for initiating an AED. A clinical diagnosis of epilepsy is met when a child has two
or more unprovoked seizures or if the predicted risk of recurrence of seizure is
60% or more after the first seizure.4 Factors linked to increased recurrence after a
first seizure include the presence of an abnormal neurologic examination,
abnormal brain MRI, abnormal EEG, and a history of nocturnal seizures.5 The
presence of any one of these factors elevates the risk of seizure recurrence by
twofold in many adult patients. Abnormal EEG and seizures related to remote
brain injury predict increased risk for recurrence in children.6,7
Naturally, the question about AED initiation arises after the first seizure.5
Early initiation of AED treatment after the first seizure has been shown to
delay the second seizure, although the long-term prognosis is unchanged
irrespective of whether treatment is started early or delayed until a second
seizure. Delaying a second seizure by immediately starting AEDs may be
clinically meaningful for many adults who wish to regain and retain their
ability to drive as soon as possible; this consideration is clearly not relevant for
young children.
Three key factors alter the approach to the initiation of treatment for epilepsy
in children.
Benign Natural History of Some Epilepsies in Children

In selected children with self-limited childhood focal epilepsies (with
centrotemporal or occipital spikes), long-term daily AED treatment may not be
needed because children invariably outgrow their tendency to have seizures.6
Because seizures are frequently focal and nocturnal, many families are
comfortable not initiating a daily AED, even after several recurrent seizures.
Daytime seizures, longer seizures, generalized tonic-clonic seizures, and
sometimes caregiver preferences may necessitate the need for initiation
of treatment.
382 APRIL 2019

Cognitive-Behavioral Side Effects of Antiepileptic Drugs KEY POINTS
AEDs may have a potentially negative impact on cognition and behavior in
● A clinical diagnosis of
children. The effect on an individual child is largely unpredictable, and data from epilepsy is met when a child
newer AEDs in children are lacking. Cognitive adverse effects are of utmost has two or more unprovoked
importance to parents and influence the decision to initiate medication. seizures or if the predicted
risk of recurrence of seizure
is 60% or more after the
Young Children Do Not Drive
first seizure.
Because of the absence of driving expectations in young children, daily AEDs are
rarely, if ever, started after the first seizure.7 In clinical settings with a very ● In selected children with
high risk of seizure recurrence, such as in tuberous sclerosis, malformation of self-limited childhood
cortical development, and acute encephalitis, early initiation of an AED after the focal epilepsies (with
centrotemporal or occipital
first seizure may be considered. In most other situations, AED initiation could spikes), long-term daily
be delayed until the occurrence of a second seizure. antiepileptic drug treatment
Initiation of treatment before the onset of epilepsy has been debated in may not be needed because
selected diseases known to have high risk of developing epilepsy. For example, children invariably outgrow
their tendency to have
children with tuberous sclerosis are at high risk of developing infantile spasms in seizures.
the first year of life, and this is linked to poor cognitive outcome. If a child is
diagnosed with tuberous sclerosis early in life before the occurrence of seizures, ● The single most important
initiation of treatment before the onset of seizures has been proposed by some determinant of treatment
choice is the type of seizure
experts. European experts have proposed serial EEGs in infancy and early
that is being targeted.
treatment (with vigabatrin) when epileptiform abnormalities emerge on EEG.8
There is limited evidence to support this approach, especially because of the
neurotoxic risks related to vigabatrin therapy. However, if the EEG shows
features to suggest early hypsarrhythmia, then one could consider therapy even
if overt spasms have not occurred.
SELECTING AN ANTIEPILEPTIC DRUG

More than 25 different antiepileptic medications are available for clinical use. In
the past 2 decades, several newer medications have become available, which, in
general, have fewer side effects and minimal drug interactions.9 Selection of the
first AED requires a careful consideration of several factors, with a focus on
efficacy and tolerability. Factors guiding the choice of an AED include the
epilepsy syndrome, patient characteristics, AED pharmacology, and
socioeconomic/cultural factors, as shown in TABLE 5-1.10
Epilepsy Syndrome Diagnosis

The single most important determinant of treatment choice is the patient’s
epilepsy syndrome diagnosis, especially the range of seizure types being targeted.
Most seizure types are recognizable with careful history taking supported by
other data from the neurologic examination, EEG, and neuroimaging. The
dichotomous classification of seizures into focal and generalized seizure types is
still helpful in guiding the treatment choice. Some patients may have seizures
with both focal and generalized features, and in others, a clear distinction
between focal and generalized seizures may not be possible in the early stages
with only a few seizures to analyze. Two seizure types—epileptic spasms and
absence seizures—typically seen in children have unique treatment options,
discussed later in this review. Both are easily recognizable by history and
interictal EEG; when in doubt, a brief video-EEG monitoring session to confirm
the seizure type may be rewarding, as illustrated in CASE 5-1. Children with
epileptic spasms and absences typically have frequent seizures, which enables

capturing on video-EEG with shorter periods of monitoring. Myoclonic and tonic

seizures are other seizure types that need careful AED selection to ensure efficacy
and avoid exacerbation or worsening of seizure frequency or severity.
Antiepileptic Drug Pharmacology

Some AEDs are considered broad spectrum because they are effective against a
wide range of seizure types. Some drugs are known to be ineffective for certain
seizure types and, in rare instances, may worsen specific seizure types. The side
TABLE 5-1 Factors Determining the Selection of Antiepileptic Drugsa
Epilepsy Characteristics
◆ Seizure type
◆ Epilepsy syndrome
◆ Seizure frequency
◆ Disease
Patient Characteristics
◆ Age
◆ Sex
◆ Pregnancy
◆ Comorbidity
◆ Previous medications
◆ Other concurrent medications
◆ Allergies
◆ Ethnicity/genomics
Antiepileptic Drug Characteristics
◆ Efficacy spectrum
◆ Side effects
◆ Antiepileptic drug interactions
◆ Rapid/slow titration
◆ Half-life and dosing frequency requirements (eg, once a day versus 2 times daily)
◆ Forms (pill/liquid)
◆ Interaction with oral contraceptives
◆ Teratogenic risk
Others
◆ Cost
◆ Insurance
◆ Availability
◆ Personal choice
a
Reprinted with permission from Cross JH, et al, Front Neurol.10 © 2017 Cross, Auvin, Falip, Striano and
Arzimanoglou.
384 APRIL 2019

effect profile of the drugs is the next major determinant. Drugs with a low risk of KEY POINTS
drowsiness and cognitive dulling are preferred. Mechanism of action is
● Drugs with low risk of
particularly important when selecting drugs for polytherapy because drugs with drowsiness and cognitive
a similar mechanism of action may cause synergistic adverse effects. Other dulling are preferred.
factors that influence AED choice include the half-life of the drug (determining
the number of doses per day), ability to titrate rapidly (the need for slower ● A higher risk of
Stevens-Johnson syndrome
titration with lamotrigine precludes its use in children with frequent seizures and
in patients of Asian ancestry
in the intensive care unit setting), interaction with other AEDs and with oral with the HLA-B*1502 allele
contraceptives, and availability in IV and oral liquid formulations. The and a risk of liver failure with
cross-allergenic potential between drugs also may need to be considered in valproic acid in patients with
patients with prior allergies. a POLG1 mutation are two
noteworthy situations in
which pharmacogenomics
Patient Characteristics factors influence
Patient variables unrelated to epilepsy phenotype shape the choice of therapy, as antiepileptic drug selection.
shown in TABLE 5-2. Age and sex have major influences on the choice of AED. In
newborns, phenobarbital has long been used as the first-line agent. Special
challenges related to epilepsy in women of childbearing potential are addressed
elsewhere; refer to the article “Treatment of Women With Epilepsy” by Mona
Sazgar, MD, FAES, in this issue of Continuum.11 Comorbidities such as liver
disease, renal dysfunction, behavioral problems, obesity, and comedications (eg,
warfarin) may influence AED choice. A higher risk of Stevens-Johnson
syndrome in patients of Asian ancestry with HLA-B*1502 allele and a risk of liver
failure with valproic acid in patients with a POLG1 mutation are two noteworthy
situations in which pharmacogenomic factors influence AED selection.12
SOCIOECONOMIC AND PERSONAL PREFERENCES. Although every effort should

be made to prescribe the most effective medication with the fewest side effects,
other personal and socioeconomic factors should also be taken into account. The
cost of the medication, insurance coverage, and the copay costs may influence
drug selection for many families. Risk aversion to side effects of medications
varies among individuals. Although some families of children with epileptic
spasms may object to injectable steroids, others may view the risk of irreversible
peripheral visual field loss with vigabatrin as unacceptable. Availability of the
medication also may be a limiting factor.
EVIDENCE-BASED CHOICE OF ANTIEPILEPTIC DRUGS IN FOCAL AND

GENERALIZED EPILEPSIES
Until the mid-1990s, only a handful of AEDs was available for routine use. Since
then, about 20 newer medications have become available for clinical use. Newer
drugs are initially tested in adults (typically for focal epilepsies) and are typically
approved for clinical use if they show more than a 50% reduction in seizures at 12
to 16 weeks of therapy. Randomized placebo-controlled trials are critical to
establishing the efficacy of a new AED because of the natural fluctuations in
seizure frequency. A review by the International League Against Epilepsy on
evidence for initial monotherapy provided a summary of the level of evidence in
specific epilepsy syndromes, as shown in TABLE 5-3.13 This review also highlights
flaws in many trials and reiterates that some AEDs not listed in the table may be
useful based on clinical experience, case series, or by expert consensus. Although
placebo-controlled trials are crucial for testing new medications, practicing
clinicians often look for information that compares the efficacy of a new drug to

old established drugs to identify opportunities to improve seizure control.

Well-conducted randomized controlled trials providing such information in
children are lacking for several AEDs.
Head-to-head comparison studies between newer and older AEDs are not
undertaken frequently. Findings from SANAD (Standard and New Antiepileptic
Drugs) trials (A and B) established good efficacy of lamotrigine for focal epilepsy
compared with carbamazepine; valproic acid, an older AED, remains the most
effective drug for generalized seizures.14,15 Network meta-analysis of several
CASE 5-1 A 4-year-old boy was referred for management of refractory epilepsy
that started at 18 months of age. His seizures were described as episodes
of confused behavior and walking aimlessly with intermittent, sudden
simultaneous movements of “arms up and head down” occurring every
few seconds to a minute, about 10 to 15 times in a cluster. He had four to
five such clusters a day. His language abilities regressed after the onset of
seizures. He had persistent seizures despite treatment trials with
levetiracetam, valproic acid, topiramate, zonisamide, clobazam,
felbamate, ethosuximide, and ketogenic diet. His prior evaluation at age
2 years had shown a normal brain MRI and multifocal and generalized
epileptiform discharges on EEG. Epilepsy gene panel and CSF analysis for
metabolic disorders were negative.
Video-EEG evaluation at 4 years of age showed features consistent
with hypsarrhythmia, as shown in FIGURE 5-1. The seizures were confirmed
to be epileptic spasms with diffuse ictal patterns. Brain MRI showed
features of subtle cortical dysplasia centered on the right middle frontal
gyrus. This subtle lesion that had not been evident at a younger age likely
became apparent later as myelination was more complete by age 4 years,
providing better gray-white contrast. The child was started on oral
prednisolone for the treatment of epileptic spasms. His seizures
completely stopped within 1 week. Follow-up EEG showed resolution of
hypsarhythmia but showed focal right frontal sharp waves. The child
continued to be seizure free at the 2-year follow-up and showed steady
progress in development and learning.
COMMENT This case highlights the importance of choosing the appropriate treatment
based on the seizure type, which was steroids for epileptic spasms in this
4-year-old child, whose combined features of epileptic spasms and
developmental regression are diagnostic of West syndrome. West
syndrome is commonly associated with an underlying malformation of
cortical development, as seen in this patient. Epileptic spasms may start or
persist in older children. Even in patients with focal malformations,
appropriate medical therapy (steroids or vigabatrin or both) may lead to
remission of epileptic encephalopathy. This child is at risk of focal seizures
in later life, but the resolution of epileptic encephalopathy was critical
for improving the developmental outcome. If the treatment with steroids
had failed, one would consider vigabatrin before considering surgical
treatment.
386 APRIL 2019

trials of various AEDs may provide additional information about comparative
efficacy. Network meta-analysis allows researchers to “combine direct evidence
and indirect evidence across the network of drugs”16 from many trials. In a recent
network meta-analysis, 10 AEDs (carbamazepine, lamotrigine, levetiracetam,
oxcarbazepine, valproic acid, zonisamide, topiramate, phenytoin, phenobarbital,
and gabapentin) commonly used as monotherapy in epilepsies (both partial
and generalized epilepsies) were studied.16 Seventy-seven trials were included
in this analysis, of which individual participant data were available for 69% of
FIGURE 5-1
EEG and MRI findings of the patient in CASE 5-1. A, Interictal EEG shows features of
hypsarrhythmia. B, Ictal EEG shows burst of diffuse slowing (black arrow) concurrent with
spasms followed by attenuation of background activity. C, Axial T2-weighted brain MRI
shows signal abnormalities (white arrows) in the right middle frontal gyrus consistent with
cortical dysplasia.

study patients (n = 12,391) from 36 trials. The trials included both children and
adults with focal seizures and generalized tonic-clonic seizures. Patients with
only absence seizures were excluded. The included trials had at least one pairwise
comparison with two other AEDs in the network. The primary outcome used was
the time to withdrawal of allocated treatment (retention time), which reflects
both effectiveness and tolerability, relevant to clinical decision making. The time
to a second seizure and 6-month and 12-month remission rates were also analyzed.
The results of this network meta-analysis showed that, for focal seizures,
lamotrigine and levetiracetam were significantly better than carbamazepine,
which was better than phenytoin and phenobarbital. For generalized seizures,
TABLE 5-2 Patient Characteristics to Consider in Antiepileptic Drug Selection
Patient Factors Antiepileptic Drug Choice Modifiers Comments

Young women on oral Many enzyme inducers decrease estrogen Intrauterine devices preferred; lamotrigine and
contraceptives levels: eg, phenobarbital, phenytoin, estrogen have a two-way interaction, decreasing
carbamazepine, felbamate, perampanel, the levels of each other
oxcarbazepine, topiramate, and lamotrigine
Pregnant/planning to Valproic acid carries a high (20% to 25%) risk of Risk of neural tube defects with many AEDs is
become pregnant cognitive and behavioral problems reduced by periconceptual folic acid
Obesity Topiramate, zonisamide, and felbamate may Weight gain: valproic acid, carbamazepine,
cause weight loss vigabatrin, and gabapentin
Weight neutral: lamotrigine, phenytoin, and
levetiracetam
Comorbid migraines Topiramate and valproic acid may be effective Targeting seizure type with the right AED should
for migraines supersede the urge to “dual” treatment
Comorbid depression Valproic acid, lamotrigine, and carbamazepine Some AEDs may worsen depression. The US Food
are mood stabilizers and Drug Administration (FDA) includes a warning
for all AEDs about increased suicidal risk based on
an analysis of 11 AEDs in 2008 that found a nearly
twofold increased risk for suicidal behavior or
ideation in patients treated with AEDs compared
to placebo.
Comedication with Enzyme inducers (phenobarbital, phenytoin, In general, many newer AEDs with some
warfarin carbamazepine, perampanel) may reduce exceptions are safe to use with warfarin
effectiveness
Behavioral problems/ Many AEDs are reported to cause psychosis, eg, Neuropsychiatric side effects may occur with any
psychosis levetiracetam, phenobarbital, topiramate, AED; large individual variability in susceptibility
perampanel, vigabatrin, and zonisamide
Kidney stones Topiramate, zonisamide, and ketogenic diet Potassium citrate supplementation is effective in
carry a minor risk of kidney stones preventing renal stones while on ketogenic diet
Liver disease Avoid hepatotoxic drugs: pharmacokinetic Frequent serum level monitoring may be needed
principles guide dosing to titrate; check free levels when appropriate
Renal disease Avoid nephrotoxic drugs: pharmacokinetic Caution about various dialysis types and their effect
principles guide dosing on AED removal; follow serum drug levels closely
AED = antiepileptic drug.
388 APRIL 2019

Level of Evidence for Efficacy of Antiepileptic Drugs for Each Seizure Type TABLE 5-3
and Childhood Epilepsy Syndromea
Seizure Type or Epilepsy Class I Class II Class III Level of Evidence for Efficacy for Effectiveness
Syndrome Studiesb Studiesb Studiesb (Drug Names in Alphabetical Order)c
Children with partial onset 1 0 19 Level A: Oxcarbazepine

seizures
Level B: None
Level C: Carbamazepine, phenobarbital,
phenytoin, topiramate, valproic acid, and
vigabatrin
Level D: Clobazam, clonazepam, lamotrigine, and
zonisamide
Self-limited epilepsy with 0 0 3 Level A: None

centrotemporal spikes
Level B: None
Level C: Carbamazepine and valproic acid
Level D: Gabapentin, levetiracetam,
oxcarbazepine, and sulthiame
Children with absence 1 0 7 Level A: Ethosuximide and valproic acid

seizures
Level B: None
Level C: Lamotrigine
Level D: None
Children with generalized 0 0 14 Level A: None

tonic-clonic seizures
Level B: None
Level C: Carbamazepine, phenobarbital,
phenytoin, topiramate, and valproic acid
Level D: Oxcarbazepine
Juvenile myoclonic epilepsy 0 0 1 Level A: None

Level B: None
Level C: None
Level D: Topiramate and valproic acid
a
Modified with permission from Glauser T, et al, Epilepsia.13 © 2013 John Wiley and Sons.
b
Classification of evidence of antiepileptic drug efficacy13 includes the following classes. Class I evidence: A prospective randomized controlled
trial or meta-analysis of randomized controlled trials in a representative population that meets all the following six criteria: treatment duration
of >48 weeks, double blind, efficacy as primary outcome, not forced to exit study by predetermined number of emergent seizures, demonstrated
superiority or with effectiveness lower limit (95% confidence interval) above a 20% lower boundary relative to the adequate comparator’s
points estimate of efficacy using a preprotocol study population for noninferiority trials, and appropriate statistical analysis. Class II evidence:
A double blind randomized controlled trial or meta-analysis meeting Class I except for treatment duration between >24 weeks and <48 weeks OR
effectiveness lower limit is between 21% and 30% lower boundary relative to the adequate comparator’s points estimate of efficacy. Class III
evidence: A randomized controlled trial or meta-analysis not meeting the criteria for Class I or Class II category: examples include an open-label
study, a study with forced exit criterion, and noninferiority studies lacking adequate comparator. Class IV evidence: Evidence from
nonrandomized, prospective, controlled, or uncontrolled studies, case series, or expert reports.
c
Level A: Established efficacy of antiepileptic drug as initial monotherapy (>Class I studies OR meta-analysis meeting Class I criteria OR >2
Class II studies). Level B: Probably effective antiepileptic drug as initial monotherapy (one Class II study OR meta-analysis meeting Class II criteria).
Level C: Possibly effective antiepileptic drug as initial monotherapy (>2 Class III double-blind or open-label studies). Level D: Potentially
effective antiepileptic drug as initial monotherapy (one Class III double-blind or open-label study OR >1 Class IV clinical studies OR data from
expert committee reports).

valproic acid was superior to phenobarbital, topiramate, and carbamazepine.

Phenytoin and phenobarbital were shown to delay seizure recurrence, suggesting
early seizure control, but this may occur at the expense of adverse side effects as
evident from lower retention rate compared with newer AEDs.16 Another similar
network meta-analysis of only pediatric patients, from 42 studies (n = 5652) with
22 different AEDs, yielded somewhat similar findings.17 Carbamazepine and
lamotrigine were superior in new-onset focal epilepsies, and levetiracetam and
perampanel may have some advantage in refractory focal epilepsies.
Most studies on newer AEDs are conducted in adults, and the approval for
pediatric patients required additional studies. Collaborative efforts between the
Pediatric Epilepsy Academic Consortium for Extrapolation (PEACE) and the US
Food and Drug Administration (FDA) have paved the way for early approval
of newer AEDs in children.18 Drugs approved for focal seizures in adults
(monotherapy or adjunct therapy) may be approved for children older than
4 years with focal seizures without the need for additional efficacy studies in
children because of similarities in disease profile. Age-targeted pharmacokinetic
TABLE 5-4 Response to Treatment With Steroids and/or Vigabatrin in Children With
Infantile Spasms
Study Outcome Measurea Steroidsb Vigabatrin

First Monotherapy
UKISS (only patients Spasm cessation on days 13 and 14 72% (40/55) 54% (28/52)
without tuberous
sclerosis)23 Sustained spasm control with no relapse until 12–14 months 40% (22/55) 37% (19/52)
of age
PERC24 Cessation of spasms in 2 weeks of therapy, with resolution 49% (74/151) 36% (17/47)
of hypsarrhythmia sustained at 3 months
Second Monotherapyc
UKISS25 Any period of 48-hour freedom from spasms after day 14 74% (14/19) 75% (9/12)
of therapy
PERC study26 Cessation of spasms in 2 weeks of therapy, with resolution of 56% (10/18) 55% (17/31)
hypsarrhythmia sustained at 3 months
Steroids Plus
Study Outcome Measurea Steroids Onlyb Vigabatrin
Combined Therapy Versus Steroids Only
ICISS21 No witnessed spasms between days 14 and 42 57% (108/191) 72% (133/186)
Electroclinical responders (no spasms and EEG resolution of 55% (104/189) 66% (123/186)
hypsarrhythmia)
EEG = electroencephalogram; ICISS = International Collaborative Infantile Spasms Study; PERC = Pediatric Epilepsy Research Consortium;
UKISS = United Kingdom Infantile Spasms Study.
a
Differences in outcome measures may account for response rates; outcome measures reported by PERC and ICISS are more akin to clinical
practice.
b
Steroids group included children with either oral prednisolone or IM adrenocorticotropic hormone.
c
Second monotherapy refers to use of the medication in patients for whom the other drug failed. For example, children on second monotherapy
with steroids for whom vigabatrin had failed.
390 APRIL 2019

studies to determine dosages to achieve levels comparable to adults still need to KEY POINTS
be conducted to achieve the indications. Using this approach, newer AEDs, such
● The results of a network
as eslicarbazepine, brivaracetam, and lacosamide, are now approved for use in meta-analysis showed
children aged 4 years and older with partial onset seizures. that, for focal seizures,
lamotrigine and
TREATMENT APPROACHES IN SELECTED PEDIATRIC levetiracetam were
significantly better than
EPILEPSY PHENOTYPES
carbamazepine, which
In this section, treatment approaches to four epilepsy phenotypes with unique was better than phenytoin
electroclinical characteristics, frequently seen in pediatric epilepsy practice, and phenobarbital.
are discussed.
● The results of a network
meta-analysis showed that,
Epileptic Spasms (West Syndrome) for generalized seizures,
Epileptic spasms typically occur in infants (hence, these were previously called valproic acid was superior to
infantile spasms) but may occur in older children with early or extensive brain phenobarbital, topiramate,
injury. Typical epileptic spasms are characterized by sudden, brief stiffening of and carbamazepine.
the whole body, predominantly the axial muscles, that occur in clusters lasting ● The American Academy
several seconds to minutes. Because epileptic spasms occur in the setting of an of Neurology quality
epileptic encephalopathy, early diagnosis and aggressive treatment are critical to committee recommends
optimize both seizure and cognitive outcomes. Delay in treatment initiation with initiation of therapy within
7 days of the onset of
epileptic spasms has been shown to negatively affect developmental outcome.19
epileptic spasms to improve
The American Academy of Neurology (AAN) quality committee recommends developmental outcome.
initiation of therapy within 7 days of the onset of spasms to improve
developmental outcome.20 Recent data suggest that spasm control is also better ● Steroids (oral prednisone
with early treatment initiation.21 For these reasons, at our and many other or IM adrenocorticotropic
hormone) and vigabatrin are
centers, children with new-onset epileptic spasms are emergently admitted to the mainstays of therapy for
initiate workup and treatment. epileptic spasms.
Steroids (oral prednisone or IM adrenocorticotropic hormone [ACTH]) and
vigabatrin are the mainstays of therapy for epileptic spasms. Vigabatrin has ● Reversible hyperintense
signal abnormalities in white
superior efficacy in epileptic spasms caused by tuberous sclerosis complex. With matter, basal ganglia,
all other etiologies, steroids are preferred. In the United States, most clinicians thalamus, and brainstem
prefer IM ACTH over oral steroids, unless parental choices steer toward oral have been reported in as
steroids. There is insufficient evidence to suggest that oral steroids are equivalent many as 30% of infants
treated with vigabatrin.
to ACTH.21,22 Children for whom one form of treatment fails frequently respond
to other treatment. TABLE 5-4 shows the response rate of steroids and vigabatrin
in various settings: as initial monotherapy, second monotherapy (after failure of
one of the two therapies), and combination therapy.21,23–26 The steroids of choice
(prednisolone versus synthetic ACTH versus natural ACTH), dosage regimen
(low-dose or high-dose ACTH), and duration of treatment vary among centers.
A suggested treatment schedule proposed by the Pediatric Epilepsy Research
Consortium is shown in TABLE 5-5.24
Steroid treatment carries the risk of hypertension, hyperglycemia, weight
gain, and reversible hypertrophic cardiomyopathy; these risks are higher with
high-dose ACTH therapy. Vigabatrin carries the risk of retinal toxicity leading
to irreversible peripheral field constriction. The exact clinical impact of such
effects in this group of children with spasms is unknown because many may not
be able to undergo a visual field examination. Periodic eye examinations may
help in recognizing signs of retinal toxicity. Reversible hyperintense signal
abnormalities in white matter, basal ganglia, thalamus, and brainstem have been
reported in as many as 30% of infants treated with vigabatrin.27 Doses of
vigabatrin higher than 150 mg/kg should be avoided, particularly in young

infants. Every attempt should be made to minimize the duration of vigabatrin

exposure to 1 year or less even in good responders.
In previously healthy children with new-onset epileptic spasms and no
apparent etiology, most clinicians prefer to use ACTH because some children will
develop sustained remission after the initial course of steroid therapy.28 A
positive response to treatment is defined as control of spasms and resolution of
hypsarrhythmia on EEG performed 2 weeks after initiation of steroids.22,24 If
children do not achieve seizure freedom or if hypsarrhythmia persists, vigabatrin
therapy should be considered. Concurrent therapy with another AED, such
as topiramate or zonisamide, is frequently used along with steroids, especially
in cases with a known etiology (such as malformations or ischemic brain
injury), but the usefulness of this practice is not proven. The simultaneous
administration of steroids and vigabatrin improves treatment response
compared with steroids alone (TABLE 5-4). However, with such a strategy,
some children may be exposed to the long-term risks of vigabatrin unnecessarily.
TABLE 5-5 Treatment Guidelines for Infantile Spasms Recommended by the Pediatric
Epilepsy Research Consortiuma
Days Dose
Adrenocorticotropic Hormoneb
1–14 75 U/m2 IM 2 times daily
15–17 30 U/m2 IM in the morningc
18–20 15 U/m2 IM in the morning
21–23 10 U/m2 IM in the morning
24–29 10 U/m2 IM every other morning (3 doses)
Prednisolone
1–14 10 mg orally 4 times dailyc,d
15–19 10 mg orally 3 times daily
20–24 10 mg orally 2 times daily
25–29 10 mg/d orally
Vigabatrine
1–3 50 mg/kg/d, divided 2 times daily
4–6 100 mg/kg/d, divided 2 times daily
>7 150 mg/kg/d, divided 2 times dailyc
IM = intramuscular.
a
Modified with permission from Knupp KG, et al, Ann Neurol.24 © 2016 John Wiley and Sons.
b
Check blood pressure 2 times a day for 2 days, then weekly; check urine for glucose at 48 hours and then
weekly.
c
If there is no clinical response by day 14, consider an alternative treatment.
d
If there is no clinical response after day 7, the dose can be increased to 20 mg 3 times a day. If done so, for
days 15–19, it would be 10 mg 4 times a day and proceed per table from day 20.
e
Side effects (eg, sedation, hypotonia) may necessitate slower titration.
392 APRIL 2019

Children in whom therapy with steroids and vigabatrin fails should be
evaluated for epilepsy surgery. Often in this setting, surgery is guided by the
lesion detected on brain MRI, sometimes supported by electroclinical features
suggesting concordant focality, and brain fludeoxyglucose positron emission
tomography (FDG-PET) findings. Nearly 80% achieve an Engel class I outcome
in lesional surgical series of epileptic spasms in infants with lesional MRI.29
Lesions are often obvious, but smaller and subtle lesions may not be apparent
until follow-up imaging is performed. Ketogenic diet is an option for medically
refractory cases in whom surgery is not an option. Anecdotal reports of
successful seizure control have been reported with other medications, including
clonazepam, clobazam, topiramate, zonisamide, and valproate, but these should
not be accepted as initial therapy.28
Genetic causes are increasingly diagnosed with wider availability of
panel-based gene testing for early childhood epilepsies. Treatable metabolic
causes, such as pyridoxine-dependent seizures, should be considered early, when
the etiology is not obvious. Empiric challenge with pyridoxine may be considered
when awaiting results of genetic/metabolic testing.30
Absence Seizures
Absence seizures are characterized by sudden, brief periods of unawareness and
unresponsiveness variably accompanied by subtle head nods, eye flutter, and
other facial movements. They often occur several times per day. EEG during the
episodes show 3-Hz generalized spike-and-wave complexes in classical cases.
Typical absence seizures occur in the setting of childhood absence epilepsy and
juvenile absence epilepsy syndromes in which the absence seizures are the
primary seizure type. Other seizure types, such as generalized tonic-clonic
seizures and myoclonic jerks, may coexist in other syndromes, and this may
influence the choice of therapy. Atypical absences may occur in children with
epileptic encephalopathies, such as Lennox-Gastaut syndrome and Doose
syndrome. Typical absence seizures compared with atypical absences respond
better to medical therapy.
TABLE 5-6 shows the response rate of ethosuximide, valproic acid, and
lamotrigine, the three most effective AEDs for absence seizures, from two major
Response to Treatment in Children With Absence Seizures TABLE 5-6
Initial Monotherapy (Blinded Study)28,30 Ethosuximide (n = 154) Valproic Acid (n = 146) Lamotrigine (n = 146)
“Freedom from failure” a

16–20 weeks 53% 58% 29%
12 months 45% 44% 21%

b 29
Second Monotherapy (Open Label) Ethosuximide (n = 75) Valproic Acid (n = 78) Lamotrigine (n = 55)
“Freedom from failure”a 16–20 weeks 63% 65% 45%
12 months 57% 49% 36%
a
Failure was defined as persistent absence seizures or occurrence of new generalized tonic-clonic seizures or other seizure types or intolerable
side effects (as defined in the study protocol) at or before the specified end point (eg, 16 weeks or 12 months).
b
Second monotherapy refers to use of the medication in patients for whom one of the other two drugs failed. For example, ethosuximide was
prescribed as a second monotherapy for children after failure of valproic acid or lamotrigine as initial therapy.

KEY POINTS studies.31,32 Both valproate and ethosuximide produce comparable rates of
seizure control, but ethosuximide has a favorable side effect profile, particularly
● Both valproate and
ethosuximide produce
in the behavioral domain. For these reasons, in children with only absence
comparable rates of seizure seizures, ethosuximide is the drug of choice.31,33 Open-label studies on children
control for absence for whom one of the three AEDs failed showed that the success rate with an
seizures, but ethosuximide alternative medication remains as high as with initial therapy, as shown in
has a favorable side effect
TABLE 5-6. In children with a partial response to either medication,
32
profile, particularly in the
behavioral domain. combination therapy with valproate and ethosuximide may be successful.
Lamotrigine, although having a favorable side effect profile, is inferior to
● The focus of management ethosuxmide and valproic acid in seizure control. Other medications reported to
in Lennox-Gastaut have anecdotal success include clonazepam and zonisamide. Ketogenic diet is a
syndrome should be on the
overall quality of life and not promising option for refractory absences. Rare reports of response to amantadine
seizure count per se. exist, and it can be tried in refractory cases.34 In patients with other coexisting
generalized seizure types, valproate and lamotrigine may be better suited; if
● Valproic acid is absence seizures remain poorly controlled in such cases, coadministration of
frequently used as a
first-line medication for
ethosuximide may be needed.
children with Lennox-
Gastaut syndrome because Lennox-Gastaut Syndrome
of its broad spectrum of Lennox-Gastaut syndrome is an electroclinical syndrome characterized by
action against various
drug-resistant epileptic encephalopathy manifesting with multiple seizure types:
seizure types.
motor (tonic, atonic, tonic-clonic, and tonic-atonic) and nonmotor (atypical
absences and nonconvulsive status epilepticus). Generalized slow-spike-wave
complexes at 1.5 Hz to 2.5 Hz are a characteristic EEG finding. Lennox-Gastaut
syndrome constitutes 1% to 4% of all childhood epilepsies and is typically
refractory to medications with only less than 10% achieving seizure freedom.10
Most children with Lennox-Gastaut syndrome develop seizures before 5 years of
age and almost all before 8 years of age. One-third of children with epileptic
spasms (West syndrome phenotype) may evolve to Lennox-Gastaut syndrome.
The etiology is heterogeneous, and structural brain lesions and genetic causes
predominate; about 30% do not have a known etiology. Cognitive delays may
precede or follow the onset of epilepsy, depending on the etiology.
Management of Lennox-Gastaut syndrome is particularly challenging because
of the low likelihood of seizure freedom with available treatment options.
Seizures are frequent, occurring daily in many patients. Prioritizing the goals
and setting expectations early during the course will assist caregivers. “Drop
attacks” (most frequently due to tonic-atonic seizures) causing injuries are a
major factor affecting the quality of life of patients and caregivers. Preventing
injuries with use of a helmet and changes in the living environment with
occupational therapy guidance should be emphasized. Patients almost always
need polytherapy, and many may experience adverse effects of polytherapy,
affecting their quality of life. The focus should be on the overall quality of life
and not seizure count per se. Many caregivers prefer rare or even regular
breakthrough seizures over cognitive and sedating drug side effects.
A suggested treatment algorithm for newly diagnosed Lennox-Gastaut
syndrome, proposed by experts, shown in FIGURE 5-2 illustrates various
therapeutic options.10 Valproic acid is frequently used as a first-line medication
because of its broad spectrum of action against various seizure types. TABLE 5-7
shows the efficacy of various AEDs that have been tested in patients with
Lennox-Gastaut syndrome in double-blind placebo-controlled trials.10,35–40 In all
these trials, patients were on other AED(s), and the study drug was used as
394 APRIL 2019

FIGURE 5-2
A suggested treatment algorithm for a newly diagnosed patient with Lennox-Gastaut
syndrome.
AED = antiepileptic drug.
a
Use for intermittent, short-term treatment of crisis episodes.
b
In combination with valproate or clobazam.
Reprinted with permission from Cross JH, et al, Front Neurol.10 © 2017 Cross, Auvin, Falip, Striano
and Arzimanoglou.
adjunct therapy. A Cochrane review concluded that no drug is superior to

another.41 Overall, lamotrigine, rufinamide, felbamate, clobazam, topiramate,
and cannabidiol offer a greater than 50% reduction in the overall seizure burden
in 26% to 65% of patients. High-dose clobazam has been particularly effective in
reducing drop seizures.39 Long-term follow-up of patients on clobazam suggests
that the effect in responders is sustained in about 85% of patients at 1 and 3 years
of treatment. Of note, despite its popular use in Lennox-Gastaut syndrome, no
controlled trials have been done with valproic acid. An addition to the

armamentarium for treatment of Lennox-Gastaut syndrome occurred in June

2018, with the FDA approval of cannabidiol oral solution for the treatment of
seizures associated with Lennox-Gastaut syndrome and Dravet syndrome in
patients 2 years of age and older.42
Nonpharmacologic treatments include ketogenic diet, vagal nerve stimulation
(VNS), corpus callosotomy, and resective surgery.10 Multiple open-label
(retrospective and prospective) studies have reported more than 50% reduction
in seizures in 40% to 51% of patients treated with ketogenic diet. A dedicated
ketogenic diet team is necessary to implement the diet in this challenging
population. A treatment trial of 3 months on the diet is generally warranted to
TABLE 5-7 Seizure Reduction Rates of Selected Antiepileptic Drugs in Lennox-Gastaut

Syndrome
Reduction in Drop Reduction in Generalized

Reductionb in Overall Seizures Seizures Compared Tonic-Clonic Seizures
Antiepileptic Druga Compared With Placebo With Placebo Compared With Placebo
Felbamate Felbamate 26% reduction versus 5% increase 44% decrease 40% reduction versus 12%
Study Group, 199335 versus 7% decrease increase
Lamotrigine 32% reduction versus 9% increase 34% reduction 36% reduction versus 10%
Motte et al, 199736 versus 9% reduction increase
33% versus 16% had >50% reduction 37% versus 22% had 43% versus 20% had
>50% reduction >50% reduction
Topiramate 20.6% reduction versus 8.8% reduction 15% reduction Not analyzed separately
Sachdeo et al, 199937 versus 5% increase
33% versus 8% for >50% reduction of drop
and tonic-clonic seizures
Rufinamide 33% reduction versus 12% decrease 42.5% reduction 46% reduction versus 18%
Glauser et al, 200838 versus 1.4% increase decrease
Clobazamc High-dose clobazam 65% reduction High-dose Not analyzed separately

Ng et al, 201139 versus 9% reduction clobazam: 68%
reduction versus
12% reduction
78% had >50%

reduction versus
32% with placebo
Cannabidiol 41% reduction versus 14% reduction 44% reduction Not analyzed separately
Thiele et al, 201840 versus 22%
reduction
44% had >50%

reduction versus
20% with placebo
a
All medications were used as adjunctive therapy. Results are not comparable across studies because several differences exist between the
studies, including the seizure types, etiology, and other medications. Valproic acid, the most frequently used medication in Lennox-Gastaut
syndrome, has not been tested against a placebo.
b
Posttreatment reduction of seizures reported at about 12 to 14 weeks of therapy.
c
The clobazam trial included three different dose regimens (0.25 mg/kg/d: low dose; 0.5 mg/kg/d: medium dose; and 1 mg/kg/d: high dose).
There was a linear trend toward increasing dose and increased efficacy in seizure control (not shown in the table). However, higher doses of
clobazam were associated with higher dropout rates, the majority of which were related to adverse effects.
396 APRIL 2019

determine if the diet if efficacious. Corpus callosotomy may be useful in KEY POINT
preventing injurious falls due to tonic-atonic seizures. The seizure reduction with
● Two commonly used
VNS in Lennox-Gastaut syndrome is comparable to its response in other epilepsy treatment options for
types. A pooled analysis of four studies reported a responder (50% or more electrical status epilepticus
seizure reduction) rate of 55% with VNS in Lennox-Gastaut syndrome. VNS is in slow-wave sleep include
less effective for atonic seizures and may be more effective for myoclonic high-dose nightly
benzodiazepines and
jerks.10,41 All patients with structural abnormalities on brain MRI should be
steroids.
carefully evaluated for resective epilepsy surgery. Electroclinical features are
frequently unhelpful for localization of epilepsy with a Lennox-Gastaut
syndrome phenotype.43 Additional testing, such as positron emission
tomography (PET), ictal single-photon emission computed tomography
(SPECT), and magnetoencephalography (MEG), may also be of only minimal
value. Experience of the center is of paramount importance in selecting surgical
candidates, both for including and excluding surgical options.
Electrical Status Epilepticus in Slow-Wave Sleep

Electrical status epilepticus in slow-wave sleep (ESES), also referred as continuous
spikes in slow-wave sleep (CSWS), is a unique epileptic encephalopathy seen in
children, often between the ages of 4 and 8 years, characterized by variable
regression of language or cognitive function with other neurobehavioral
problems secondary to near-continuous epileptiform discharges in slow-wave
sleep. ESES involving the posterior language cortices manifesting with auditory
agnosia and language regression is referred to as Landau-Kleffner syndrome. As
with many other epileptic encephalopathies, the etiology is heterogeneous,
ranging from structural to genetic causes. ESES may also occur in the setting of
self-limited childhood focal epilepsy. For research criteria, ESES is diagnosed
when 85% or more of slow-wave sleep is occupied by spike-and-slow waves. In
patients with clinical dysfunction attributable to ESES, clinicians may start
treatment even with a lower spike burden.44 The treatment decision is not always
straightforward because some children with self-limited childhood focal epilepsy
may have ESES based on EEG criteria without the accompanying cognitive
dysfunction, and such children are best monitored clinically for regression.
Conversely, in children with major chronic cognitive dysfunction at baseline, the
impact of the spike-wave burden on cognition may be unclear.
Treatment of ESES poses several challenges with fewer effective treatment
options.45,46 Evidence-based recommendations for the treatment of ESES are
lacking because of a lack of well-conducted prospective studies in this
heterogeneous epilepsy syndrome with various etiologies. Two commonly
used treatment options for ESES include steroids and high-dose nightly
benzodiazepines. Conventional AEDs may also reduce the spike burden,
and frequently used medications for this purpose include valproic acid,
ethosuximide, levetiracetam, clobazam, and sulthiame.47 Sulthiame is not
available in the United States but is used in European countries. Most clinicians
avoid using drugs such as carbamazepine, oxcarbazepine, phenytoin, and
phenobarbital because they have been reported to worsen ESES.47
From a pooled analysis of 112 studies with a total of 575 patients undergoing
910 treatments for ESES, the EEG and cognitive improvement rates of 282
consecutive patients with 585 treatment responses are shown in TABLE 5-8.46
Among medical therapies, steroids had the highest chance for success. Two
factors are evaluated to monitor response to treatment: spike burden in sleep on

EEG and improvement in cognitive, language, or behavioral symptoms. Children

who were developmentally and cognitively normal before the onset of the ESES
respond favorably to treatment and have a better prognosis. A longer duration of
ESES before treatment also influences the outcome negatively. It is not
uncommon to see improvement in EEG without notable improvement in clinical
symptoms because there may be a lag between EEG resolution and clinical
improvement. A lack of clinical improvement despite sustained resolution of
ESES on EEG may indicate that the behavioral and cognitive problems in such a
patient may be secondary to the underlying etiology per se and not linked to its
EEG expression. In patients who achieve remission, close clinical monitoring and
periodic assessment of EEG in sleep would be needed to monitor the response.
For medically refractory ESES secondary to focal epileptogenic lesions,
epilepsy surgery is often an effective option.46 Frequently, lesions that cause the
ESES phenotype are large with involvement of the thalamus, thereby suggesting
a role for the thalamus in the genesis of ESES syndrome. Children with structural
lesions that are not surgically amenable tend to have the worst outcome. The role
of multiple subpial transection of the posterior perisylvian language/auditory
cortices for the treatment of Landau-Kleffner syndrome remains controversial.48
DISEASE-MODIFYING OR DISEASE-SPECIFIC TREATMENT

In most instances, the treatment of epilepsy has been essentially symptomatic
with an aim to reduce the recurrence of seizures with conventional AEDs. The
natural history of the underlying etiology most often determines if long-term
remission occurs. In few selected etiologies, there may be opportunities for
disease-modifying treatment, such as in autoimmune epilepsies and
genetic/metabolic disorders. The field of autoimmune epilepsy is still in an
evolving phase, and the treatment opportunities will become clearer over the
next decade. In children, metabolic and genetic causes may offer unique
TABLE 5-8 Response Rate of Commonly Used Treatments for Electrical Status
Epilepticus in Slow-Wave Sleep/Continuous Spikes in Slow-Wave Sleep
From a Pooled Analysis of 575 Casesa
Electroencephalogram
Improvement, Clinical Improvement, Any Improvement, Odds Ratio of
Treatment % (95% CI) % (95% CI) % (95% CI) Favorable Response
Conventional antiepileptic 33 (28–38) 32 (26–37) 34 (29–39) Reference (for other

drugs (n = 310) odds ratios below)
Steroids (n = 100) 68 (58–77) 70 (60–79) 75 (67–83) 4.4 (2.9–6.7)
Benzodiazepines (n = 171) 46 (37–56) 45 (35–54) 59 (50–68) 2.2 (1.5–3.2)
Surgery (n = 30) 74 (58–91) 83 (70–97) 93b 9.8 (4.1–23.1)
Others (n = 38) 26 (8–44) 71 (53–89) 58 (42–74) 1.2 (0.9–1.6)
CI = confidence interval.
a
Modified with permission from Van Den Munckhof B, et al, Epilepsia.46 © 2013 John Wiley and Sons.
b
No 95% confidence interval is available because of the small sample size and large proportion of patients who showed improvement.
398 APRIL 2019

therapeutic options targeted to specific diseases or disease pathogenesis, as
shown in TABLE 5-9.49,50
Therapeutic options in specific genetic/metabolic diseases may occur
in several forms. In some instances, supplementation with deficient
cofactors/nutrients leads to better seizure control. Examples include pyridoxine
for pyridoxine-dependent seizures; ketogenic diet in glucose transporter
deficiency, providing alternative fuel when glucose transport to the brain is
defective; and biotin supplementation in biotinidase deficiency. Some of these
syndromes may present as a commonly seen epilepsy syndrome, such as absence
epilepsy, as illustrated in CASE 5-2.49,51
Specific AEDs are known to worsen the epilepsy in certain diseases. Examples
include phenytoin, carbamazepine, and lamotrigine in Dravet syndrome due to
SCNA1 mutation.49,50 Specific AEDs may be selectively more effective in certain
channelopathies. Examples include phenytoin and other sodium channel
Selected List of Genetic Diagnosis Modifying Treatment of Epilepsies TABLE 5-9
Genetic Disease Preferred Treatment Comments
SLC2A1 (glucose Ketogenic diet Glucose transport across the blood-brain barrier defective;
transporter 1 defect) providing ketones and fatty acids as alternative fuel by
ketogenic diet
ALDH7A1 Pyridoxine ± folinic acid Pyridoxine-dependent epilepsy
PNPO Pyridoxal 50 -phosphate Pyridoxal phosphate deficiency
SCN8A Phenytoin, carbamazepine, and Sodium channel blockers useful with “gain of function”
oxcarbazepine mutations; supratherapeutic levels of phenytoin may be
needed
SCN2A Phenytoin, carbamazepine, and Likely useful in case of “gain of function” mutations;
lamotrigine supratherapeutic levels of phenytoin may be needed
KCNQ2 Phenytoin and carbamazepine Retigabine offered some promise but was pulled from
market because of long-term retinal and skin adverse
effects
TSC1 and TSC2 (tuberous Vigabatrin and rapamycin Vigabatrin is effective for epileptic spasms; rapamycin
sclerosis types 1 and 2) inhibits hyperactive mammalian target of rapamycin
pathway and reduces tumor size and seizures
GRIN2A Memantine (N-methyl-D-aspartate Few case reports; likely only in selected mutation
[NMDA]) receptor antagonist)
KCNT1 Quinidine Initial case report reported benefit; not shown to be useful
in a controlled trial
SCN1A Valproic acid, clobazam, Avoid sodium channel blockers such as phenytoin,
topiramate, and stiripentol carbamazepine, and lamotrigine; case series of benefits
with fenfluramine
POLG1 No preferred antiepileptic drug High risk of hepatic failure with valproic acid
treatment

CASE 5-2 A 4-year-old girl was evaluated for management of refractory absence
seizures. Staring spells were noted since the age of 9 months but were
not diagnosed as seizures until about 3 years of age. Brief episodes of
behavioral arrest with unresponsiveness lasting 5 to 10 seconds were
noted several times per day at the time of diagnosis. EEG showed 3 Hz
spike-wave complexes consistent with the diagnosis of absence
epilepsy, as shown in FIGURE 5-3. Seizures decreased but persisted
despite adequate trials of ethosuximide and valproic acid, initially as
monotherapy and later in combination. She had mild gross and fine
motor delays and needed physical and occupational therapy. No major
cognitive difficulties were reported.
A genetic evaluation at age 4 years revealed a heterozygous mutation
c.227G>A (p.Gly76Asp) in the SLC2A1 gene. SCL2A1 gene mutation causes
glucose transporter type 1 (GLUT1) deficiency syndrome, which may
manifest with early-onset absence seizures. The child was started on the
ketogenic diet with a dramatic decrease in seizures.
FIGURE 5-3
EEG of the patient in CASE 5-2 shows generalized 3-Hz spike-wave complexes.
COMMENT This case highlights the importance of identifying the etiology of epilepsy,
which may have unique treatment implications: in this case, ketogenic diet
for GLUT1 deficiency in this child. The SLC2A mutation, more popularly
described as GLUT1 disease, causes defective transport of glucose across
the blood-brain barrier. Low glucose in the central nervous system is
known to cause cognitive delays, seizures, ataxia, paroxysmal dyskinesias,
and eye movement abnormalities. In a series of 34 children with early-onset
(before the age of 4 years) absence epilepsy, four children (12%) had
pathogenic mutations in the SLC2A mutation; these findings were
replicated in another cohort of 55 patients.50 Genetic testing for GLUT1
deficiency should be considered in early-onset absences as well as in
refractory absences in older children as well.
400 APRIL 2019

blockers in channelopathies such as SCN8A, SCN2A, and KCNQ2 mutations. In KEY POINTS
particular, higher doses of phenytoin are effective in SCN8A epilepsy.49,50,52
● Examples of antiepileptic
Unconventional medications targeted toward dysfunctional receptors and or drugs known to worsen
channels and pathways is another approach. Examples include the use of epilepsy in certain diseases
quinidine in KCNT1 mutation, memantine in GRIN2A mutation, and everolimus include phenytoin,
in mammalian target of rapamycin (mTOR) signaling. The response in this group carbamazepine, and
lamotrigine in Dravet
is variable and may be affected by the specific type(s) of mutations. Missense
syndrome due to SCNA1
mutations in different locations in the same gene may have varied effects, with mutation.
one being responsive to a specific medication, whereas another may not be
responsive.50 Clinical evidence has suggested specific medications to be ● A pooled analysis of five
particularly more effective in some disease phenotypes, although the exact trials in children provided
evidence to support a
mechanisms are poorly understood, for example, the use of vigabatrin in minimum 2-year
epileptic spasms secondary to tuberous sclerosis. seizure-free period before
Experience with many of the above-described therapies is limited to relatively considering the need for
few patients. A few hypothetical precision medicine options based on the continued treatment.
pathophysiology of the genetic effects have been proposed. With a better
understanding of the pathogenesis and cellular electrophysiology of genetic
epilepsies, more treatment options are likely to emerge.
DISCONTINUATION OF ANTIEPILEPTIC DRUGS

Once a remission is achieved for a sustained period, the question of duration of
therapy arises, and every attempt should be made to minimize the duration of
AED exposure in children. A pooled analysis of five trials in children provided
evidence to support a minimum 2-year seizure-free period before considering the
need for continued treatment.53 At the time of weaning of AEDs, the concerns of
clinicians, patients, and families center on the risk of recurrence after stopping
AEDs, the risk of evolving intractable seizures after weaning, and the potential
risks of serious harm from recurrent seizures, including status epilepticus,
sudden unexpected death in epilepsy (SUDEP), and injuries or accidents.53
The risk of recurrence should be reassessed periodically, primarily determined
by epilepsy syndrome/etiology and EEG findings. In practice, one may categorize
children whose epilepsy is in remission into three groups, based on their risk
of recurrence, that influence the duration of AED therapy.54,55
In the first group, the natural history of epilepsy is predictably benign, and
chances for remission are very high. Examples include self-limited childhood
focal epilepsies with centrotemporal spikes and Panayiotopoulos syndrome, in
which the majority of children outgrow their tendency to have seizures by
puberty. These constitute 15% to 20% of all children with epilepsies. In this
category, one would consider weaning of AEDs after 2 years of seizure freedom
or even earlier in some cases. Some children with these syndromes may have
persistent interictal abnormalities but may have outgrown the tendency to have
clinical seizures. If seizures were to recur, their long-term outlook is not altered.
A second group consists of children with very high risk of seizure recurrence,
with very low chances for sustained remission. Examples include epileptic
encephalopathies secondary to genetic disorders and epilepsy secondary to major
bilateral malformations. Achieving seizure freedom in such patients is often
difficult. When children are seizure free for a longer period, families and
clinicians may attempt to reduce polypharmacy. EEG in most cases continues
to show epileptiform abnormalities. This group constitutes about 10% to 15%
of all children with epilepsy.

The majority of children with epilepsy belong in the “uncertain” group for
whom the long-term chances for seizure remission without AEDs are not very
predictable. Long-term remission rates may vary between etiologies, ranging
from 70% in childhood absence epilepsy to 15% with juvenile myoclonic epilepsy
(in adulthood).54–56 Many symptomatic focal epilepsies and those in which the
etiology is unknown also belong to this category. The presence of structural
abnormalities does not always indicate the need for long-term therapy. For
example, a child with seizures secondary to stroke may be able to come off
the AED after a period of remission. In a population-based cohort of children
with epilepsy (the study excluded self-limited childhood focal epilepsy with
centrotemporal spikes, previously referred to as benign rolandic epilepsy) 70%
CASE 5-3 A 6-year-old boy with global delay and left hemiparesis was evaluated
for medically intractable epilepsy. His seizure symptomatology
suggested generalized tonic and atypical absence seizures. Between the
ages of 1 and 5 years, he had two phases of seizure freedom for 2 years on
medications. Subsequently, he had subtle seizures with momentary
leaning to one side and eye flutter, along with subtle changes in
awareness and behavior, several times per hour during wakefulness. He
was listless and lethargic for most of the day. Treatment failed with four
antiepileptic drugs (AEDs) (valproate, lamotrigine, levetiracetam, and
clobazam) and a trial of oral prednisolone. A video-EEG study showed
abundant diffuse slow-spike-wave complexes, higher over the right
hemisphere, as shown in FIGURE 5-4. Runs of slow-spike-wave complexes
occupied 60% to 70% of the awake period and 80% to 90% of the
sleep period. Subtle eye flutter, head bobbing, and alterations of
responsiveness occurred several times concurrent with runs of spike-
wave complexes. Focal abnormalities in the right frontocentral region
were also noted. Brain MRI showed extensive right hemispheric dysplasia
(FIGURE 5-4C).
A right disconnective hemispherectomy was performed. The boy
became seizure free after surgery; a mild worsening of left hemiparesis
recovered to the preoperative baseline. Postoperative EEGs showed the
expected findings after the hemispherectomy. At 6 weeks postsurgery,
one of the three AEDs was weaned and stopped. A second AED was
weaned off after 6 months of freedom from seizures. On follow-up, he
became more attentive, interactive, and verbal, partly attributed to
reduction of AEDs. He is maintained on one medication with a plan to
wean after seizure freedom for 1 year after surgery.
COMMENT This case highlights the role of epilepsy surgery in patients with medically
refractory lesional epilepsies. Surgery could have been considered earlier
in this child, but he had periods of “remission” on medical therapy. Such
malformations carry very low chances for sustained long-term seizure
freedom. This case also illustrates that, after successful epilepsy surgery,
AED weaning could be considered early in patients on polytherapy.
402 APRIL 2019

of children who had achieved sustained seizure freedom remained seizure free
after discontinuation of AEDs, and 30% had one or more recurrences.55 The
authors of this study estimated that stopping the AEDs increases the risk of
recurrence twofold, suggesting that continued treatment does not guarantee
against recurrence. Risk factors for increased recurrence include epileptiform
abnormalities on EEG, epilepsy onset before 2 years of age and after 10 years
of age, intellectual disability (IQ <70), history of status epilepticus, and higher
seizure burden before and during treatment.53 The risk of recurrence is high
in the initial 1 to 2 years after discontinuation. In one series, 97% of children who
had recurrence after AED withdrawal achieved remission with reinitiation
of AEDs.55
FIGURE 5-4
EEG and MRI findings of the patient in CASE 5-3. A, Interictal EEG shows generalized
slow-spike-wave complexes (2 Hz) and right frontal sharp waves. B, Ictal EEG shows
bisynchronous spike-wave complexes, higher on the right side, concurrent with clinical
seizures. C, Axial fluid-attenuated inversion recovery (FLAIR) brain MRI shows extensive
right hemispheric dysplasia with volume loss.

There are no systematic data available about the risk of SUDEP and other
injuries related to seizures in patients after AED discontinuation; the risk is
estimated to be very low. Driving needs; ease with which initial control was
achieved; the severity of seizures; and psychological, economic, and cultural
factors further shape the decision making. Some risk-averse patients and families
may choose to continue medications and not “rock the boat” despite a low
predicted risk, whereas another family may decide to wean and stop medications
even with a higher risk. Once the patient and family elect to wean, most clinicians
wean the AED over 6 weeks to 3 months and one drug at a time for patients
receiving polytherapy. A comparison of slow weaning over 9 months versus
relatively faster weaning over 6 weeks showed no significant difference in
seizure recurrence risk.57
In summary, in most children with epilepsy, if the child is seizure free on
medication(s) for 2 years, then AED tapering should be strongly considered
unless the recurrence risk is estimated to be very high. For patients who become
seizure free after epilepsy surgery, withdrawal could be considered at 1 year of
seizure freedom. In patients who are on multiple AEDs, tapering of some AEDs
may be appropriate even as early as a few weeks after a potentially successful
surgery, as shown in CASE 5-3.
RESCUE MEDICATIONS FOR HOME USE TO PREVENT STATUS

EPILEPTICUS
Prescribing non-IV rescue medications for outside hospital use to prevent status
epilepticus should be considered in children at risk of prolonged seizures or acute
repetitive seizures. Rectal diazepam is the only FDA-approved medication for
“bouts of increased seizure activity” and is the most commonly used rescue
medication in the United States.58 The approval for rectal diazepam by the FDA
was based on two randomized trials that showed a decrease in seizures in
children with acute repetitive seizures.59,60 Multiple trials have shown
midazolam is effective for prolonged seizures and acute repetitive seizures
when administered through IM, intranasal, or buccal routes.61–63 In a network
meta-analysis of efficacy of non-IV medication for acute convulsive seizures, IM
and intranasal midazolam had superior efficacy over rectal diazepam. Rectal
diazepam is available as prepackaged syringes in specific dosages. Off-label use of
intranasal midazolam is common in practice. The IV midazolam formulation is
readily available, and this may be used with an appropriate delivery device, such
as an atomizer or syringe, based on the route of administration. Oral clonazepam,
either as liquid or dissolvable wafer, is frequently prescribed, but the evidence to
support its use is limited. Somnolence is a common adverse effect with all rescue
medications, but respiratory depression is rare.59–63 The choice of rescue
medication is guided by the age of the patient, ease of administration, and
personal preferences. For older children and teenagers, families prefer to use
intranasal midazolam over rectal diazepam.
CONCLUSION
An individualized approach tailored to each patient and the family’s goals and
expectations is critical to the successful management of epilepsy in children. For
patients with new-onset epilepsies, specific evidence-based recommendations
404 APRIL 2019

are available for choosing medications primarily based on specific characteristics KEY POINTS
of epilepsy. These decisions are further shaped by patient characteristics, AED
● The majority of children
side effects, and cultural and socioeconomic factors. AED tapering should be with epilepsy belong in the
considered after 2 years or more of seizure freedom. In children with medically “uncertain” group for whom
refractory epilepsy, nonpharmacologic therapies, including epilepsy surgery and the long-term chances for
ketogenic diet, should be considered. Precision medicine opportunities based seizure remission without
antiepileptic drugs are not
on specific genetic diagnosis are emerging and may lead to more treatment
very predictable.
opportunities in many genetic epilepsies in the future. In all children with
epilepsy, general precautions to avoid injuries during seizures should always be ● Pooled analyses of
emphasized, and rescue medications, such as rectal diazepam or intranasal studies in children who had
midazolam, to avoid a prolonged seizure should be considered. achieved sustained seizure
freedom show that 70%
remained seizure free
after discontinuation of
REFERENCES antiepileptic drugs, and
30% had one or more
recurrences.
1 Sillanpää M, Schmidt D. Natural history of treated 9 Golyala A, Kwan P. Drug development for
childhood-onset epilepsy: Prospective, refractory epilepsy: the past 25 years and
long-term population-based study. Brain 2006; beyond. Seizure 2017;44:147–156. doi:10.1016/ ● Risk factors for increased
129(pt 3):617–624. doi:10.1093/brain/awh726. j.seizure.2016.11.022. recurrence include
epileptiform abnormalities
2 Sillanpää M, Saarinen M, Schmidt D. Clinical 10 Cross JH, Auvin S, Falip M, et al. Expert opinion on
on EEG, epilepsy onset
conditions of long-term cure in childhood- the management of Lennox-Gastaut syndrome:
onset epilepsy: a 45-year follow-up study. treatment algorithms and practical before 2 years of age and
Epilepsy Behav 2014;37:49–53. doi:10.1016/ considerations. Front Neurol 2017;8:505. after 10 years of age,
j.yebeh.2014.05.029. doi:10.3389/fneur.2017.00505. intellectual disability
(IQ <70), history of status
3 Kwan P, Brodie MJ. Early identification of 11 Sazgar M. Treatment of women with epilepsy.
epilepticus, and higher
refractory epilepsy. N Engl J Med 2000;345(5): Continuum 2019;25(2, Epilepsy):408–430. doi: 10.1212/
314–319. doi:10.1056/NEJM200002033420503. CON.0000000000000713. seizure burden before and
during treatment.
4 Fisher RS, Acevedo C, Arzimanoglou A, et al. ILAE 12 Balestrini S, Sisodiya SM. Pharmacogenomics in
official report: a practical clinical definition of epilepsy. Neurosci Lett 2018;667:27–39.
● A comparison of slow
epilepsy. Epilepsia 2014;55(4):475–482. doi:10.1111/ doi:10.1016/j.neulet.2017.01.014.
epi.12550.
weaning over 9 months
13 Glauser T, Ben-Menachem E, Bourgeois B, et al. versus relatively faster
5 Krumholz A, Wiebe S, Gronseth GS, et al. Updated ILAE evidence review of antiepileptic drug weaning over 6 weeks
Evidence-based guideline: management of an efficacy and effectiveness as initial monotherapy showed no significant
unprovoked first seizure in adults: report of for epileptic seizures and syndromes. Epilepsia
difference in seizure
the guideline development subcommittee of 2013;54(3):551–563. doi:10.1111/epi.12074.
the American Academy of Neurology and the
recurrence risk.
American Epilepsy Society. Neurology 2015;
SANAD study of effectiveness of valproate,
84(16):1705–1713. doi:10.1212/WNL.
lamotrigine, or topiramate for generalised and
0000000000001487.
unclassifiable epilepsy: an unblinded randomised
6 Bouma PA, Bovenkerk AC, Westendorp RG, controlled trial. Lancet 2007;369(9566):1016–1026.
Brouwer OF. The course of benign partial doi:10.1016/S0140-6736(07)60461-9.
epilepsy of childhood with centrotemporal
spikes: a meta-analysis. Neurology 1997;48(2):
SANAD study of effectiveness of carbamazepine,
430–437. doi:10.1212/WNL.48.2.430.
gabapentin, lamotrigine, oxcarbazepine or
7 Hirtz D, Berg A, Bettis D, et al. Practice parameter: topiramate for treatment of partial epilepsy: an
treatment of the child with a first unprovoked unblinded randomised controlled trial. Lancet
seizure: report of the quality standards 2007;369(9566):1000–1015. doi:10.1016/S0140-
subcommittee of the American Academy of 6736(07)60460-7.
Neurology and the practice committee of the
16 Nevitt SJ, Sudell M, Weston J, et al. Antiepileptic
Child Neurology Society. Neurology 2003;60(2):
drug monotherapy for epilepsy: a network
166–175. doi:10.1212/01.WNL.0000033622.27961.B6.
meta-analysis of individual participant data.
8 Curatolo P, Jóźwiak S, Nabbout R. Management Cochrane Database Syst Rev 2017;6:CD011412.
of epilepsy associated with tuberous sclerosis doi:10.1002/14651858.CD011412.pub2.
complex (TSC): clinical recommendations. Eur J
17 Rosati A, Ilvento L, Lucenteforte E, et al.
Paediatr Neurol 2012;16(6):582–586. doi:10.1016/
Comparative efficacy of antiepileptic drugs
j.ejpn.2012.05.004.
in children and adolescents: a network
meta-analysis. Epilepsia 2018;59(2):297–314.
doi:10.1111/epi.13981.

18 Arzimanoglou A, D'Cruz O, Nordli D, et al. 29 Chugani HT, Ilyas M, Kumar A, et al. Surgical
A review of the new antiepileptic drugs for treatment for refractory epileptic spasms: the
focal-onset seizures in pediatrics: role of Detroit series. Epilepsia 2015;56(12):1941–1949.
extrapolation. Paediatr Drugs 2018;20(3): doi:10.1111/epi.13221.
249–264. doi:10.1007/s40272-018-0286-0.
30 Pellock JM, Hrachovy R, Shinnar S, et al. Infantile
19 Darke K, Edwards SW, Hancock E, et al. spasms: a U.S. consensus report. Epilepsia
Developmental and epilepsy outcomes at age 2010;51(10):2175–2189. doi:10.1111/j.1528-1167.2010.
4 years in the UKISS trial comparing hormonal 02657.x.
treatments to vigabatrin for infantile spasms: a
31 Glauser TA, Cnaan A, Shinnar S, et al.
multi-centre randomised trial. Arch Dis Child
Ethosuximide, valproic acid, and lamotrigine in
2010;95(5):382–386. doi:10.1136/adc.2009.160606.
childhood absence epilepsy. N Engl J Med 2010;
20 Patel AD, Berg AT, Billinghurst L, et al. Quality 362(9):790–799. doi:10.1056/NEJMoa0902014.
improvement in neurology: child neurology
32 Cnaan A, Shinnar S, Arya R, et al. Second
quality measure set. Neurology 2018;90(2):67–73.
monotherapy in childhood absence epilepsy.
doi:10.1212/WNL.0000000000004806.
Neurology 2017;88(2):182–190. doi:10.1212/
21 O’Callaghan FJ, Edwards SW, Alber FD, et al. WNL.0000000000003480.
Safety and effectiveness of hormonal treatment
versus hormonal treatment with vigabatrin for
infantile spasms (ICISS): a randomised,
childhood absence epilepsy: initial monotherapy
multicentre, open-label trial. Lancet Neurol 2017;
outcomes at 12 months. Epilepsia 2013;54(1):
16(1):33–42. doi:10.1016/S1474-4422(16)30294-0.
141–155. doi:10.1111/epi.12028.
22 Go CY, Mackay MT, Weiss SK, et al.
34 Perry MS, Bailey LJ, Kotecha AC, et al.
Evidence-based guideline update: medical
Amantadine for the treatment of refractory
treatment of infantile spasms. Report of the
absence seizures in children. Pediatr Neurol 2012;
guideline development subcommittee of the
46(4):243–245. doi:10.1016/j.pediatrneurol.
American Academy of Neurology and the
2012.02.004.
practice committee of the Child Neurology
Society. Neurology. 2012;78(24):1974–1980. 35 Felbamate Study Group in Lennox-Gastaut
doi:10.1212/WNL.0b013e318259e2cf. Syndrome. Efficacy of felbamate in childhood
epileptic encephalopathy (Lennox-Gastaut
23 Lux AL, Edwards SW, Hancock E, et al. The United
syndrome). N Engl J Med 1993;328(1):29–33.
Kingdom Infantile Spasms Study comparing
doi:10.1056/NEJM199301073280105.
vigabatrin with prednisolone or tetracosactide at
14 days: a multicentre, randomised controlled 36 Motte J, Trevathan E, Arvidsson JF, et al.
trial. Lancet 2004;364(9447):1773–1778. Lamotrigine for generalized seizures associated
doi:10.1016/S0140-6736(04)17400-X. with the Lennox-Gastaut syndrome. Lamictal
Lennox-Gastaut Study Group. N Engl J Med
24 Knupp KG, Coryell J, Nickels KC, et al. Response
1997;337(25):1807–1812. doi:10.1056/
to treatment in a prospective national infantile
NEJM199712183372504.
spasms cohort. Ann Neurol 2016;79(3):475–484.
doi:10.1002/ana.24594. 37 Sachdeo RC, Glauser TA, Ritter F, et al. A double-
blind, randomized trial of topiramate in Lennox-
25 Lux AL, Edwards SW, Hancock E, et al. The United
Gastaut syndrome. topiramate YL study group.
Kingdom Infantile Spasms Study (UKISS)
Neurology. 1999;52(9):1882–1887. doi:10.1212/
comparing hormone treatment with vigabatrin on
WNL.52.9.1882.
developmental and epilepsy outcomes to age
14 months: a multicentre randomised trial. 38 Glauser T, Kluger G, Sachdeo R, et al. Rufinamide
Lancet Neurol 2005;4(11):712–717. doi:10.1016/ for generalized seizures associated with
S1474-4422(05)70199-X. Lennox-Gastaut syndrome. Neurology 2008;
70(21):1950–1959. doi:10.1212/01.wnl.0000303813.
26 Knupp KG, Leister E, Coryell J, et al. Response to
95800.0d.
second treatment after initial failed treatment in
a multicenter prospective infantile spasms cohort. 39 Ng YT, Conry JA, Drummond R, et al. Randomized,
Epilepsia 2016;57(11):1834–1842. doi:10.1111/epi.13557. phase III study results of clobazam in Lennox-
Gastaut syndrome. Neurology 2011;77(15):
27 Pearl PL, Vezina LG, Saneto RP, et al. Cerebral
1473–1481. doi:10.1212/WNL.0b013e318232de76.
MRI abnormalities associated with vigabatrin
therapy. Epilepsia 2009;50(2):184–194. doi:10.1111/ 40 Thiele EA, Marsh ED, French JA, et al. Cannabidiol
j.1528-1167.2008.01728.x. in patients with seizures associated with Lennox-
Gastaut syndrome (GWPCARE4): a randomised,
28 Wilmshurst JM, Gaillard WD, Vinayan KP, et al.
double-blind, placebo-controlled phase 3 trial.
Summary of recommendations for the
Lancet 2018;391(10125):1085–1096. doi:10.1016/
management of infantile seizures: task force
S0140-6736(18)30136-3.
report for the ILAE Commission of Pediatrics.
Epilepsia 2015;56(8):1185–1197. doi:10.1111/epi.13057. 41 Hancock EC, Cross JH. Treatment of Lennox-
Gastaut syndrome. Cochrane Database Syst Rev
2013;(2):CD003277. doi:10.1002/14651858.
CD003277.pub3.
406 APRIL 2019

42 Drug Enforcement Administration, Department 53 Strozzi I, Norlan SJ, Sperling MR, et al. Early versus
of Justice. Schedules of controlled substances: late antiepileptic drug withdrawal for people
placement in schedule V of certain FDA- with epilepsy in remission. Cochrane Database
approved drugs containing cannabidiol; Syst Rev 2015;(2):CD001902. doi:10.1002/
corresponding change to permit requirements. 14651858.CD001902.pub2.
Final order. Fed Regist 2018;83(189):48950–48953.
54 Camfield P, Camfield C. When is it safe to
43 Wyllie E, Lachhwani DK, Gupta A, et al. Successful discontinue AED treatment? Epilepsia 2008;
surgery for epilepsy due to early brain lesions 49(suppl 9):25–28. doi:10.1111/j.1528-1167.2008.
despite generalized EEG findings. Neurology 01923.x.
2007;69(4):389–397. doi:10.1212/01.wnl.
55 Camfield P, Camfield C. The frequency of
0000266386.55715.3f.
intractable seizures after stopping AEDs in
44 Sánchez Fernández I, Chapman KE, Peters JM, seizure-free children with epilepsy. Neurology
et al. Continuous spikes and waves during sleep: 2005;64(6):973–975. doi:10.1212/01.WNL.
electroclinical presentation and suggestions for 0000154517.82748.A7.
management. Epilepsy Res Treat 2013;2013:
56 Vorderwülbecke BJ, Kowski AB, Kirschbaum A,
583531. doi:10.1155/2013/583531.
et al. Long-term outcome in adolescent-onset
45 Sánchez Fernández I, Chapman K, Peters JM, generalized genetic epilepsies. Epilepsia 2017;
et al. Treatment for continuous spikes and waves 58(7):1244–1250. doi:10.1111/epi.13761.
during sleep (CSWS): Survey on treatment
57 Ranganathan LN, Ramaratnam S. Rapid versus
choices in North America. Epilepsia 2014;55(7):
slow withdrawal of antiepileptic drugs. Cochrane
1099–1108. doi:10.1111/epi.12678.
Database Syst Rev 2006;(2):CD005003.
46 Van Den Munckhof B, Van Dee V, Sagi L, et al. doi:10.1002/14651858.CD005003.pub2.
Treatment of electrical status epilepticus in
58 Food and Drug Administration. Diazepam
sleep: a pooled analysis of 575 cases. Epilepsia
rectal gel. accessdata.fda.gov/drugsatfda_
2015;56(11):1738–1746. doi:10.1111/epi.13128.
docs/label/2016/020648s014lbl.pdf. Updated
47 Caraballo RH, Veggiotti P, Kaltenmeier MC, et al. December 2016. Accessed February 4, 2019.
Encephalopathy with status epilepticus during
59 Dreifuss FE, Rosman NP, Cloyd JC, et al.
sleep or continuous spikes and waves during
A comparison of rectal diazepam gel and
slow sleep syndrome: a multicenter, long-term
placebo for acute repetitive seizures. N Engl J
follow-up study of 117 patients. Epilepsy Res 2013;
Med 1998;338(26):1869–1875. doi:10.1056/
105(1–2):164–173. doi:10.1016/j.eplepsyres.2013.02.010.
NEJM199806253382602.
48 Downes M, Greenaway R, Clark M, et al.
60 Cereghino JJ, Mitchell WG, Murphy J, et al.
Outcome following multiple subpial transection
Treating repetitive seizures with rectal diazepam
in Landau-Kleffner syndrome and related
formulation: a randomized study. The North
regression. Epilepsia 2015;56(11):1760–1766.
American Diastat Study Group. Neurology 1998;
doi:10.1111/epi.13132.
51(5):1274–1282.
49 EpiPM Consortium. A roadmap for precision
61 Silbergleit R, Durkalski V, Lowenstein D, et al.
medicine in the epilepsies. Lancet Neurol
Intramuscular versus intravenous therapy for
2015;14(12):1219–1228. doi:10.1016/S1474-
prehospital status epilepticus. N Engl J Med 2012;
4422(15)00199-4.
366(7):591–600. doi:10.1056/NEJMoa1107494.
50 Reif PS, Tsai MH, Helbig I, et al. Precision
62 Bhattacharyya M, Kalra V, Gulati S. Intranasal
medicine in genetic epilepsies: break of dawn?
midazolam vs rectal diazepam in acute childhood
Expert Rev Neurother 2017;17(4):381–392.
seizures. Pedatr Neurol 2006;34(5):355–359.
doi:10.1080/14737175.2017.1253476.
doi:10.1016/j.pediatrneurol.2005.09.006.
51 Suls A, Mullen SA, Weber YG, et al. Early-onset
63 McIntyre J, Robertson S, Norris E, et al. Safety
absence epilepsy caused by mutations in the
and efficacy of buccal midazolam versus rectal
glucose transporter GLUT1. Ann Neurol 2009;
diazepam for emergency treatment of seizures in
66(3):415–419. doi:10.1002/ana.21724.
children: a randomised controlled trial. Lancet
52 Barker BS, Ottolini M, Wagnon JL, et al. The 2005;366(9481):205–210. doi:10.1016/S0140-
SCN8A encephalopathy mutation p.Ile1327Val 6736(05)66909-7.
displays elevated sensitivity to the anticonvulsant
phenytoin. Epilepsia 2016;57(9):1458–1466.
doi:10.1111/epi.13461.

REVIEW ARTICLE
Treatment of Women
With Epilepsy

ONLINE
By Mona Sazgar, MD, FAES
ABSTRACT
PURPOSE OF REVIEW: This article provides the latest information to guide
practitioners in counseling and treating women with epilepsy.
RECENT FINDINGS: There is an increasing body of literature on the

multidirectional effects of sex hormones on seizure frequency and severity
and of seizures altering areas of the brain involved in neuroendocrine
CITE AS: function. Ongoing pregnancy outcome data from pregnancy registries and
CONTINUUM (MINNEAP MINN) meta-analysis of observational studies have provided key information on
2019;25(2, EPILEPSY):408–430.
the safety of using antiseizure medications during pregnancy and the risk to
the fetus.
Dr Mona Sazgar, Department of
Neurology, University of SUMMARY:In treating and counseling women with epilepsy from puberty to
California Irvine Health,
Comprehensive Epilepsy menopause, it is important to understand the complex interactions of sex
Program, 101 The City Dr S, hormones, seizures, and antiseizure medications on reproductive health
Pavilion I, Ste 123, Orange, CA
and pregnancy outcomes.
92868-3201, msazgar@uci.edu.
Dr Sazgar has received personal
compensation for serving on the INTRODUCTION
E
scientific advisory board for UCB
SA and for serving on the
pilepsy is estimated to affect 70 million people worldwide1,2 and is one
speaker’s bureau of Eisai Co, Ltd; of the most common neurologic diseases globally. Nearly 1.5 million
Sunovion Pharmaceuticals Inc; women of childbearing age in the United States live with epilepsy,
and UCB SA. Dr Sazgar has
received grants or research and each year approximately 24,000 give birth. Specific challenges
support from Biogen, Sunovion arise in caring for women with epilepsy throughout their life cycle
Pharmaceuticals Inc, and UCB SA from menarche to menopause. Important considerations in the care of women
as the principal investigator for
clinical trials and has received with epilepsy include whether antiepileptic drugs (AEDs) interact with
publishing royalties from contraceptives; how AEDs impact fertility, teratogenic risk, and libido; whether
Springer for her book
Controversies in Caring for
seizures and AED concentrations change during pregnancy, postpartum, and
Women with Epilepsy. lactation; and the risk of osteoporosis associated with AEDs and epilepsy.
UNLABELED USE OF
CATAMENIAL EPILEPSY
USE DISCLOSURE: The term catamenial is derived from the Greek word katamenios, meaning
Dr Sazgar discusses the “monthly.” In 1857, Sir Charles Locock first described that some seizures are
unlabeled/investigational use of
acetazolamide, clobazam, associated with the menstrual cycle.3,4 William Gowers in 1881 reported that the
depot-medroxyprogesterone majority of women with epilepsy in his practice had reported a worsening of their
acetate, and progesterone
seizures perimenstrually.5,6 Women with catamenial epilepsy have a cyclic
products for the treatment of
women with catamenial epilepsy. exacerbation of their seizures at certain points in their menstrual cycle that are
attributed to fluctuations in their sex hormones.7,8 Estrogen and progesterone
variations throughout the menstrual cycle and their neuroactive effects on brain
of Neurology. structures with epileptic potential are thought to be the principal behind
408 APRIL 2019

catamenial epilepsy. This condition is reported in at least one-third of women KEY POINTS
with epilepsy. Herzog and colleagues9,10 proposed existence of three patterns
● Nearly 1.5 million women
of catamenial seizure exacerbation (FIGURE 6-1 and TABLE 6-1): perimenstrual of childbearing age in
(C1: days –3 to 3) and periovulatory (C2: days 10 to –13) in ovulatory cycles and the United States live
the entire luteal phase (C3: day 10 of one cycle to day 3 of the next cycle) in with epilepsy.
anovulatory cycles, where day 1 is the first day of menstrual flow and day 14 is
● Catamenial epilepsy is
the day of ovulation. The diagnosis is made by using seizure diaries and charting
reported in at least
the time of ovulation (by using basal body temperature or ovulation kits) and one-third of women
menstruation for three cycles. If most seizures (twofold or higher) occur during with epilepsy.
one of the above periods, a diagnosis of catamenial epilepsy is made.
● Women with catamenial
epilepsy have a cyclic
MECHANISM OF CATAMENIAL EPILEPSY exacerbation of their
Estrogen and progesterone have neuroactive properties, meaning that they can seizures with the fluctuation
alter neuronal excitability and seizure susceptibility by their effect on of their hormones.
brain structures. As a rule, estrogens are proconvulsant and progesterone
● As a rule, estrogens are
anticonvulsant. Progesterone’s
proconvulsant and
reduced metabolite, progesterone is
allopregnanolone, is a potent anticonvulsant.
positive allosteric modulator of
γ-aminobutyric acid type A
(GABA-A) neurotransmission,11
which may mediate
anticonvulsant potential. In
rodent models of catamenial
epilepsy developed by Reddy
and colleagues,12 withdrawal
of progesterone or
allopregnanolone, mimicking
the premenstrual state,
enhanced neuronal excitability
and seizures.12
Evidence exists for moderate
proconvulsant activity of
estrogen-containing compounds.
Ethinyl estradiol significantly
increased seizure frequency and
severity in kindled female mice.13
Estradiol may potentiate seizures
by increasing the density of
dendritic spines and N-methyl- FIGURE 6-1
Patterns of catamenial epilepsy. A, The normal
D-aspartate (NMDA)
menstrual cycle. B, The inadequate luteal phase
receptor–containing excitatory cycle. The three patterns of catamenial epilepsy are
synapses in the hippocampus. (1) perimenstrual (C1, day –3 to 3), (2) periovulatory
Estradiol may also potentiate (C2, day 10 to –13) for the normal ovulatory cycles,
and (3) the entire second half of the cycle for
non-NMDA kainate-mediated
inadequate luteal phase (C3, day 10 to 3). Day 1 is
and quisqualate-mediated the first day of menstrual flow, and day 14 is the day
neurotransmission,14–16 which of ovulation.
may explain the increase in F = follicular phase; L = luteal phase; M = menstruation;
O = ovulation.
seizure risk with exposure to Reprinted with permission from Herzog AG, Seizure.10
oral contraceptives reported by © 2007 British Epilepsy Association.

TREATMENT OF WOMEN WITH EPILEPSY
Herzog and colleagues.17–19 This large, community-based survey found that 28.2%
of women with epilepsy reported a change in seizures when using hormonal
contraception, with the relative risk of seizure being 4.5 times higher in women
using hormonal contraception than in women using nonhormonal contraception.
Estrogen effects on seizure exacerbation are complex. The dose, route of
administration, acute versus chronic administration, and natural hormonal
environment may change its proconvulsant potential.20
Finally, antimüllerian hormone, another neuroactive peptide that is a measure
of ovarian reserve and follicular activity and a member of the transforming
growth factor β family, has been implicated as having seizure-protective
properties. Women with epilepsy who have active seizures have a lower
concentration of antimüllerian hormone compared with those without
seizures and healthy controls,21 and the hormone protects neurons against
NMDA-mediated neurotoxicity both in vitro and in vivo.22,23 The antimüllerian
hormone receptor is expressed in several brain regions that are also involved in
epileptic networks, including the hippocampus, cortex, and hypothalamus.
Further studies are needed to establish whether antimüllerian hormone is
protective against seizures in women with epilepsy.24
THE HYPOTHALAMIC-PITUITARY-OVARIAN AXIS AND EPILEPSY

Brain regulation of sex hormones occurs via the hypothalamic-pituitary-
ovarian axis (FIGURE 6-2). Gonadotropin-releasing hormone produced by the
hypothalamus stimulates the production of follicle-stimulating hormone and
luteinizing hormone for secretion by the anterior pituitary gland. Follicle-
stimulating hormone stimulates the growth of ovarian follicles, which, in turn,
secrete estradiol (the main form of estrogen). The dominant ovarian follicle
develops into the oocyte. This is called the follicular phase (days 1 to 14).
Estrogen production results in negative feedback, reducing follicle-stimulating
hormone but stimulating gonadotropin-releasing hormone. Gonadotropin-
releasing hormone stimulation leads to a luteinizing hormone surge, further
oocyte maturation, ovulation, and the conversion of the follicle to corpus luteum,
signaling the end of the follicular phase and the start of the luteal phase
TABLE 6-1 Patterns of Catamenial Epilepsya
Perimenstrual or C1 Pattern
◆ Significant increase in average seizure frequency around day –3 to day 3 of the menstrual
cycle compared with the rest of the menstrual cycle
Periovulatory or C2 Pattern
◆ Significant increase in average seizure frequency during ovulation around days 10 to –13
compared with the rest of the menstrual cycle
Inadequate Luteal Phase or C3 Pattern
◆ Significant increase in average seizure frequency between day 10 of one cycle and day 3 of
the next cycle, which includes ovulatory, luteal, and menstrual phases in women with an
inadequate and anovulatory luteal phase
a
Data from Herzog AG, Seizure.10
410 APRIL 2019

(days 15 to 28). Corpus luteum
secretes progesterone, which
inhibits gonadotropin-
releasing hormone, luteinizing
hormone, and follicle-
stimulating hormone production.
In the absence of pregnancy,
corpus luteum regresses, and
both progesterone and estradiol
decline, resulting in
menstruation. The cycle
repeats as lower levels of
progesterone decrease
gonadotropin-releasing
hormone inhibition.
The hypothalamic-pituitary-
ovarian axis regulation is
affected by the abnormal
neurophysiology of seizures, FIGURE 6-2
and the hypothalamic-pituitary- Brain regulation of sex hormones through the
hypothalamic-ovarian-pituitary axis.
ovarian–associated hormones are Gonadotropin-releasing hormone (GnRH)
affected by medications produced by the hypothalamus stimulates the
used to treat seizures in women production of follicle-stimulating hormone (FSH)
with epilepsy. Ictal and interictal and luteinizing hormone (LH) by the anterior
pituitary gland. FSH stimulates the growth of
discharges can disrupt the ovarian follicles. Estrogen production results in
normal activity of brain negative feedback, reducing FSH but stimulating
structures, including the GnRH. GnRH stimulation leads to LH surge,
limbic system, amygdala, ovulation, and conversion of the follicle to corpus
luteum. Corpus luteum secretes progesterone,
hypothalamus, and pituitary
which inhibits GnRH, LH, and FSH production.
gland.25 Hepatic enzyme–
inducing AEDs, specifically
cytochrome P450 3A4 (CYP3A4) inducers, affect the metabolism of endogenous
sex hormones and thyroid hormones and, therefore, contribute to the
dysregulation of the hypothalamic-pituitary-ovarian axis.10
TREATMENT OF CATAMENIAL EPILEPSY

No specific treatment is approved by the US Food and Drug Administration
(FDA) for catamenial epilepsy. The treatment is usually divided into hormonal
and nonhormonal. Acetazolamide, which has been in use for more than 50 years,
is one of the oldest treatment options for catamenial epilepsy. There have not
been any randomized clinical trials to prove the efficacy of acetazolamide in
treating catamenial epilepsy.26,27
Benzodiazepines, such as clonazepam and clobazam, are used in the treatment
of seizure clusters during hormonally related exacerbation of seizures.
Benzodiazepines are positive allosteric modulators of the GABA-A receptor, and
of the benzodiazepines, clobazam has a particularly broad spectrum of efficacy
against a variety of seizure types. In a double-blind, placebo-controlled crossover
study, clobazam resulted in complete control in most women during the
10-day trial period.28,29 In this study, clobazam was effective when used at a
dose of 20 mg/d to 30 mg/d, administered intermittently starting 2 to 4 days

before menses. The most common adverse effects of clobazam are sedation
and depression.
Certain antiseizure medication doses can be temporarily increased during the
catamenial seizure exacerbation period. This approach may not be safe with
some antiseizure medications, such as phenytoin and carbamazepine.
Synthetic progestin depot-medroxyprogesterone acetate at a dose of 150 mg
every 3 months has been used for reducing seizure exacerbation in catamenial
epilepsy. Reductions in seizure frequency of up to 39% over a 1-year period have
been reported.30,31 A risk of osteoporosis occurs with the prolonged use of
depot-medroxyprogesterone acetate, and the delay to conception with
depot-medroxyprogesterone acetate may last up to 1 year.
A National Institutes of Health–sponsored randomized, double-blind,
placebo-controlled Phase 3 multicenter clinical trial by Herzog and colleagues32
assessed the response to treatment with natural progesterone lozenges in women
with medically refractory catamenial partial epilepsy. Patients were randomly
assigned 2:1 to progesterone or placebo. Overall results were negative for a
beneficial effect, but a post hoc analysis showed a significantly higher responder
rate in women with perimenstrual seizure exacerbation (C1). Progesterone may
provide a clinically important benefit for this subset of women with perimenstrual
TABLE 6-2 Suggested Treatment Options in Women With Catamenial Epilepsy
1 Determine True Catamenial Epilepsy

A Establish whether the seizures are, in fact, catamenial in nature by using seizure diaries.
Ask the patient to chart daily the seizure type and frequency with simultaneous recording
of ovulation and menstruation status using an ovulation kit or basal body temperature
recording for three menses.
B Determine whether there is an increase in the number and severity of seizures by twofold
or greater during the specific days of the patient’s menstrual cycle and establish C1, C2, or
C3 type of catamenial epilepsy.
2 Choose One of the Options Below
A Progesterone lozenges/natural progesterone for C1 pattern
For the C1 type, consider using progesterone lozenges 200 mg 3 times daily around the
days of seizure exacerbation or days 14 to 28 of the cycle.
B Synthetic progestin
Consider oral daily synthetic progestin or intrauterine devices with progestin versus
depot- medroxyprogesterone acetate.
C Acetazolamide
Consider using at 250 mg twice daily or 500 mg twice daily to be taken around the
7–10 days of seizure exacerbation as determined by the seizure diary.
D Clobazam
20 milligrams to 30 mg divided twice a day or one dose at night for 10 days, starting 2 days
before and throughout the identified seizure exacerbation dates.
E Small increase in baseline antiepileptic drugs
These can be taken approximately 2 days before the identified period of seizure
exacerbation for up to 10 days. Be cautious about phenytoin, carbamazepine, or other
medications with a higher risk of toxicity.
412 APRIL 2019

catamenial epilepsy, which is the most prevalent form (TABLE 6-1). TABLE 6-2 KEY POINT
shows a suggested treatment algorithm for catamenial epilepsy.
● Menstrual disorders are
estimated to occur in
REPRODUCTIVE DISORDERS one-third of women with
Reproductive disorders in women with epilepsy may result from direct effects of epilepsy as compared with
seizures or secondary to use of certain antiepileptic drugs. 12% to 14% of women in the
general population.
Direct Effects of Seizures

A higher rate of reproductive disorders, including polycystic ovary syndrome,
infertility, and decreased libido, occurs in women with epilepsy. Menstrual
disorders are estimated to occur in one-third of women with epilepsy compared
with 12% to 14% of women in the general population.33 Polycystic ovary syndrome
is characterized by enlarged ovaries with multiple small cysts and a
hypervascularized, androgen-secreting stroma leading to the associated signs
of androgen excess (hirsutism, alopecia, acne), obesity, and menstrual cycle
disturbance (oligomenorrhea or amenorrhea).34 It occurs in 4% to 7% of women
of reproductive age in the general population but in 10% to 25% of women with
epilepsy.35,36 The amygdala has reciprocal connections with the hypothalamus,
and seizures originating from the mesial temporal lobe can disrupt the
gonadotropin-releasing hormone-producing cells in the preoptic area of the
hypothalamus, resulting in abnormal luteinizing hormone and follicle-stimulating
hormone levels and, therefore, abnormal levels of sex hormones and sexual
dysfunction. Herzog and colleagues37 studied 50 consecutive women with
temporal lobe epilepsy and found that 56% of them reported amenorrhea,
oligomenorrhea, or abnormally long or short menstrual cycle intervals, and 68%
showed reproductive endocrine disorders, such as polycystic ovary syndrome,
hypoandrogenism, premature menopause, and hyperprolactinemia.
The fertility data in women with epilepsy are conflicting. The Rochester
population study of residents for the years 1935 to 1974 found the fertility rate was
reduced to 85% in women with epilepsy.38 Another population study in Iceland
found no difference in birth rates of women with epilepsy compared with
sex-matched residents without epilepsy except in patients with epilepsy with
mental retardation and cerebral palsy.39 More recently, Pennell and colleagues40
conducted an observational cohort study that compared the fertility rate in
89 women with epilepsy without previously known infertility with 108 control
women. They found no difference in achieving pregnancy, sexual activity,
ovulatory rates, time to pregnancy, and live birth rates among the two groups.
Fertility is affected by multiple factors, including the social and psychological
state and frequency of unprotected intercourse.
Antiepileptic Drug Effects

AEDs are known to cause endocrine side effects resulting in abnormalities in
fertility, thyroid hormones, sexual function, and bone health. Microsomal
hepatic enzyme–inducing AEDs, such as phenytoin, carbamazepine, and
phenobarbital, can reduce the circulating bioavailable steroid hormones and,
therefore, increase sex hormone–binding globulin concentrations.41 Valproic
acid is also known to cause endocrine side effects. In 1993, Isojärvi conducted the
first systematic study of 238 women with epilepsy receiving valproic acid
monotherapy. Approximately 45% of these women had menstrual disorders, and
of those, 90% had polycystic ovary syndrome or hyperandrogenism, or both.42

The risk of developing polycystic ovary syndrome and high testosterone levels
was age dependent and highest in women 26 years old and younger. In a
prospective study, an increase in serum testosterone and androstenedione levels
was seen in half of the women within 3 months of starting valproic acid
therapy.41 The endocrine adverse effects of valproic acid are at least partly
reversible. Valproic acid–induced weight gain may exacerbate its
endocrine effects.
Sexual dysfunction and decreased libido and satisfaction with sex lives are
also reported in women taking enzyme-inducing AEDs, such as carbamazepine
and phenytoin.43 Mattson and colleagues44 reported sexual dysfunction and
decreased libido in 16% of patients with epilepsy after starting monotherapy
TABLE 6-3 Antiepileptic Drugs and Contraceptive Failure
Antiepileptic Drugs Causing Contraceptive Failure

◆ Carbamazepine
◆ Clobazam
◆ Eslicarbazepine acetate
◆ Oxcarbazepine
◆ Phenobarbital
◆ Phenytoin
◆ Primidone
◆ Rufinamide
Antiepileptic Drugs Causing Contraceptive Failure at Higher Doses
◆ Felbamate
◆ Perampanel
◆ Topiramate
Antiepileptic Drugs With No Known Effect on Contraceptive Failure
◆ Clonazepam
◆ Ethosuximide
◆ Gabapentin
◆ Lacosamide
◆ Lamotrigine
◆ Levetiracetama
◆ Retigabine/ezogabine
◆ Tiagabine
◆ Valproateb
◆ Vigabatrin
◆ Zonisamide
a
Increased testosterone concentrations reported in men on levetiracetam.
b
Decreased free testosterone concentrations in men and increased androgen concentrations in women
taking valproate.
414 APRIL 2019

with phenobarbital, 12% after starting phenytoin, and 13% after starting KEY POINTS
carbamazepine.
● The oral contraceptive
The full endocrine side effect profile of newer AEDs is unknown. Lamotrigine failure rate is 1% in healthy
is not found to cause endocrine side effects. In fact, switching from valproic acid women but 3% to 6% in
to lamotrigine resulted in normalization of endocrine function after a year. the population of women
Levetiracetam can reduce basal estrogen secretion from ovarian follicles, but it with epilepsy.
does not affect the gonadotropin-stimulated estrogen secretion.45 To date, no
● Neurologists should be
clinically significant reproductive endocrine side effects have been associated aware of the complex and
with the use of levetiracetam in women or prepubertal children with epilepsy.46 at times bidirectional
However, there have not been any randomized studies assessing the potential interactions between
endocrine side effects of other newer antiseizure medications. antiepileptic drugs and
hormonal contraception.
CONTRACEPTION
A variety of hormonal contraception methods are available, including oral
contraceptive tablets, topical patches, IM depot injections, implants, and
intrauterine devices. Combined oral contraceptives contain both synthetic
estrogen and progestin. Contraceptives work with inhibition of ovulation or
fertilization. The progestin component is responsible for the contraceptive effect
of combined oral contraceptives, including inhibition of ovulation, increased
viscosity of the cervical mucus, and reduced endometrial suitability for ovum
implantation.43 The oral contraceptive failure rate is 1% in healthy women but
3% to 6% in the population of women with epilepsy.47–49 The Epilepsy Birth
Control Registry, a web-based survey, reported data on 1144 women with
epilepsy aged 18 to 47 years. Most (78.9%) reported having at least one
unintended pregnancy, and 65.0% of their pregnancies were unplanned.50 In
comparison, in 2011, nearly half of pregnancies (45% or 2.8 million of 6.1 million)
in the United States were unintended.51 Others found that fewer than 55% of
women with epilepsy plan their pregnancy, and contraceptive failure is the cause
of one in four unplanned pregnancies.43,48 The consequences can be devastating
considering that women with epilepsy may be taking antiseizure medications
with significant teratogenic side effects. Neurologists should be aware of the
complex and, at times, bidirectional interactions between AEDs and
hormonal contraception.
Enzyme-inducing AEDs, such as phenytoin, carbamazepine, phenobarbital,
primidone, oxcarbazepine, and eslicarbazepine acetate, can result in
contraceptive failure because these drugs lower the level of the progestin
component of hormonal contraceptives through CYP3A4 induction52,53
(TABLE 6-3). Perampanel, felbamate, and topiramate are less potent hepatic
enzyme inducers and can cause contraceptive failure at higher doses.54–56
However, levetiracetam, gabapentin, pregabalin, vigabatrin, tiagabine,
zonisamide, and lacosamide have no known interactions with oral contraceptives
(TABLE 6-3). Women who take enzyme-inducing AEDs may be at risk of
contraceptive failure when taking the progestin-only pills. Theoretically,
enzyme-inducing AEDs may reduce the efficacy of depot-medroxyprogesterone
acetate injections, and it is common practice to administer the injections every
10 weeks in women using enzyme-inducing AEDs, as opposed to the
recommended 12-week intervals for women in the general population. Women
with epilepsy receiving enzyme-inducing AEDs who receive levonorgestrel
subdermal capsules are also at risk of contraceptive failure. Two of nine women
with epilepsy who received enzyme-inducing AEDs became pregnant in a 1 year

study comparing contraceptive efficacy in women receiving levonorgestrel

subdermal implants, while no pregnancies occurred in 10 control women with
epilepsy.57 Both women were taking phenytoin, and their plasma concentrations
of levonorgestrel were low at the time of conception. In a separate report, there
were significant reductions in etonogestrel concentrations seen in 10 women
receiving carbamazepine 600 mg/d who were concurrently using a contraceptive
implant.53
Intrauterine devices (IUDs) are T-shaped devices that are fitted into the
uterus. The two common types are copper IUDs (nonhormonal) and
levonorgestrel-releasing IUDs. The copper IUD prevents pregnancy by releasing
into the uterus a small amount of copper, which is toxic to sperm, and the
levonorgestrel-releasing IUD causes a thickening of the cervical mucus, thereby
blocking sperm passage. IUDs are highly efficacious and are not affected by
enzyme-inducing AEDs, making them a reasonable form of contraception for
women with epilepsy.
PREGNANCY AND PERINATAL COUNSELING

Approximately 24,000 women with epilepsy become pregnant each year. In the
majority of women with epilepsy, pregnancy has no effect on their seizure
frequency; therefore, if seizures are well controlled, they are likely to remain so
during pregnancy. However, in approximately 20% to 35% of pregnancies in
women with epilepsy, an increased seizure frequency occurs during pregnancy.58,59
In an analysis of data from the European Registry of Antiepileptic Drugs and
Pregnancy (EURAP), an international AED and pregnancy registry, 33.4% of
pregnant women experienced seizures during pregnancy. Seizure frequency was
unchanged in 70.5%, reduced in 12.0%, and increased in 15.8% of women with
epilepsy.60 The Australian Pregnancy Registry for women with epilepsy found
that, between 1998 and late 2016, seizures had occurred during pregnancy in
approximately 43%. Seizures of any type occurred in 78.4% of pregnancies when
seizures had occurred in the previous year (active epilepsy) and in 22.3% of those
associated with inactive epilepsy. From these data, having a seizure disorder
that was active in the year before pregnancy and in early pregnancy appears to be
the best predictor of seizure recurrence during pregnancy.61
Cagnetti and colleagues62 reported better seizure control during pregnancy
in women with catamenial epilepsy compared with women with epilepsy in
general. They prospectively followed women referred to their Epilepsy Center
for pregnancy planning. In this study, 59 women with catamenial epilepsy and
215 with noncatamenial epilepsy were included. Forty-seven women (79.7%)
with catamenial epilepsy and 48 women (22.3%) with noncatamenial epilepsy
remained seizure free throughout pregnancy.62
The reasons for seizure recurrence during pregnancy are multifactorial,
including lowering or stopping antiseizure medications because of fear of
harming the baby, hormonal fluctuations, and a higher estrogen-to-progesterone
ratio especially in weeks 8 to 16 of pregnancy, sleep deprivation, and psychosocial
stress. The most common cause of seizure recurrence in pregnancy, however,
is likely reduced plasma concentration of AEDs and changes in AED metabolism.
The plasma AED concentration fluctuates during pregnancy because of several
physiologic reasons, including increased renal clearance, altered hepatic
absorption, increased plasma volume of distribution, and hepatic enzyme
induction by steroid hormones.63 A reduction in AED blood levels during
416 APRIL 2019

pregnancy by 35% or more compared with baseline levels may be the cause KEY POINTS
of an increase in seizures.64 During pregnancy, the dose of antiseizure medications
● Intrauterine devices are a
should be adjusted based on the patient’s seizure history and prepregnancy levels. safe and effective method
Lamotrigine metabolism through hepatic glucuronidation is enhanced during of contraception in women
pregnancy by elevated concentrations of sex hormones. Declining plasma with epilepsy.
concentrations of lamotrigine during pregnancy, therefore, may be associated
● In most women with
with increased seizure frequency in more than 40% of patients.65–67 Worsening
epilepsy, pregnancy has no
seizure control in the second and third trimester is more common in women effect or a protective effect
taking lamotrigine than those taking carbamazepine or valproic acid.60 on their seizure frequency.
Lamotrigine clearance during pregnancy is 2 to 3 times higher than before
pregnancy. The levels after delivery reach prepregnancy levels within 1 to 3 weeks. ● Having a seizure disorder
that was active in the year
Based on an analysis of data from six hospitals in Norway and Denmark, before pregnancy appears
zonisamide serum concentration may fall by more than 40% during pregnancy. to be the best predictor
Reimers and colleagues68 found that the lowest zonisamide serum concentration of seizure recurrence
occurred in gestational months 6 to 9, with significant interindividual variability, during pregnancy.
and 4 of 10 patients who were previously seizure free developed breakthrough
● There is evidence for
seizures during pregnancy. The plasma concentration of the active metabolite better seizure control during
of oxcarbazepine declines by 36% to 50% in the late stages of pregnancy and pregnancy in women with
is associated with increased seizure frequency in 50% of women.69,70 Therefore, catamenial epilepsy
compared with women with
an adjustment in the dose of oxcarbazepine is necessary during pregnancy.
epilepsy in general.
The elimination rate of levetiracetam is significantly increased during
pregnancy because of increased renal glomerular filtration in late pregnancy. ● The most common cause
Therapeutic drug monitoring and adjustment of the dose, therefore, is needed.71 of seizure recurrence in
The American Academy of Neurology (AAN) practice guidelines suggest pregnancy is likely a
reduced plasma
checking AED levels at baseline before conception and monthly thereafter.59 concentration of
Dose adjustments should be considered to maintain an effective and stable antiepileptic drugs and
level throughout pregnancy, at least for women with epilepsy who are taking changes in antiepileptic
lamotrigine, oxcarbazepine, levetiracetam, carbamazepine, and phenytoin.59 drug metabolism.
Despite limited evidence for changes in levels of other AEDs during pregnancy,
● American Academy of
AED levels should be regularly monitored during each trimester of pregnancy Neurology guidelines
because of the potential for altered AED bioavailability, metabolism, and recommend checking
clearance. antiepileptic drug levels at
baseline before conception
and monthly thereafter.
Risks to the Fetus Because of Maternal Seizure Recurrence
Women with epilepsy who experience seizures during their pregnancy are at
increased risk of delivering preterm, low-birth-weight, and small-for-
gestational-age newborns. A nationwide population-based study of 1016 women
with epilepsy in Taiwan reported that epileptic seizures during pregnancy
were independently associated with an increased risk of adverse outcomes,
with a 1.36-fold increased risk of low birth weight, 1.63-fold increased risk of
preterm delivery, and 1.37-fold increased risk of small-for-gestational-age
babies.72 This seizure-related risk may be related to fetal hypoxia, acidosis,
decreased blood flow to the placenta, deceleration of fetal heart rate, and trauma
as a result of a maternal fall.73–75 In another study, frequent maternal tonic-clonic
seizures during pregnancy were associated with a lower verbal IQ in their
offspring.76 To avoid these harmful effects of convulsive seizures on the fetus,
controlling seizures during pregnancy is crucial.72 The benefits of using AEDs
during pregnancy may outweigh the potential teratogenic side effects.
In counseling women with epilepsy who plan to become pregnant,
practitioners should discuss the need for staying on antiseizure medication,

simplifying the medication regimen, attempting monotherapy, and selecting

medications with more favorable side effect profiles. Baseline prepregnancy AED
levels should be obtained to guide adjustment of the dose during pregnancy.
Monthly levels during pregnancy will help in detecting a significant decrease in
levels. In case of any seizure recurrence or fall, the patient should be evaluated at
the obstetrics clinic to ensure maternal and fetus well-being. The patient with
active epilepsy should be managed by a team of high-risk obstetricians and
epileptologists working together to manage her throughout her pregnancy,
delivery, and immediate postpartum period.
Risk of Birth Defects or Fetal Death Because of Antiepileptic Drugs

Most women with epilepsy give birth to healthy offspring, but since the 1960s
there have been reports that children of women taking AEDs during pregnancy
may be at increased risk of birth defects, premature birth, low birth weight, and
low Apgar scores.77 A systematic review of 59 studies found that the overall
frequency of major congenital malformations was 3 times greater in children of
women with epilepsy than without epilepsy.78 This increased risk could be
attributed to either AEDs or maternal epilepsy. In a pooled analysis of data from
26 studies, the rate of major congenital malformations was highest in children of
AED-treated women with epilepsy (6.1%) when compared to children of
untreated women with epilepsy (2.8%) and control population women (2.2%).79
Polytherapy (6.8%) resulted in a higher rate of fetal malformation than
monotherapy (4%), and the risk was dose dependent at higher versus lower
doses of valproic acid (>700 mg/d), carbamazepine (>400 mg/d), phenobarbital
(>150 mg/d), and lamotrigine (>300 mg/d).
Multiple worldwide AED pregnancy registries prospectively collect data on
women with and without epilepsy who receive AEDs to determine risk for
adverse pregnancy outcomes and major congenital malformations. The North
American AED Pregnancy Registry (NAAPR) recently reported on the risk of
major congenital malformation for 11 AEDs for pregnancy outcomes of 9294
women with and without epilepsy who were receiving AEDs for any reason
between February 1997 and December 2015, as summarized in FIGURE 6-3.
Major malformations were defined as overt structural, consequential
abnormalities identified between birth and age 5 days. AEDs identified as
having a major malformation risk above controls from highest to lower level
of risk were (drug, with reported percent frequency in parentheses): valproate
(8.9%), phenobarbital (5.9%), topiramate (4.4%), carbamazepine (3.0%),
phenytoin (2.8%), lamotrigine (2.1%), and levetiracetam (2.0%).80 A recent
systematic review of the literature81 similarly found that valproic acid,
topiramate, phenobarbital, phenytoin, ethosuximide, and 11 polytherapies
were significantly associated with major congenital malformations.
Levetiracetam and lamotrigine did not significantly increase the risk of major
congenital malformation.81
A recent NAAPR analysis also showed that women taking AEDs for any
indications are at a higher risk of delivering prematurely (<37 weeks) and giving
birth to a small-for-gestational-age newborn.82 The Australian Pregnancy
Register recently reported an increased risk of intrauterine fetal death when the
fetus was exposed to AEDs, which was statically significant for carbamazepine.83
A prospective observational study from EURAP also found that the most
important risk factor for intrauterine death (spontaneous abortion and stillbirth
418 APRIL 2019

KEY POINT
● In counseling women with

epilepsy who plan to
become pregnant, the
practitioners should discuss
the need for staying on
antiseizure medication,
simplifying the medication
regimen, attempting
monotherapy, and selecting
medications with a more
favorable side effect
profile.
FIGURE 6-3
Risk of major congenital malformations in women taking an antiepileptic drug (AED) as
monotherapy during pregnancy compared with women taking no AEDs based on the data
provided by the North American Antiepileptic Drug Pregnancy Registry (NAAPR) (1997–2005).80
combined) was maternal exposure to AED polytherapy and a parent with a

history of major congenital malformation.84
NEUROMODULATION AND PREGNANCY

Vagal nerve stimulation (VNS) therapy is an adjunctive therapy for patients with
medically refractory epilepsy. Using data from EURAP85 in 26 pregnancies in women
who received VNS therapy during pregnancy (the largest report to date), no evidence
for teratogenicity was found. Another case series of four patients also reported
the safety of VNS during pregnancy.86 In women with drug-resistant epilepsy
who are planning future pregnancy and are not amenable to epilepsy surgery,
VNS placement may be an option to be considered because this nonpharmacologic
therapy may also enable simplification of the patient’s AED polytherapy regimen,
thereby potentially reducing teratogenic risk. CASE 6-1 illustrates several aspects
of caring for a young woman with medically refractory catamenial epilepsy.
Neurodevelopment and Fetal Antiepileptic Drug Exposure

Exposure of the fetus to AEDs through the placenta may have adverse cognitive
and behavioral effects, such as lower IQ, language deficits, autism, and
attention deficit hyperactivity disorder. Several factors contribute to IQ scores,
such as maternal IQ, maternal age, gestational age at birth, prenatal folic acid
supplementation, socioeconomic status, and maternal use of tobacco or alcohol.
Meador and colleagues87 in the NEAD (Neurodevelopmental Effects of
Antiepileptic Drugs) study reported that fetal valproic acid exposure was
associated with a lower IQ and reduced cognitive abilities across a range of
domains at 6 years of age in a dose-dependent manner. The adjusted mean IQ
score of children exposed to high-dose (more than 800 mg/d) valproic acid was
significantly lower than that of controls by 9.7 points. Exposure to other AEDs,
including carbamazepine, lamotrigine, and phenytoin, was not associated with a
lower IQ in children.
Adaptive Behavioral Scale assessments on children ages 3 years to 6 years
exposed to valproic acid, lamotrigine, and carbamazepine in utero demonstrated

CASE 6-1 A 19-year-old woman presented to a tertiary epilepsy center for

evaluation of medically refractory epilepsy. She had developed viral
encephalitis complicated by status epilepticus at 15 years of age requiring
a medication-induced coma in an intensive care setting for 1 month.
Eventually, she was discharged to rehabilitation. Her condition
dramatically improved except for mild psychomotor slowing and
development of medically refractory epilepsy, which clustered around
her menstrual period.
At age 19, the patient was taking a combination of oxcarbazepine
1200 mg total daily dose, topiramate 200 mg total daily dose, and
lamotrigine 600 mg total daily dose. She reported approximately 10 focal
impaired awareness seizures per month, 7 of which clustered
approximately 2 days before the onset of her menstrual period and
3 days afterward. A trial of progesterone lozenges, clobazam, and
acetazolamide around the time of her menstrual period did not result in
successful treatment of her catamenial seizures.
Because the patient was of reproductive age and sexually active, she
was given a prescription for folic acid 1 mg/d and counseled to consider
an intrauterine device with progesterone release for contraception.
Two years later, the patient informed her epileptologist that she was
planning to marry and have a family. She wanted to plan in advance to
reduce the risk of anticonvulsant medications on her future offspring.
Epilepsy monitoring captured seizures of multifocal independent left and
right hemispheric onset, indicating that she was not a candidate for
epilepsy surgery. Topiramate was successfully weaned during inpatient
monitoring, but subsequent attempts to streamline her antiepileptic drug
polytherapy to monotherapy to reduce adverse effects resulted in
increased seizure severity and frequency, especially surrounding the time
of her menstrual periods.
She was counseled about the potential risks of polytherapy, and she
elected to undergo vagal nerve stimulation (VNS) in the hope of further
simplification of her antiseizure medication regimen. Approximately
6 months later and after multiple adjustments of the VNS settings, she
finally was successfully weaned off oxcarbazepine and continued
with lamotrigine monotherapy. Once seizures stabilized, a baseline
lamotrigine serum level was obtained to guide future adjustments of
her dose in case of pregnancy.
COMMENT This patient had medically refractory catamenial epilepsy of multifocal

origin that was not amenable to resective surgery. Unfortunately, she did
not respond to targeted therapy for her hormonally exacerbated seizures,
and further simplification of her antiseizure medications was not possible,
putting her future pregnancy at risk of teratogenicity because of the
antiepileptic drug polytherapy. VNS therapy has not been shown to
adversely affect pregnancy outcome or carry additional risk for
teratogenicity and provided a chance to achieve monotherapy in this
patient with medically refractory catamenial epilepsy.
420 APRIL 2019

specific deficits in socialization, motor function, and a relative weakness in KEY POINT
communication among children exposed to valproate, but not in those exposed
● Exposure to valproate in
to lamotrigine and carbamazepine.88 A systematic meta-analysis of 29 cohort utero may adversely affect
studies found that valproate had similarly detrimental effects on child IQ and contribute to
neurodevelopmental outcomes, while oxcarbazepine and lamotrigine autistic traits.
exposure was associated with an increased occurrence of autism.89
FOLIC ACID SUPPLEMENTATION

Folic acid is essential during gestation because of its involvement in nucleic
acid and amino acid synthesis, cell division, DNA methylation, and tissue
growth. Some AEDs, such as valproic acid, carbamazepine, oxcarbazepine,
phenobarbital, phenytoin, and primidone, alter folic acid metabolism and may
decrease folic acid levels in the blood.90 A randomized double-blind prevention
trial conducted at 33 centers in seven countries by the Medical Research
Council Vitamin Study Research Group determined that folic acid 4 mg/d
supplementation around the time of conception can prevent neural tube defects.
In this study, women with epilepsy were excluded. Other studies recently found
that periconceptional folate supplementation has a positive association with a
better neurodevelopmental outcome, a lower rate of autism spectrum disorder in
the general population,91 and a higher IQ in children exposed to AEDs in utero.92
The US Public Health Service, Centers for Disease Control and Prevention,
and AAN each recommend a dose of 0.4 mg/d of folate for all women of
childbearing age to prevent neural tube defects.93 There are no solid data and,
therefore, no specific recommendation regarding the benefits of higher doses
of folic acid in women with epilepsy taking AEDs. A systematic review
concluded that folic acid 5 mg/d in women without epilepsy may provide 85%
protection against neural tube defects.94 However, a recent population-based
study of 2302 mother-child pairs in Spain suggested that a dosage of folate above
1 mg/d may be associated with an increased risk of small-for-gestational-age
newborns. In another analysis, more than 1 mg/d perinatal folate supplementation
was associated with poorer cognitive development in children 4 years to 5 years
old.95 Some experts have recommended the use of a higher folic acid dose of
4 mg/d to 5 mg/d in women taking valproic acid, carbamazepine, phenobarbital,
phenytoin, and primidone; however, this practice is not uniform, and given
contradictory data concerning higher-dose folic acid safety, further data are needed
before adopting higher-dose folic acid use as a standard practice in the treatment of
women with epilepsy who are receiving valproate or an enzyme-inducing AED.
BREAST-FEEDING
There are many known benefits of breast-feeding, including reduced risk of
lower respiratory tract infections, atopic dermatitis, asthma, acute otitis media,
gastroenteritis, obesity, diabetes mellitus, childhood leukemia, sudden unexpected
death, and necrotizing enterocolitis in babies. Breast-feeding is also beneficial to
the mother in reducing the risk of breast cancer, ovarian cancer, postpartum
depression, and diabetes mellitus. All AEDs can be transmitted into the breast milk
to some degree, but this amount is much less than that previously transmitted
through the placenta to the fetus. The AED level transmitted to an infant via breast
milk depends on multiple factors, including the amount of AED excreted into
breast milk and AED absorption and clearance by the infant. The actual infant
AED exposure from breast milk is usually low, and the benefits of breast-feeding

are ultimately felt to outweigh the potential risks for most women with epilepsy
and their newborns. For barbiturates and benzodiazepines, the risk-benefit ratio
should be evaluated more carefully because of the reports of sedation, lethargy,
weight loss, and higher drug levels in the child than in the mother.
In an extension of the NEAD study, pregnant women with epilepsy who were
taking a single AED (carbamazepine, lamotrigine, phenytoin, or valproate) during
pregnancy were followed to focus on the effects of breast-feeding while receiving
an AED on cognitive outcome of these children at age 3. A total of 42% of these
women who were taking a single AED during pregnancy went on to breast-feed
their newborns. Breast-fed children exposed to AEDs in utero and through breast
milk exhibited higher IQs and enhanced verbal abilities compared with children
exposed to AEDs in utero who were not breast-fed.96 Our present knowledge is not
conclusive regarding the recommendation for breast-feeding in mothers taking
AEDs, however. Current practice is to educate women with epilepsy regarding the
known benefits of breast-feeding and potential risks for infant exposure to AEDs
so that women can make an informed decision. Except for barbiturates and
benzodiazepines, the reported side effects in infants who are breast-fed on other
AEDs have been rare or infrequent, and the benefit may outweigh the risks.
PERIMENOPAUSE AND MENOPAUSE

Perimenopause is characterized by decreased ovarian progesterone secretion,
leading to increased anovulatory menstrual cycles. Early in perimenopause, estrogen
secretion remains high, creating an excitatory environment and contributing to
seizure exacerbation. When menopause is achieved, because of diminished levels of
follicle-stimulating hormone and a hypogonadal state, seizures stabilize. A recent
questionnaire found that two-thirds of menopausal or perimenopausal women
experiencing the early signs of menstrual changes reported seizure exacerbation.97
A history of catamenial epilepsy and the use of hormonal replacement therapy
were associated with an increase in seizure frequency. Menopausal women
generally report a decrease in seizure frequency.
A follow-up randomized, double-blind, placebo-controlled trial of hormone
replacement therapy in menopausal women with epilepsy using 0.625 mg
conjugated equine estrogen and 2.5 mg medroxyprogesterone found that
seizure frequency significantly increased in a dose-related manner with the
use of hormone replacement therapy. The study was terminated because of an
increased risk of breast cancer with this form of hormone replacement therapy.98
For women with catamenial epilepsy who are at risk of seizure exacerbation
during the perimenopause state, antiseizure medications may need to be adjusted
to higher therapeutic levels. Once menopause is reached, AED doses may be
reduced back to their baseline. The 0.625 mg conjugated equine estrogen and
2.5 mg medroxyprogesterone formulation of hormone replacement therapy
should be avoided in these patients. For women in whom hot flashes are
disturbing sleep, consultation with the patient’s gynecologist is warranted to
explore other hormonal treatment approaches.
BONE HEALTH
Some AEDs are identified as an independent risk factor for low bone density
and secondary osteoporosis. The duration of epilepsy and cumulative drug load
both correlate with a progressive reduction in bone mineral density, predisposing
patients to fractures. Persons with epilepsy have a risk of fracture that is 2 to
422 APRIL 2019

6 times higher than that of the general population because of altered bone KEY POINTS
metabolism, decreased bone density, and a propensity to fall as a result of a
● In the
seizure or AED-induced loss of balance.99,100 Risk factors for bone loss other Neurodevelopmental
than epilepsy and AEDs include female sex, postmenopausal status, sedentary Effects of Antiepileptic
lifestyle, smoking, excessive alcohol intake, inadequate sun exposure, and Drugs study, breast-fed
certain endocrine conditions. Many AEDs are known to alter bone metabolism, children exposed to
antiepileptic drugs in utero
especially the enzyme-inducing AEDs.
and through breast milk
The enzyme-inducing AEDs, such as phenytoin, phenobarbital, and exhibited higher IQs and
primidone, are most consistently associated with a low bone mineral density enhanced verbal abilities
and bone disorders.101,102 However, the data regarding the effect of valproic compared with children
exposed to antiepileptic
acid, carbamazepine, oxcarbazepine, lamotrigine, gabapentin, vigabatrin,
drugs in utero who were
levetiracetam, and topiramate on bone metabolism and bone density are not breast-fed.
limited and show conflicting results. Levels of vitamin D metabolites, such as
25-hydroxyvitamin D, may be low in people with epilepsy who are taking ● At perimenopause,
enzyme-inducing AEDs. Elevated bone turnover markers in these patients seizures may worsen
because of a temporary
reflect increased bone remodeling, are associated with a higher rate of bone increased estrogen-
loss, and are an independent predictor of bone fracture.99,101,103 to-progesterone ratio. After
Long-term gabapentin use in several studies was associated with bone loss in menopause is established,
the hip and spine.104–106 Findings with carbamazepine, valproic acid, and women with catamenial
epilepsy may experience
lamotrigine are mixed. In one systematic review of the literature, only 3 of 11 improvement in seizures.
carbamazepine monotherapy studies and 6 of 11 valproic acid monotherapy
studies showed a significant reduction in bone mineral density.107 In a study of ● Antiepileptic drugs used
bone metabolism and density in premenopausal women with epilepsy receiving to treat epilepsy can
adversely affect bone
AED monotherapy (phenytoin, carbamazepine, valproic acid, and lamotrigine),
health and calcium
phenytoin altered bone metabolism and increased bone turnover, while metabolism.
carbamazepine and valproic acid were associated with low calcium levels.108
Lamotrigine did not result in low calcium levels or increased bone turnover
markers. A more recent study on adults treated with carbamazepine,
oxcarbazepine, valproic acid, lamotrigine, topiramate, and levetiracetam
monotherapy found reduced bone mineral density with levetiracetam and
oxcarbazepine.109 Previous studies demonstrated decreased bone mineral
density in children treated with oxcarbazepine.110,111 Another study found no
harmful effects of levetiracetam on bone density, indicating the need for further
research.112 TABLE 6-4 summarizes our current knowledge about the effect of
AEDs on bone turnover and bone mineral density.
Monitoring calcium and vitamin D metabolites is important in patients who
take AEDs. Based on the Society for Endocrinology guidelines, 25-hydroxyvitamin
D concentrations should be maintained at greater than 30 ng/mL, which may
require at least 1500 IU/d to 2000 IU/d of vitamin D supplementation. Dual energy
x-ray absorptiometry (DEXA) scans should be performed periodically to
monitor bone mineral density. If osteopenia or osteoporosis is detected,
consideration should be given to starting bisphosphonates or other therapeutic
agents, increasing calcium and vitamin D supplementation, and/or replacing
enzyme-inducing AEDs. The patient may benefit from a referral to an
endocrinologist. Other treatment options for osteoporosis may include estrogens
and hormonal therapy, parathyroid hormone (teriparatide), and estrogen
agonist/antagonist (raloxifene). TABLE 6-5 summarizes the suggested
considerations in the management of women with epilepsy at risk of osteoporosis.
CASE 6-2 demonstrates some important points in counseling a woman of
menopausal age.

TABLE 6-4 Effect of Antiepileptic Drugs on Bone Mineral Density and Bone Turnover
(Based on Majority of Studies)
Lowering Bone Mineral Density/

Antiepileptic Drug Affecting Bone Metabolism
Benzodiazepines Yes
Carbamazepine Yes
Gabapentin Yes
Lamotrigine No
Levetiracetam Maybe
Oxcarbazepine Yes
Phenobarbital Yes
Phenytoin Yes
Primidone Yes
Topiramate Maybe
Valproic acid Yes
Zonisamide Yes
TABLE 6-5 Recommendations for Women With Epilepsy at Risk of Osteoporosis
If Osteopenia or Osteoporosis Is Not Detected

◆ Monitor calcium and vitamin D levels 1 to 2 times per year
◆ Follow bone density (dual energy x-ray absorptiometry [DEXA] scan), every 2 years,
especially if postmenopausal
◆ Consider calcium (at least 1200 mg/d) and vitamin D (at least 600 IU/d) supplements or
greater amounts to achieve a level of >30 ng/mL
◆ Encourage weight-bearing exercises
◆ Encourage cessation of smoking
◆ Suggest limiting alcohol consumption
◆ Recommend avoiding excessive caffeine
If Osteopenia or Osteoporosis Is Detected
◆ Increase vitamin D supplement to 1500–2000 IU/d to achieve a level of >30 ng/mL
◆ Lifestyle modifications to decrease the risk of falls and fractures
◆ Discuss advantages and disadvantages of switching to another non–enzyme-inducing
antiseizure medication
◆ Consider bisphosphonates (alendronate, ibandronate, risedronate, and zoledronic acid)
◆ Consider estrogens and hormone therapy, parathyroid hormone (teriparatide), and estrogen
agonist/antagonist (raloxifene)
◆ Consider referring to an endocrinologist
424 APRIL 2019

A 52-year-old postmenopausal woman, who had been seizure free on CASE 6-2
oxcarbazepine for 10 years, presented with osteoporosis of her lumbar
spine and hip that had been diagnosed by her primary care physician. Her
seizures had started at the time of puberty and occurred in clusters for
about 3 days of the month around the time of her ovulation. She had a
feeling of déjà vu and a rising butterfly sensation in her stomach followed
by confusion. There were rare instances of progression of focal impaired
awareness to bilateral tonic-clonic seizures. The patient had first been
tried on phenytoin, to which she developed bleeding of her gums while
brushing teeth and gingival hyperplasia. Her seizures did not respond to
valproic acid. Finally, oxcarbazepine was started, and once the dose was
adjusted to high therapeutic levels, she became seizure free.
She presented to discuss her options and plan of care. Her vitamin D
level was low at 20 ng/mL. Long-term inpatient video-EEG monitoring
while the patient was tapered off oxcarbazepine showed independent
left and right temporal interictal sharp waves. Four seizures were
captured, one from the left anterior temporal area and three from the
right anterior and middle temporal region. Her medication was switched
over the course of 2 months to lamotrigine, and she was started on
vitamin D 2000 IU/d with 1200 mg of calcium. Weight-bearing exercises
were encouraged, and the patient was referred to endocrinology and
started on bisphosphonate therapy. The patient’s seizures responded to
lamotrigine, and a repeat bone density study showed significant
improvement in bone mineral density over the next 2 years.
This case highlights the added risk of hepatic enzyme–inducing COMMENT

antiepileptic drugs such as oxcarbazepine in accelerating bone loss and
compromising bone health in women with epilepsy. One challenge is the
risk of seizure recurrence when trying to switch the patient from a hepatic
enzyme–inducing antiepileptic drug. There is no guarantee that the patient
will remain seizure free, and seizure recurrence may result in significant
patient safety, social, and psychological consequences. Up to 21.7% of
patients switched from phenytoin and carbamazepine to other antiseizure
medications are at risk of seizure recurrence within 6 months of the
switch.113 The inpatient epilepsy monitoring in this patient suggested a very
limited possibility for seizure freedom off anticonvulsant therapy.
Fortunately, her seizures responded to lamotrigine, and with proper
vitamin D and calcium replacement and bisphosphonate therapy, the
patient’s osteoporosis improved.

CONCLUSION
The professionals who care for women with epilepsy should be aware of the
challenges these women are facing because of complex hormonal interactions
with their antiseizure medications and seizure disorders. In every step of their
lives from puberty, menses, and birth control to conception, pregnancy,
childbirth, breast-feeding, childcare, bone health, and menopause, women with
epilepsy need to be educated to make informed decisions. Practitioners need to
keep up-to-date with the latest treatment recommendations to help women with
epilepsy live safe and productive lives.
REFERENCES
1 Ngugi AK, Kariuki SM, Bottomley C, et al. 14 Murphy DD, Cole NB, Greenberger V, Segal M.
Incidence of epilepsy: a systematic review and Estradiol increases dendritic spine density
meta-analysis. Neurology 2011;77(10):1005–1012. by reducing GABA neurotransmission in
doi:10.1212/WNL.0b013e31822cfc90. hippocampal neurons. J Neurosci 1998;18(7):
2550–2559. doi:10.1523/JNEUROSCI.18-07-02550.
2 Ngugi AK, Bottomley C, Kleinschmidt I, et al.
Estimation of the burden of active and life-time 15 Rudick CN, Woolley CS. Estrogen regulates
epilepsy: a meta-analytic approach. Epilepsia 2010; functional inhibition of hippocampal CA1
51(5):883–890. doi:10.1111/j.1528-1167.2009.02481.x. pyramidal cells in the adult female rat. J Neurosci
2001;21(17):6532–6543. doi:10.1523/JNEUROSCI.
3 Locock C. Discussion. In: Sieveking EH, editor.
21-17-06532.
Analysis of fifty-two cases of epilepsy observed
by the author. Med Times Gaz 1857;14:524–526. 16 Woolley CS, Weiland NG, McEwen BS,
Schwartzkroin PA. Estradiol increases the
4 Foldvary-Schaefer N. Introduction. Atlas of
sensitivity of hippocampal CA1 pyramidal cells
epilepsies. London, UK: Springer, 2010:1295–1299.
to NMDA receptor-mediated synaptic input:
5 Patel S, Foldvary-Schaefer N. Catamenial correlation with dendritic spine density.
epilepsy. In: Women with epilepsy: a practical J Neurosci 1997;17(5):1848–1859. doi:10.1523/
management handbook. New York, NY: JNEUROSCI.17-05-01848.
Cambridge University Press, 2014:101–112.
17 Herzog AG. Differential impact of antiepileptic
6 Gowers WR. Epilepsy and other chronic convulsive drugs on the effects of contraceptive methods
disorders. 1st ed. London, UK: J and A Churchill, 1881. on seizures: interim findings of the Epilepsy
Birth Control Registry. Seizure 2015;28:71–75.
7 Bäckström T. Epileptic seizures in women related
doi:10.1016/j.seizure.2015.02.011.
to plasma estrogen and progesterone during the
menstrual cycle. Acta Neurol Scand 1976;54(4): 18 Herzog AG, Mandle HB, Cahill KE, et al.
321–347. Differential impact of contraceptive methods on
seizures varies by antiepileptic drug category:
8 Laidlaw J. Catamenial epilepsy. Lancet 1956;
findings of the Epilepsy Birth Control Registry.
271(6955):1235–1237.
9 Herzog AG, Klein P, Ransil BJ. Three patterns of j.yebeh.2016.04.020.
catamenial epilepsy. Epilepsia 1997;38(10):
19 Herzog AG, Mandle HB, Cahill KE, et al.
1082–1088. doi:10.1111/j.1528-1157.1997.tb01197.x.
Contraceptive practices of women with
10 Herzog AG. Catamenial epilepsy: definition, epilepsy: findings of the Epilepsy Birth Control
prevalence pathophysiology and treatment. Registry. Epilepsia 2016;57(4):630–637.
Seizure 2008;17(2):151–159. doi:10.1016/j. doi:10.1111/epi.13320.
seizure.2007.11.014.
20 Reddy DS. Neuroendocrine aspects of
11 Reddy DS. Pharmacology of endogenous catamenial epilepsy. Horm Behav 2013;63(2):
neuroactive steroids. Crit Rev Neurobiol 2003; 254–266. doi:10.1016/j.yhbeh.2012.04.016.
15(3–4):197–234.
21 Harden CL, Pennell PB, French JA, et al.
12 Reddy DS, Kim HY, Rogawski MA. Neurosteroid Anti-Mullerian hormone is higher in seizure-free
withdrawal model of perimenstrual catamenial women with epilepsy compared to those with
epilepsy. Epilepsia 2001;42(3):328–336. ongoing seizures. Epilepsy Res 2016;127:66–71.
doi:10.1016/j.eplepsyres.2016.08.003.
13 Younus I, Reddy DS. Seizure facilitating activity
of the oral contraceptive ethinyl estradiol.
Epilepsy Res 2016;121:29–32. doi:10.1016/
j.eplepsyres.2016.01.007.
426 APRIL 2019

22 Lebeurrier N, Launay S, Macrez R, et al. 35 Bauer J, Cooper-Mahkorn D. Reproductive
Anti-Mullerian-hormone-dependent dysfunction in women with epilepsy: menstrual
regulation of the brain serine-protease inhibitor cycle abnormalities, fertility, and polycystic
neuroserpin. J Cell Sci 2008;121(pt 20):3357–3365. ovary syndrome. Int Rev Neurobiol 2008;83:
doi:10.1242/jcs.031872. 135–155. doi:10.1016/S0074-7742(08)00007-X.
23 Gazzaley AH, Weiland NG, McEwen BS, Morrison 36 Knochenhauer ES, Key TJ, Kahsar-Miller M, et al.
JH. Differential regulation of NMDAR1 mRNA Prevalence of the polycystic ovary syndrome in
and protein by estradiol in the rat hippocampus. unselected black and white women of the
J Neurosci 1996;16(21):6830–6838. doi:10.1523/ southeastern United States: a prospective study.
JNEUROSCI.16-21-06830. J Clin Endocrinol Metab 1998;83(9):3078–3082.
doi:10.1210/jcem.83.9.5090.
24 Cimino I, Casoni F, Liu X, et al. Novel role for
anti-Müllerian hormone in the regulation of GnRH 37 Herzog AG, Seibel MM, Schomer DL, et al.
neuron excitability and hormone secretion. Reproductive endocrine disorders in women
Nat Commun 2016;7:10055. doi:10.1038/ with partial seizures of temporal lobe origin. Arch
ncomms10055. Neurol 1986;43(4):341–346. doi:10.1001/
archneur.1986.00520040029014.
25 Herzog AG, Coleman AE, Jacobs AR, et al.
Interictal EEG discharges, reproductive 38 Webber MP, Hauser WA, Ottman R, Annegers JF.
hormones, and menstrual disorders in epilepsy. Fertility in persons with epilepsy: 1935-1974.
Ann Neurol 2003;54(5):625–637. doi:10.1002/ Epilepsia 1986;27(6):746–752. doi:10.1111/
ana.10732. j.1528-1157.1986.tb03605.x.
26 Lim LL, Foldvary N, Mascha E, Lee J. 39 Olafsson E, Hauser WA, Gudmundsson G. Fertility
Acetazolamide in women with catamenial in patients with epilepsy: a population-based
epilepsy. Epilepsia 2001;42(6):746–749. study. Neurology 1998;51(1):71–73. doi:10.1212/
doi:10.1046/j.1528-1157.2001.33600.x. WNL.51.1.71.
27 Bergstrom W, Carzoli RF, Lomroso C, et al. 40 Pennell PB, French JA, Harden CL, et al. Fertility
Observations on the metabolic and clinical and birth outcomes in women with epilepsy
effects of carbonic-anhydrase inhibitors in seeking pregnancy. JAMA Neurol 2018;75(8)
epileptics. AMA Am J Dis Child 1952;84(6): 962–969. doi:10.1001/jamaneurol.2018.0646.
771–772.
41 Isojärvi JI, Taubøll E, Herzog AG. Effect of
28 Feely M, Calvert R, Gibson J. Clobazam in antiepileptic drugs on reproductive endocrine
catamenial epilepsy. A model for evaluating function in individuals with epilepsy. CNS Drugs
anticonvulsants. Lancet 1982;2(8289):71–73. 2005;19(3):207–223.
doi:10.1016/S0140-6736(82)91691-9.
42 Isojärvi JI, Laatikainen TJ, Pakarinen AJ, et al.
29 Feely M, Gibson J. Intermittent clobazam for Polycystic ovaries and hyperandrogenism in
catamenial epilepsy: tolerance avoided. J Neurol women taking valproate for epilepsy. N Engl J
Neurosurg Psychiatry 1984;47(12):1279–1282. Med 1993;329(19):1383–1388. doi:10.1056/
doi:10.1136/jnnp.47.12.1279. NEJM199311043291904.
30 Zimmerman AW, Holden KR, Reiter EO, 43 Reimers A, Brodtkorb E, Sabers A. Interactions
Dekaban AS. Medroxprogesterone acetate in between hormonal contraception and
the treatment of seizures associated with antiepileptic drugs: clinical and mechanistic
menstruation. J Pediatr 1973;83(6):959–963. considerations. Seizure 2015;28:66–70.
doi:10.1016/S0022-3476(73)80529-3. doi:10.1016/j.seizure.2015.03.006.
31 Mattson RH, Cramer JA, Caldwell BV, 44 Mattson RH, Cramer JA, Collins JF, et al.
Siconolfi BC. Treatment of seizures with Comparison of carbamazepine, phenobarbital,
medroxyprogesterone acetate: preliminary phenytoin, and primidone in partial and
report. Neurology 1984;34(9):1255–1258. secondarily generalized tonic-clonic seizures.
doi:10.1212/WNL.34.9.1255. N Engl J Med 1985;313(3):145–151. doi:10.1056/
NEJM198507183130303.
32 Herzog AG, Fowler KM, Smithson SD, et al.
Progesterone vs placebo therapy for women 45 Taubøll E, Gregoraszczuk EL, Tworzydø A, et al.
with epilepsy: a randomized clinical trial. Comparison of reproductive effects of
Neurology 2012;78(24):1959–1966. doi:10.1212/ levetiracetam and valproate studied in
WNL.0b013e318259e1f9. prepubertal porcine ovarian follicular cells.
Epilepsia 2006;47(9):1580–1583. doi:10.1111/
33 Herzog AG. Disorders of reproduction in patients
j.1528-1167.2006.00668.x.
with epilepsy: primary neurological mechanisms.
Seizure 2008;17(2):101–110. doi:10.1016/ 46 Svalheim S, Taubøll E, Leuf G, et al. Differential
j.seizure.2007.11.025. effects of levetiracetam, carbamazepine, and
lamotrigine on reproductive endocrine function
34 Balen A. Pathogenesis of polycystic ovary
in adults. Epilepsy Behav 2009;16(2):281–287.
syndrome—the enigma unravels? Lancet 1999;
doi: 10.1016/j.yebeh.2009.07.033.
354(9183):966–967. doi:10.1016/S0140-
6736(99)00218-4.

47 Coulam CB, Annegers JF. Do anticonvulsants 60 Battino D, Tomson T, Bonizzoni E, et al. Seizure
reduce the efficacy of oral contraceptives? control and treatment changes in pregnancy:
Epilepsia 1979;20(5):519–525. doi:10.1111/ observations from the EURAP epilepsy
j.1528-1157.1979.tb04834.x. pregnancy registry. Epilepsia 2013;54(9):
1621–1627. doi:10.1111/epi.12302.
48 Fairgrieve SD, Jackson M, Jonas P, et al.
Population based, prospective study of the care 61 Vajda FJE, O'Brien TJ, Graham JE, et al. Predicting
of women with epilepsy in pregnancy. BMJ 2000; epileptic seizure control during pregnancy.
321(7262):674–675. doi:10.1136/bmj.321.7262.674. Epilepsy Behav 2018;78:91–95. doi:10.1016/
j.yebeh.2017.10.017.
49 Johnston CA, Crawford PM. Anti-epileptic drugs
and hormonal treatments. Curr Treat Options 62 Cagnetti C, Lattanzi S, Foschi N, et al. Seizure
Neurol 2014;16(5):288. doi:10.1007/s11940-014- course during pregnancy in catamenial epilepsy.
0288-3. Neurology 2014;83(4):339–344. doi:10.1212/
WNL.0000000000000619.
50 Herzog AG, Mandle HB, Cahill KE, et al. Predictors
of unintended pregnancy in women with 63 Leppik IE, Rask CA. Pharmacokinetics of
epilepsy. Neurology 2017;88(8):728–733. antiepileptic drugs during pregnancy. Semin
doi:10.1212/WNL.0000000000003637. Neurol 1988;8(3):240–246. doi:10.1055/
s-2008-1041385.
51 Finer LB, Zolna MR. Declines in unintended
pregnancy in the United States, 2008-2011. 64 Reisinger TL, Newman M, Loring DW, et al.
N Engl J Med 2016;374(9):843–852. doi:10.1056/ Antiepileptic drug clearance and seizure
NEJMsa1506575. frequency during pregnancy in women with
epilepsy. Epilepsy Behav 2013;29(1):13–18.
52 Davis AR, Westhoff CL, Stanczyk FZ.
doi:10.1016/j.yebeh.2013.06.026.
Carbamazepine coadministration with an
oral contraceptive: effects on steroid 65 EURAP Study Group. Seizure control and
pharmacokinetics, ovulation, and bleeding. treatment in pregnancy: observations from the
Epilepsia 2011;52(2):243–247. doi:10.1111/ EURAP epilepsy pregnancy registry. Neurology
j.1528-1167.2010.02917.x. 2006;66(3):354–360. doi:10.1212/01.
wnl.0000195888.51845.80.
53 Lazorwitz A, Davis A, Swartz M, Guiahi M.
The effect of carbamazepine on etonogestrel 66 Ohman I, Beck O, Vitols S, Tomson T. Plasma
concentrations in contraceptive implant users. concentrations of lamotrigine and its 2-N-
Contraception 2017;95(6):571–577. doi:10.1016/ glucuronide metabolite during pregnancy in
j.contraception.2017.03.004. women with epilepsy. Epilepsia 2008;49(6):
1075–1080. doi:10.1111/j.1528-1167.2007.01471.x.
54 Brodie MJ, Kwan P. Newer drugs for focal
epilepsy in adults. BMJ 2012;344:e345. 67 Pennell PB, Peng L, Newport DJ, et al. Lamotrigine
doi:10.1136/bmj.e345. in pregnancy: clearance, therapeutic drug
monitoring, and seizure frequency. Neurology
55 Brodie MJ, Mintzer S, Pack AM, et al. Enzyme
2008;70(22 pt 2):2130–2136. doi:10.1212/01.
induction with antiepileptic drugs: cause for
wnl.0000289511.20864.2a.
concern? Epilepsia 2013;54(1):11–27. doi:10.1111/
j.1528-1167.2012.03671.x. 68 Reimers A, Helde G, Becser Andersen N, et al.
Zonisamide serum concentrations during
56 Perucca E, Tomson T. The pharmacological
pregnancy. Epilepsy Res 2018;144:25–29.
treatment of epilepsy in adults. Lancet
doi:10.1016/j.eplepsyres.2018.05.002.
Neurol 2011;10(5):446–456. doi:10.1016/S1474-
4422(11)70047-3. 69 Christensen J, Sabers A, Sidenius P.
Oxcarbazepine concentrations during
57 Haukkamaa M. Contraception by Norplant
pregnancy: a retrospective study in patients with
subdermal capsules is not reliable in epileptic
epilepsy. Neurology 2006;67(8):1497–1499.
patients on anticonvulsant treatment.
doi:10.1212/01.wnl.0000240047.11166.0e.
Contraception 1986;33(6):559–565.
doi:10.1016/0010-7824(86)90044-2. 70 Petrenaite V, Sabers A, Hansen-Schwartz J.
Seizure deterioration in women treated with
58 Sabers A. Influences on seizure activity in
oxcarbazepine during pregnancy. Epilepsy
pregnant women with epilepsy. Epilepsy
Res 2009;84(2–3):245–249. doi:10.1016/
Behav 2009;15(2):230–234. doi:10.1016/
j.yebeh.2009.03.031.
71 Tomson T, Palm R, Källén K, et al. Pharmacokinetics
59 Harden CL, Hop J, Ting TY, et al. Practice
of levetiracetam during pregnancy, delivery, in the
parameter update: management issues for
neonatal period, and lactation. Epilepsia 2007;
women with epilepsy–focus on pregnancy
48(6):1111–1116. doi:10.1111/j.1528-1167.2007.01032.x.
(an evidence-based review): obstetrical
complications and change in seizure frequency: 72 Chen YH, Chiou HY, Lin HC, Lin HL. Affect of
report of the Quality Standards Subcommittee seizures during gestation on pregnancy
and Therapeutics and Technology Assessment outcomes in women with epilepsy. Arch
Subcommittee of the American Academy of Neurol 2009;66(8):979–984. doi:10.1001/
Neurology and American Epilepsy Society. archneurol.2009.142.
Neurology 2009;73(2):126–132. doi:10.1212/
WNL.0b013e3181a6b2f8.
428 APRIL 2019

73 Hiilesmaa VK. Pregnancy and birth in women with 87 Meador KJ, Baker GA, Browning N. Fetal
epilepsy. Neurology 1992;42(4 suppl 5):8–11. antiepileptic drug exposure and cognitive
outcomes at age 6 years (NEAD study): a
74 Teramo K, Hiilesmaa V, Bardy A, Saarikoski S. Fetal
prospective observational study. Lancet
heart rate during a maternal grand mal epileptic
Neurol 2013;12(3):244–252. doi:10.1016/S1474-
seizure. J Perinat Med 1979;7(1):3–6. doi:10.1515/
4422(12)70323-X.
jpme.1979.7.1.3.
88 Deshmukh U, Adams J, Macklin EA, et al.
75 Sveberg L, Svalheim S, Taubøll E. The impact of
Behavioral outcomes in children exposed
seizures on pregnancy and delivery. Seizure 2015;
prenatally to lamotrigine, valproate, or
28:35–38. doi:10.1016/j.seizure.2015.02.020.
carbamazepine. Neurotoxicol Teratol 2016;54:
76 Adab N, Kini U, Vinten J, et al. The longer term 5–14. doi:10.1016/j.ntt.2016.01.001.
outcome of children born to mothers with
89 Veroniki AA, Rios P, Cogo E, et al. Comparative
epilepsy. J Neurol Neurosurg Psychiatry 2004;
safety of antiepileptic drugs for neurological
75(11):1575–1583. doi:10.1136/jnnp.2003.029132.
development in children exposed during
77 Meadow SR. Anticonvulsant drugs and pregnancy and breast feeding: a systematic
congenital abnormalities. Lancet 1968; review and network meta-analysis. BMJ Open 2017;
2:1296. 7(7):e017248. doi:10.1136/bmjopen-2017-017248.
78 Meador K, Reynolds MW, Crean S, et al. 90 Dansky LV, Rosenblatt DS, Andermann E.
Pregnancy outcomes in women with epilepsy: Mechanisms of teratogenesis: folic acid and
a systematic review and meta-analysis of antiepileptic therapy. Neurology 1992;
published pregnancy registries and cohorts. 42(4 suppl 5):32–42.
Epilepsy Res 2008;81(1):1–13. doi:10.1016/
91 Levine SZ, Kodesh A, Viktorin A, et al. Association
of maternal use of folic acid and multivitamin
79 Shorvon SD, Tomson T, Cock HR. The management supplements in the periods before and during
of epilepsy during pregnancy—progress is painfully pregnancy with the risk of autism spectrum
slow. Epilepsia 2009;50(5):973–974. doi:10.1111/ disorder in offspring. JAMA Psychiatry 2018;75(2):
j.1528-1167.2009.02134.x. 176–184. doi:10.1001/jamapsychiatry.2017.4050.
80 Holmes LB, Hernandez-Diaz S, Quinn M, et al. 92 Gao Y, Sheng C, Xie RH, et al. New perspective
Update on monotherapy findings: comparative on impact of folic acid supplementation during
safety of 11 antiepileptic drugs used during pregnancy on neurodevelopment/autism in the
pregnancy (Winter 2016). The North American offspring children—a systematic review. PLoS
Pregnancy Registry newsletter. www. One 2016;11(11):e0165626. doi:10.1371/journal.
aedpregnancyregistry.org/wp-content/ pone.0165626.
uploads/2016-newsletter-Winter-2016.pdf.
93 Harden CL, Meador KJ, Pennell PB, et al. Practice
Accessed February 7, 2019.
parameter update: management issues for
81 Veroniki AA, Cogo E, Rios P, et al. Comparative women with epilepsy—focus on pregnancy
safety of anti-epileptic drugs during pregnancy: (an evidence-based review): teratogenesis and
a systematic review and network meta-analysis perinatal outcomes: report of the Quality
of congenital malformations and prenatal Standards Subcommittee and Therapeutics and
outcomes. BMC Med 2017;15(1):95. doi:10.1186/ Technology Assessment Subcommittee of the
s12916-017-0845-1. American Academy of Neurology and American
Epilepsy Society. Neurology 2009;73(2):133–141.
82 Hernández-Díaz S, McElrath TF, Pennell PB, et al.
doi:10.1212/WNL.0b013e3181a6b312.
Fetal growth and premature delivery in pregnant
women on antiepileptic drugs. Ann Neurol 2017; 94 Wald NJ, Hackshaw AD, Stone R, Sourial NA.
82(3):457–465. doi:10.1002/ana.25031. Blood folic acid and vitamin B12 in relation to
neural tube defects. Br J Obstet Gynaecol 1996;
83 Vajda FJE, O'Brien TJ, Graham J, et al.
103(4):319–324. doi:10.1111/j.1471-0528.1996.
Anti-epileptic drug exposure and risk of foetal
tb09735.x.
death in utero. Acta Neurol Scand 2018;137(1):
20–23. doi:10.1111/ane.12816. 95 Valera-Gran D, Navarrete-Muñoz EM, Garcia de
la Hera M, et al. Effect of maternal high dosages
84 Tomson T, Battino D, Bonizzoni E, et al.
of folic acid supplements on neurocognitive
Antiepileptic drugs and intrauterine death: a
development in children at 4-5 y of age: the
prospective observational study from EURAP.
prospective birth cohort Infancia y Medio
Neurology 2015;85(7):580–588. doi:10.1212/
Ambiente (INMA) study. Am J Clin Nutr 2017;
WNL.0000000000001840.
106(3):878–887. doi:10.3945/ajcn.117.152769.
85 Sabers A, Battino D, Bonizzoni E, et al. Maternal
96 Meador KJ, Baker GA, Browning N, et al.
and fetal outcomes associated with vagus nerve
Breastfeeding in children of women taking
stimulation during pregnancy. Epilepsy Res 2017;
antiepileptic drugs: cognitive outcomes at age
137:159–162. doi:10.1016/j.eplepsyres.2017.05.013.
6 years. JAMA Pediatr 2014;168(8):729–736.
86 Rodríguez-Osorio X, López-González FJ, doi:10.1001/jamapediatrics.2014.118.
Garamendi Í, et al. VNS and pregnancy: a
multicentric experience of four cases. Acta Neurol
Scand 2017;136(4):372–374. doi:10.1111/ane.12780.

97 Harden CL, Pulver MC, Ravdin L, Jacobs AR. 105 El-Hajj Fuleihan G, Dib L, Yamout B, et al.
The effect of menopause and perimenopause on Predictors of bone density in ambulatory
the course of epilepsy. Epilepsia 1999;40(10): patients on antiepileptic drugs. Bone 2008;43(1):
1402–1407. doi:10.1111/j.1528-1157.1999.tb02012.x. 149–155. doi:10.1016/j.bone.2008.03.002.
98 Harden CL, Herzog AG, Nikolov BG, et al. 106 Vestergaard P, Rejnmark L, Mosekilde L.
Hormone replacement therapy in women Anxiolytics and sedatives and risk of fractures:
with epilepsy: a randomized, double-blind, effects of half-life. Calcif Tissue Int 2008;82(1):
placebo-controlled study. Epilepsia 2006;47(9): 34–43. doi:10.1007/s00223-007-9095-0.
1447–1451. doi:10.1111/j.1528-1167.2006.00507.x.
107 Vestergaard P. Effects of antiepileptic drugs on
99 Pack AM. Treatment of epilepsy to optimize bone health and growth potential in children
bone health. Curr Treat Options Neurol 2011;13(4): with epilepsy. Paediatr Drugs 2015;17(2):141–150.
346–354. doi:10.1007/s11940-011-0133-x. doi:10.1007/s40272-014-0115-z.
100 Souverein PC, Webb DJ, Petri H, et al. Incidence 108 Pack AM, Morrell MJ, Marcus R. Bone mass and
of fractures among epilepsy patients: a turnover in women with epilepsy on antiepileptic
population-based retrospective cohort study in drug monotherapy. Ann Neurol 2005;57(2):
the General Practice Research Database. 252–257. doi:10,1002/ana.20378.
Epilepsia 2005;46(2):304–310. doi:10.1111/
109 Beniczky SA, Viken J, Jensen LT, Andersen NB.
j.0013-9580.2005.23804.x.
Bone mineral density in adult patients treated
101 Pack AM, Walczak TS. Bone health in women with various antiepileptic drugs. Seizure 2012;
with epilepsy: clinical features and potential 21(6):471–472. doi:10.1016/j.seizure.2012.04.002.
mechanisms. Int Rev Neurobiol 2008;83:305–328.
110 Babayigit A, Dirik E, Bober E, Cakmakci H, et al.
doi:10.1016/S0074-7742(08)00018-4.
Adverse effects of antiepileptic drugs on bone
102 Verrotti A, Coppola G, Parisi P, et al. Bone and mineral density. Pediatr Neurol 2006;35(3):
calcium metabolism and antiepileptic drugs. Clin 177–181. doi:10.1016/j.pediatrneurol.2006.03.004.
Neurol Neurosurg 2010;112(1):1–10. doi:10.1016/
111 Cansu A, Yesilkaya E, Serdaroğlu A, et al.
j.clineuro.2009.10.011.
Evaluation of bone turnover in epileptic children
103 Verrotti A, Greco R, Morgese G, Chiarelli F. using oxcarbazepine. Pediatr Neurol 2008;39(4):
Increased bone turnover in epileptic patients 266–271. doi:10.1016/j.pediatrneurol.2008.07.001.
treated with carbamazepine. Ann Neurol 2000;
112 Koo DL, Joo EY, Kim D, Hong SB. Effects of
47(3):385–388. doi:10.1002/1531-8249(200003)47:
levetiracetam as a monotherapy on bone mineral
3<385::AID-ANA18>3.0.CO;2-W.
density and biochemical markers of bone
104 Ensrud KE, Walczak TS, Blackwell TL, et al. metabolism in patients with epilepsy. Epilepsy
Antiepileptic drug use and rates of hip bone loss Res 2013;104(1–2):134–139. doi:10.1016/
in older men: a prospective study. Neurology j.eplepsyres.2012.09.002.
2008;71(10):723–730. doi:10.1212/01.
113 Wang SP, Mintzer S, Skidmore CT. Seizure
wnl.0000324919.86696.a9.
recurrence and remission after switching
antiepileptic drugs. Epilepsia 2013;54(1):187–193.
doi:10.1111/j.1528-1167.2012.03652.x.
430 APRIL 2019

Electroencephalography REVIEW ARTICLE
in Epilepsy Evaluation

C O N T I N U UM A U D I O
ONLINE
By Hai Chen, MD, PhD; Mohamad Z. Koubeissi, MD, FAAN
ABSTRACT
PURPOSE OF REVIEW: Epilepsy is a heterogeneous disorder that is often
associated with abnormal electroencephalogram (EEG) findings. This
article provides an overview of common EEG findings in epileptic
disorders. The physiologic basis of EEG and intracranial EEG studies is also
discussed.
RECENT FINDINGS:EEG is widely used in clinical practice. Because of the

paroxysmal nature of seizure disorders, interictal epileptiform discharges,
such as spikes and sharp waves, are often used to support the diagnosis
of epilepsy when a habitual seizure is not captured by EEG. Interictal and
ictal EEG findings also underlie the classification of seizures and epilepsy.
Continuous critical care EEG monitoring has become an invaluable study in
the diagnosis and treatment of subclinical seizures and nonconvulsive CITE AS:
status epilepticus. Intracranial EEG with subdural or intraparenchymal
2019;25(2, EPILEPSY):431–453.
electrodes is warranted when localization of the seizure focus and
mapping of eloquent brain areas are required to plan epilepsy surgery. Address correspondence to
Dr Mohamad Z. Koubeissi,
George Washington University,
SUMMARY: The EEG is a key tool in the diagnosis of epilepsy. Interictal and
Department of Neurology, 2150
ictal EEG findings are crucial for the confirmation and classification of Pennsylvania Ave NW,
seizure disorders. Intracranial EEG monitoring is also indispensable for Washington, DC 20037,
mkoubeissi@mfa.gwu.edu.
planning surgery for some patients.
Dr Chen reports no disclosure.
Dr Koubeissi serves on the
INTRODUCTION editorial boards of Epilepsy
E
Currents and Functional
lectroencephalographic (EEG) signals, which are generated by cortical Neurology and as the surgery
neuronal postsynaptic potentials, provide real-time assessment of and device editor of Epilepsy.com.
cerebral physiologic function. The analysis of EEG findings, Dr Koubeissi has received
personal compensation for
particularly seizures and interictal epileptiform discharges, plays a serving on the speakers’
crucial role in the diagnosis of epilepsy. In addition, EEG findings are bureaus of Sunovion
essential for classifying seizure types and epilepsy syndromes. This review Pharmaceuticals Inc and UCB SA,
and has received publishing
focuses on common interictal and ictal findings on the scalp as well as pertinent royalties from Springer for his
intracranial EEG findings. book, Epilepsy Board Review.
UNLABELED USE OF
PHYSIOLOGIC BASIS OF ELECTROENCEPHALOGRAPHY PRODUCTS/INVESTIGATIONAL
USE DISCLOSURE:
The EEG records the summation of postsynaptic potentials generated by Drs Chen and Koubeissi report
pyramidal cells in the cerebral cortex. The postsynaptic potentials are generated no disclosures.
by ion flow across postsynaptic cell membranes.1 Depending on the type of ion
channel activated, postsynaptic potentials can be excitatory or inhibitory. A © 2019 American Academy
current sink is generated at the site of an excitatory postsynaptic potential with of Neurology.

EEG IN EPILEPSY EVALUATION
an inflow of positively charged ions (such as Na+ or Ca+) into the postsynaptic
membrane. A corresponding source occurs at a distance, where positive ions exit
the neuron into the extracellular space. Electric currents flow from the source to
the sink in the extracellular space, giving rise to the extracellular field potential.
Thus, an electrode that is close to the current sink records a negative potential
because the inward flow of positive ions causes negativity in the extracellular
space nearby. Similarly, an electrode that is close to a current source records a
positive potential because of the inward flow of negatively charged ions (such as
Cl−) into the intracellular space.
Pyramidal neurons are perpendicularly aligned to the cortical surface. This
allows for summated potentials with a vertical dipole that can be detected by
surface electrodes. The electric activity from deeper generators gets dispersed
and attenuated by volume conduction effects. EEG detects the summated
extracellular field potential from the underlying cortex, and it has been proposed
that at least 6 cm2 of a synchronously active area of the cortex is required to
produce a detectable scalp EEG signal with the use of an in vitro cadaver skull
model. A more recent study using concomitant intracranial and scalp electrodes
found that intracranial spikes synchronizing over an area of less than 6 cm2 were
never associated with a scalp EEG signal. Rather, 90% of spikes with a source area
of more than 10 cm2 yielded detectable spikes on a scalp EEG, whereas only 10%
of cortical spikes with less than 10 cm2 of source area were seen on a scalp EEG.2
INTERICTAL EPILEPTIFORM DISCHARGES

Interictal epileptiform discharges aid in the diagnosis of epilepsy and are
sometimes the only abnormal finding encountered on routine EEG.
Definition and Clinical Significance

Interictal epileptiform discharges are defined as sharply contoured waveforms or
complexes that are distinct from background activity and resemble those
recorded in a proportion of human subjects with epileptic disorders.3 Spikes and
sharp waves are the most commonly recognized interictal epileptiform
discharges. A spike on a scalp EEG has a duration between 20 ms and 70 ms
(FIGURE 7-1A), whereas a sharp wave has a duration of 70 ms to 200 ms
(FIGURE 7-1B). Thus, the difference between spikes and sharp waves is based only
on the morphology, namely the duration, but they have the same clinical
significance and are often seen in the same EEG. When a slow wave follows the
spike or sharp wave, the discharges are defined as a spike-and-slow-wave
complex (or a sharp-wave-and-slow-wave complex). The cellular mechanisms
behind the generation of interictal epileptiform discharges are different from
those of the baseline EEG. Simultaneous intracellular and cortical surface
recordings in the penicillin-induced epilepsy model have demonstrated that a
paroxysmal depolarizing shift, rather than a postsynaptic potential, underlies the
interictal epileptiform discharges. A paroxysmal depolarizing shift consists of
a steep depolarization that triggers a series of action potentials, followed by a
plateau of continued depolarization, and then by a steep repolarization with or
without a following hyperpolarization.
On the EEG, the slopes of interictal epileptiform discharges are often
asymmetric with the initial negative component typically steeper, followed by
a slower positive component. The asymmetric morphology could be used to
differentiate an EEG artifact or a physiologic waveform with a sharply contoured
432 APRIL 2019

KEY POINTS
● The EEG signal is

generated by the summation
of extracellular
postsynaptic potentials.
● Paroxysmal depolarizing
shifts are considered the
intracellular correlate of
the interictal epileptiform
discharge on EEG
recordings.
FIGURE 7-1 ● The finding of interictal

EEG of a 29-year-old man with a history of epilepsy. A, EEG showing spike discharge with epileptiform discharges
maximal phase reversal over the left frontotemporal region (F7, T7). B, EEG showing a sharp may aid in the diagnosis
wave with maximal phase reversal over the left frontotemporal region (F7, T7). of epilepsy.
morphology. In addition, the field of interictal epileptiform discharges is

typically detected by at least a few electrodes.
Interictal epileptiform discharges occur rarely, in approximately 2% of healthy
adults. Therefore, in the appropriate clinical context, interictal epileptiform
discharges aid in the determination of epilepsy as an etiology for suspected
clinical events. However, the absence of interictal epileptiform discharges in a
routine EEG does not exclude the possibility of epilepsy. In patients with a
history of epilepsy, an initial, routine EEG detects epileptiform discharges in
approximately 50% of patients, and the yield improves to approximately 90%
with three EEG recordings. Provocation techniques, such as sleep deprivation,
hyperventilation, and photic stimulation, are used to increase the sensitivity of
detecting interictal epileptiform discharges on an EEG. Aside from
differentiating epilepsy from nonepileptic disorders, the interictal epileptiform
discharge features also aid in classifying seizure types, namely focal or
generalized seizures, and in diagnosing specific epilepsy syndromes.
SPECIFIC INTERICTAL EPILEPTIFORM DISCHARGES

Interictal epileptiform discharges are generally described under two major
categories: focal and generalized, depending on their distribution.
Focal Interictal Epileptiform Discharges

Focal epileptiform discharges occur in a focal distribution, commonly in one lobe
or region.
TEMPORAL AND FRONTAL INTERICTAL EPILEPTIFORM DISCHARGES. Temporal

lobe epilepsy is the most common focal epilepsy in adolescents and adults, and
therefore, temporal spikes or sharp waves are the most commonly observed
interictal epileptiform discharges (FIGURE 7-1). Interictal epileptiform discharges
in the frontotemporal region typically originate in the mesial temporal lobe;
whereas midtemporal and posterior temporal spikes are possibly associated with
neocortical temporal generators. Patients with unilateral temporal seizures may
show bilateral temporal interictal epileptiform discharges, presumably because
of connectivity between the temporal lobes. The presence of contralateral

temporal interictal epileptiform discharges is associated with less favorable

surgical outcomes after temporal lobectomy.4
In frontal epilepsy, interictal epileptiform discharges can be detected with
maximal negativity over frontopolar, frontotemporal, or anterior midline regions
(FIGURE 7-2). However, compared with temporal lobe epilepsy, interictal
epileptiform discharges are less frequently observed in frontal lobe epilepsy,
especially with mesial or basal frontal foci. In addition, secondary bilateral
synchrony may result in broadly distributed interictal epileptiform discharges, a
phenomenon that is more commonly observed in frontal lobe epilepsy than in
other focal epilepsies, making some frontal lobe epilepsies difficult to lateralize.5
CENTROTEMPORAL SPIKES. Centrotemporal spikes are observed in childhood

epilepsy with centrotemporal spikes, which is the most common focal epilepsy
syndrome of childhood. Seizure onset typically occurs between 3 and 14 years of
age, and most children outgrow it by 16 years of age. These children typically
have focal sensorimotor seizures that start in the face or arm during sleep and
may secondarily generalize. Spikes are observed over the centrotemporal region
unilaterally or bilaterally (FIGURE 7-3). The distribution often suggests a
horizontal dipole; the negativity is seen in the central and temporal regions,
whereas the positive end of the dipole is frontal. Not all centrotemporal spikes
are associated with clinical seizures. These spikes occur in 15% of seizure-free
siblings of children with childhood epilepsy with centrotemporal spikes, and an
autosomal dominant mode of inheritance has been suggested.
OCCIPITAL SPIKES. Occipital spikes are frequently observed in childhood occipital

epilepsy (Panayiotopoulos and Gastaut types) as well as in other types of
occipital lobe epilepsy (FIGURE 7-4). Childhood occipital epilepsy is often
associated with autonomic symptoms, such as emesis, pallor, cardiorespiratory
irregularities, ictal headache, and visual hallucinations. Occipital spikes may be
observed in patients without a
history of seizures, and this is
particularly common in the
pediatric population. In addition,
occipital spikes are seen in people
who have congenital blindness,
and, therefore, they do not always
correlate with occipital seizures.
TEMPORAL INTERMITTENT
RHYTHMIC DELTA ACTIVITY.
Temporal intermittent rhythmic
delta activity (TIRDA) refers to a
unilateral intermittent rhythmic
sinusoidal or saw-toothed
activity of 1-Hz to 4-Hz
frequency occurring over the
anterior-to-midtemporal
FIGURE 7-2
EEG of a 59-year-old man with a history of epilepsy.
derivations (FIGURE 7-5). The
Note the spike discharge with the maximal duration of the train varies,
amplitude at the right frontopolar region (Fp2). usually lasting for a few seconds.
434 APRIL 2019

FIGURE 7-3
EEG showing centrotemporal spikes recorded in a patient with childhood epilepsy with
centrotemporal spikes.
FIGURE 7-4
EEG of a 23-year-old woman with occipital lobe epilepsy showing right occipital spikes. CT of
the patient (not shown) revealed encephalomalacia over the right posterior region.

FIGURE 7-5
EEG of a 9-year-old boy showing 3-Hz spike and wave discharges typical of absence
epilepsy.
Concomitant depth and scalp electrode recordings have shown that TIRDA
correlates with intracranially recorded mesial temporal spikes and, therefore, is
considered an EEG marker of mesial temporal lobe epilepsy.6
Generalized Interictal Epileptiform Discharges

Generalized epileptiform discharges present in a bilateral symmetric fashion.
GENERALIZED SPIKE AND SLOW WAVES. Generalized spike and slow waves consist
of bilaterally synchronous spikes followed by a slow wave of high amplitude
(FIGURE 7-6). Typically, the diffuse spike and slow wave often has frontal
predominance and may occur in rhythmic runs at various frequencies. This
electrographic pattern is consistent with the diagnosis of generalized epilepsy. In
addition, several electroclinical syndromes, such as childhood absence epilepsy
and juvenile myoclonic epilepsy, are associated with generalized spike and slow
waves at different frequencies.
The 3-Hz spike-and-slow-wave discharges are the hallmark of absence
epilepsy. They often present in bursts lasting from 1 second to 3 seconds and can
be activated by hyperventilation or hypoglycemia. When the spike-and-slow-
wave discharges last longer than 6 seconds, they are often associated with
an alteration of consciousness and are, thus, considered ictal discharges
(FIGURE 7-7).
Myoclonic seizures are brief jerks of muscles associated with diffuse spike and
slow waves (FIGURE 7-8). Juvenile myoclonic epilepsy is a common generalized
epilepsy syndrome, and generalized spike-and-slow-wave discharges often occur
at a faster frequency, typically from 4 Hz to 6 Hz in this syndrome. In addition,
generalized polyspikes are also seen. Juvenile myoclonic epilepsy is often
associated with a photoparoxysmal response, whereby photic stimulation can
436 APRIL 2019

FIGURE 7-6
EEG of a 43-year-old man with epilepsy showing generalized spikes. Note the diffuse,
bilaterally symmetric spikes with anterior predominance.
FIGURE 7-7
EEG of a 29-year-old woman with absence epilepsy showing 3-Hz spike and slow wave. Blank
staring was noted in the patient when the prolonged spike and slow wave was recorded on EEG.

FIGURE 7-8
EEG of a 43-year-old man with myoclonic seizures showing diffuse spike discharges.
elicit a bilaterally synchronous spike and slow wave during or outlasting the
photic stimulation train.
Slow spike-and-wave patterns consist of generalized spike-wave or
sharp-wave discharges occurring at 1 Hz to 2.5 Hz, initially termed petit mal
variant discharges by Gibbs.7 Compared with the classic 3-Hz spike and slow
wave, the spike component is more blunted in slow-spike-and-wave discharges,
although the spatial distribution is similar. Lennox-Gastaut syndrome, a disorder
associated with multiple seizure types and intellectual disability, is commonly
associated with slow spike and slow waves on the EEG.
GENERALIZED PAROXYSMAL FAST ACTIVITY. Generalized paroxysmal fast activity

consists of bursts of generalized, sharply contoured discharges in the beta
frequency range. It is more frequently observed in non–rapid eye movement
(REM) sleep and typically lasts from 2 seconds to 4 seconds (FIGURE 7-9).
Generalized paroxysmal fast activity is traditionally associated with epileptic
encephalopathy with intellectual disability, such as in patients with
Lennox-Gastaut syndrome, although it is also occasionally seen in individuals
with generalized epilepsy and normal intelligence.8
BRIEF POTENTIALLY ICTAL RHYTHMIC DISCHARGES

Usually, a minimum duration of 10 seconds is required for a rhythmic discharge
to be considered an electrographic seizure. Brief ictal rhythmic discharges
(BIRDs) are defined as “very brief runs of focal or generalized sharply contoured
rhythmic activity greater than 4 Hz with or without evolution, that are not
consistent with any known normal or benign pattern.” Specifically, the term
BIRDs describes ictal-appearing rhythmic discharges that last less than
10 seconds.9 Although they are considered an interictal-ictal continuum pattern,
they are also associated with a high prevalence of electrographic seizures in
patients who are critically ill.10 In addition, BIRDs are also associated with
refractory epilepsy, and their location correlates with the seizure onset in patients
who are noncritically ill.9 Therefore, starting prophylactic treatment and
continuous monitoring are often warranted when BIRDs are noted on the EEG.9,10
438 APRIL 2019

FIGURE 7-9
EEG of a patient with intractable epilepsy. Note the burst of diffuse, sharply contoured
fast activity.
ICTAL DISCHARGES
An epileptic seizure is defined as “a transient occurrence of signs and/or symptoms
due to abnormal excessive or synchronous neuronal activity in the brain.”11
According to their clinical and electrographic onsets, seizures are classified as
focal, generalized, or unknown.12 A seizure is more likely to be recorded during
continuous video-EEG monitoring than during a routine EEG, and admission to
the epilepsy monitoring unit is indispensable for accurate localization of the seizure-
onset zone. Ictal patterns refer to the observed EEG changes during seizures. The ictal
discharges during seizures do not always consist of repetitive spikes or sharp waves;
instead, ictal discharges vary in morphology (eg, sharply contoured or not), frequency,
and distribution.3 Frequently, ictal discharges show evolution patterns with changes
in frequency, amplitude, field, or morphology during the course of the seizure.
Focal Seizures
Focal-onset seizures are conceptualized as originating within networks limited to
one hemisphere and may be discretely localized or more widely distributed.13
Repetitive seizures in focal epilepsy tend to have a very consistent pattern of
onset and evolution both clinically and electrographically.
Mesial temporal lobe epilepsy, which is often caused by mesial temporal
sclerosis, is the most common focal seizure disorder in adults. In approximately
80% of mesial temporal seizures, rhythmic, sharply contoured (or apiculate)
temporal theta activity appears within 30 seconds of seizure onset (FIGURE 7-10).14
An initial, regular 5-Hz to 9-Hz inferotemporal rhythm may be specific for
hippocampal-onset seizures. Such a pattern results from the synchronous
recruitment of the adjacent inferolateral temporal neocortex on intracranial
EEG studies.15 Compared with mesial temporal seizures, neocortical temporal
seizures are usually shorter in duration, less frequently associated with visceral
sensations and automatisms, and more frequently accompanied by auditory
hallucinations. Neocortical temporal seizures often evolve into unilateral clonic
activity or a generalized seizure faster than mesial temporal lobe seizures do.

FIGURE 7-10
EEG of a 29-year-old man with left temporal epilepsy. A, Semirhythmic delta discharges
evolved into sharply contoured theta activity in left temporal electrodes. B, The ictal activity
increased in frequency. C, The ictal activity evolved into spike-and-slow-wave discharges.
440 APRIL 2019

Electrographically, neocortical temporal seizure onset is more frequently
associated with irregular, polymorphic, 2-Hz to 5-Hz lateralized activity.15
Frontal lobe seizures are the second most common seizure and are further
classified as dorsolateral frontal, mesial frontal, and basal frontal epilepsy. Overall,
frontal lobe seizures are less likely to show localized ictal onsets on the EEG
(FIGURE 7-11).16 Diffuse ictal onsets, such as paroxysmal fast activity and
background attenuation, are often seen in frontal lobe seizures, especially those
of mesial frontal lobe origin, whereas localized repetitive epileptiform activity is
most common in dorsolateral frontal seizures.16 Factors contributing to the frequent
lack of EEG localization in frontal lobe epilepsy include the inaccessibility of much
of the frontal lobes to surface electrodes, the widespread anatomic connections
of the frontal lobes to other areas, and the orientation of dipoles. In addition,
hypermotor seizures are common in frontal lobe epilepsy, resulting in obscuring
artifacts.16,17
Overall, parietal lobe epilepsy is uncommon, and it represented approximately
6% of patients undergoing epilepsy surgery in one large cohort at a major
FIGURE 7-11
EEG of a 32-year-old man with intractable epilepsy. A, B, Scalp EEG failed to detect clear ictal
EEG activity during a seizure. Movement artifact was noted on the scalp EEG. Concomitant
intracranial EEG (not shown) demonstrated ictal onset over the left mesial frontal lobe.

epilepsy center.18 Scalp-recorded interictal epileptiform discharges in parietal

lobe epilepsy often show a variable field outside of the lobe of origin, such as
frontal, central, temporal, or posterior areas.18–20 Diffuse interictal findings due
to fast bilateral synchrony or a lack of interictal findings have also been
reported.18,19 Ictal findings are also variable and commonly unlocalizable on scalp
EEG studies.18,20 These findings may be due to known connections between the
parietal lobe and other brain regions. Therefore, invasive monitoring is often
required for definite localization of the seizure onset, especially in patients who
are nonlesional.18–20
Occipital lobe seizures are less common than temporal and frontal lobe
seizures. Surface ictal EEG in occipital seizures usually shows paroxysmal fast
activity or fast spiking (or both) in the occipital regions with occasional gradual
anterior propagation and generalization (FIGURE 7-12).21 A well-localized
unifocal rhythmic ictal discharge during occipital seizures is infrequent
compared with the more common bioccipital field and spread to the temporal
regions.21 In one series, approximately 40% of occipital seizures could be
localized on the EEG, and the rest either had a diffuse pattern or were falsely
localized or lateralized.16
FIGURE 7-12
EEG of a 33-year-old woman showing an occipital lobe seizure. The patient had her typical
visual hallucinations during the seizure. A, Semirhythmic spike discharges (arrow) were noted
over the left occipital region. B, Ictal discharges increased in rhythmicity and amplitude (arrow).
442 APRIL 2019

Generalized Seizures KEY POINTS
Generalized seizures are defined as seizures “originating at some point within, and
● Ictal discharges seen on
rapidly engaging, bilaterally distributed networks. Such bilateral networks can EEG in patients with epilepsy
include cortical and subcortical structures, but do not necessarily include the entire vary in morphology,
cortex.”13 In generalized seizures, ictal EEG changes are not lateralizing and indicate frequency, and distribution
bilateral brain involvement, although asymmetric onsets can be seen at times. and often show pattern
evolutions.
Spike-and-Slow-Wave Discharge ● Seizures are classified as

As mentioned previously, 3-Hz spike-and-slow-wave discharges are the hallmark of focal-onset or generalized-
absence seizures. Prolonged runs of spike-and-slow-wave discharges correlated onset seizures.
with such clinical changes as staring and unresponsiveness are considered seizures.
However, because of the difficulty of assessing a subtle, brief alteration of awareness,
the distinction between interictal and ictal events is not straightforward. The ictal
pattern of an absence seizure typically shows a change of frequency starting at
approximately 3.5 Hz and ending at approximately 2.5 Hz (FIGURE 7-7).
Runs of spike and slow waves are also found in juvenile myoclonic
epilepsy but with faster frequencies of 4 Hz to 6 Hz. When individual
spikes are accompanied by myoclonic jerking, they are also considered ictal
discharges (FIGURE 7-8). Aside from runs of spike-and-slow-wave discharges,
ictal discharges in juvenile myoclonic epilepsy also occur as bursts of
polyspikes, and the spike voltage may increase from the first spike to the last.
Electrodecrement Pattern
This ictal pattern is characterized by a sudden attenuation of amplitude for a few
seconds. The electrodecrement pattern may be preceded by a diffuse sharp and
slow wave. Generalized, low-voltage fast activity is often superimposed on the
electrodecrement. This pattern is mostly observed in tonic and atonic seizures, as
well as in epileptic spasms in infants (FIGURE 7-13).
PERIODIC DISCHARGES
According to the standardized critical care EEG terminology of the American Clinical
Neurophysiology Society, periodic discharges are defined as regularly recurring
FIGURE 7-13
EEG of a 73-year-old man showing a tonic seizure. Note the generalized slow waves followed
by background attenuation with a superimposed muscle artifact.

FIGURE 7-14
EEG of a 66-year-old woman with new-onset status epilepticus showing right-sided
lateralized periodic discharges.
waveforms with uniform morphology and duration, and with a quantifiable

interdischarge interval.22 The discharges have no more than three phases, ie, they
cross the isoelectric line no more than twice, and last no longer than 0.5 seconds,
regardless of the number of phases.22 Periodic discharges are further classified as
generalized periodic discharges (GPDs), lateralized periodic discharges (LPDs), and
bilateral independent periodic discharges. GPDs are bisynchronous and symmetric,
even if their field is restricted (eg, bifrontally). LPDs refer to unilateral or bilateral
patterns that are clearly asymmetric.22 Features that favor an ictal nature are
described as “periodic discharges plus.” For example, periodic discharges plus faster
activity are those with superimposed faster rhythms in the theta range or faster.22
LPDs, previously known as periodic lateralized epileptiform discharges (PLEDs),
are commonly seen in acute, destructive brain injury, such as herpes encephalitis
and stroke (FIGURE 7-14).23 LPDs can also emerge after focal status epilepticus.
LPDs are transient phenomena that rarely persist for more than a few weeks.
GPDs are seen in patients with systemic metabolic derangements, postanoxia,
or infections.24 Frequent GPDs are seen in postanoxic brain injury, and, when
associated with myoclonus, represent the electrographic correlate of myoclonic
status epilepticus (FIGURE 7-15).
Aside from capturing seizures, the EEG can be an important tool in evaluating
patients after cardiac arrest.25 Periodic discharges on a suppressed background,
together with complete background suppression or burst suppression, represent
a highly malignant pattern. Malignant EEG features include abundant periodic
discharges, rhythmic spike and slow waves, discontinuous or low-voltage
background, and reversed anterior-posterior gradient of nonreactive EEG. One
malignant EEG feature predicts a poor prognosis almost one-half (48%) of the
time, whereas two malignant EEG features predict poor prognosis in 96% of
cases.25 For further information about the issues involved in prognostication after
cardiac arrest, refer to the December 2018 Continuum issue on neurocritical care.26
While they most frequently represent an interictal phenomenon, periodic
discharges can occasionally be an ictal pattern. In epilepsia partialis continua,
periodic discharges are oftentimes locked to muscle twitches, and, therefore,
they are considered ictal discharges. Unified criteria for nonconvulsive status
epilepticus (TABLE 7-127) have been proposed to differentiate the ictal versus
444 APRIL 2019

FIGURE 7-15
EEG of a 59-year-old woman showing generalized periodic discharges after hypoxia that
were associated with shoulder twitching.
interictal nature of LPDs. These criteria rely on the frequency and the evolution
of the discharges and the clinical correlates to differentiate ictal from interictal
phenomena, and they may not necessarily be sensitive. Thus, a pattern that does not
fulfill these criteria could still be ictal, although this cannot be proven based on the
EEG pattern alone.28 Some authors view periodic discharges as an interictal-ictal
continuum and suggest evaluating the likelihood of neuronal injuries from periodic
discharges in individual clinical settings.28 CASE 7-1 discusses a patient with
nonconvulsive status epilepticus and the EEG finding of periodic discharges.
Triphasic Waves
Triphasic waves are high-amplitude positive sharp transients that are preceded
and followed by negative waves of relatively lower amplitude. They are referred
Working Clinical Criteria for Nonconvulsive Status Epilepticusa TABLE 7-1
Patients Without Known Epileptic Encephalopathy

◆ Epileptiform discharges >2.5 Hz or
◆ Epileptiform discharges ≤2.5 Hz or rhythmic delta/theta activity (>0.5 Hz) and one of
the following:
◇ EEG and clinical improvement after IV antiepileptic drug or
◇ Subtle clinical ictal phenomena during the EEG patterns mentioned above or
◇ Typical spatiotemporal evolution
Patients With Known Epileptic Encephalopathy
◆ Increase in prominence or frequency of the features mentioned above, when compared to
baseline with observable change in clinical state
◆ Improvement of clinical and EEG features with IV antiepileptic drugs
EEG = electroencephalography; IV = intravenous.

a
Reprinted with permission from Beniczky S, et al, Epilepsia.27 © 2013 International League Against Epilepsy.

CASE 7-1 A 58-year-old woman was admitted to the hospital for new-onset
seizures manifesting as jerking movements of all extremities.
Levetiracetam was administered and maintained at a dose of 1500 mg
every 12 hours. Her jerking movements resolved, but her mental status
continued to be altered.
On examination, she did not answer any questions and followed some
commands only intermittently. Her brain MRI showed increased T2 signal
and restricted diffusion in the medial frontal cortices. Her continuous
video-EEG monitoring revealed frequent runs of periodic (approximately
1 Hz), sharply contoured bifrontal discharges, lasting longer than
10 seconds at a time. In addition, she had an average of 10 seizures per
hour, during which her periodic discharges became more sharply
contoured and increased in rhythmicity and frequency (FIGURE 7-16). No
discernible clinical correlates were present aside from her altered
mental status.
The diagnosis of nonconvulsive status epilepticus was confirmed, and
phenytoin was added with subsequent resolution of the nonconvulsive
status epilepticus and gradual improvement of the patient’s mental
status over the following 4 days.
COMMENT This is a typical case of nonconvulsive status epilepticus. Without an EEG,

identifying the etiology of the patient’s altered awareness would have
been difficult. The EEG helps differentiate nonconvulsive status
epilepticus from a prolonged postictal confusion or other encephalopathy.
Early recognition and prompt treatment of nonconvulsive status
epilepticus are crucial to avoid further neuronal damage and more
refractory seizures. The EEG also guides treatment of nonconvulsive status
epilepticus, avoiding undertreatment or overtreatment with antiseizure
medications.
446 APRIL 2019

FIGURE 7-16
EEG of the patient in CASE 7-1. A, Note the semirhythmic bifrontal sharply contoured
discharges. B, C, The evolution of the seizure is shown; ictal activity increased in frequency
and rhythmicity and became more sharply contoured. Frequent seizures were recorded,
and the patient remained with altered mental status consistent with nonconvulsive status
epilepticus.

to as GPDs with triphasic morphology (FIGURE 7-17),22 although they can occur
singly or unilaterally. Triphasic waves previously were considered the hallmark
of hepatic encephalopathy, but they proved to be a nonspecific feature of any
cause of encephalopathy, including toxic, metabolic derangement, or structural
changes. Triphasic waves may be occasionally considered epileptic discharges in
certain clinical circumstances, and some authors even associate frequent triphasic
waves with nonconvulsive status epilepticus. Noxious or auditory stimulation
frequently increases triphasic waves in diffuse encephalopathy, whereas ictal
patterns in nonconvulsive status epilepticus are nonreactive.29 Unfortunately, all
proposed criteria to differentiate epileptic triphasic waves from nonepileptic
triphasic waves have not been validated in prospective studies.30
INTRACRANIAL RECORDINGS
Intracranial EEG recordings are widely used for seizure localization and
functional mapping as part of a patient’s presurgical evaluation. These recordings
circumvent signal attenuation caused by the filtering effect of the skull and skin
in a surface EEG; therefore, intracranial electrode–recorded spikes may be
indiscernible on scalp EEG studies. Intracranial electrodes are in direct contact
with brain tissue, including deep brain structures, thus providing great spatial
resolution. In addition, intracranial EEG signals are not susceptible to
electromyographic artifacts, which offers a major advantage, especially with
hypermotor seizures. However, intracranial EEG electrodes can be implanted only
in a confined region, necessitating that electrode implantation should be based on
solid presurgical hypotheses regarding the likely localization of the seizure focus.
Different invasive monitoring techniques, including subdural electrodes
(grids or strips) and stereo-EEG, are available, but each have distinct advantages
and limitations. Stereo-EEG implantation is less invasive than subdural grid
placement, which requires a craniotomy, and stereo-EEG can explore deeper
brain regions. However, stereo-EEG offers a relatively limited spatial sampling of
the cortical surface, and subdural grids can facilitate a more detailed mapping of
eloquent cortex.31
FIGURE 7-17
EEG of a 58-year-old woman showing triphasic waves. She was admitted for altered mental
status and was also noted to have acute kidney injury and metabolic acidosis on admission.
448 APRIL 2019

Interictal Spikes and High-Frequency Oscillations KEY POINTS
The cortical region displaying interictal spikes is defined as the irritative zone.32
● Periodic discharges can
Not all spikes have the same clinical significance. Interictal spikes that are high be considered as ictal or
amplitude, occur in a rhythmic or repetitive fashion, or occur in an abnormal interictal phenomena in
background are more likely associated with an epileptogenic process.32 various clinical settings.
High-frequency oscillations are spontaneous EEG transients in the
● Continuous EEG
frequency range of 80 Hz to 500 Hz that clearly stand out from the background
monitoring is important in
activity.33,34 High-frequency oscillations are subdivided into ripples (80 Hz to the diagnosis and treatment
250 Hz) and fast ripples (>250 Hz).33,34 They are typically recorded in an of nonconvulsive status
intracranial EEG, although they may be seen in an unfiltered scalp EEG. epilepticus.
High-frequency oscillations can be superimposed on spikes or present
independently.35
High-frequency oscillations are emerging as a promising biomarker of
epileptogenic tissue.33,34 Some studies have shown that fast ripples are more
abundant and have a higher power (defined as the square of the amplitude
divided by the frequency) in the seizure-onset zone.35,36 Studies have shown that
resection of areas that show high-frequency oscillations (both ripples and fast
ripples) is associated with greater chances of seizure freedom.34,37 Furthermore,
residual fast ripples in the postresection electrocorticogram are prognostic
markers for seizure recurrence, especially if they are abundant. Therefore, van’t
Klooster and colleagues38 have suggested careful consideration of high-frequency
oscillations when tailoring the resection borders in epilepsy surgery.
Ictal Findings
The epileptogenic zone is defined as the brain region whose removal or
disconnection is necessary and sufficient for seizure freedom.32 The
epileptogenic zone can be validated only when seizure freedom is achieved after
epilepsy surgery. However, the seizure-onset zone, which refers to the cortical
area from which clinical seizures originate, can be identified on EEG during
monitoring. Intracranially recorded seizures differ from those recorded by a
scalp EEG. A seizure appears on a scalp EEG only after the ictal discharge
synchronizes over a relatively large brain area. Therefore, seizures have often
spread outside of their onset zones by the time they are detected on a scalp EEG.
Intracranial contacts enable early detection of seizure discharges if appropriately
placed in the seizure-onset zone, thus providing earlier detection of seizure
onset. For example, ictal discharges confined in the hippocampus produce no
scalp EEG rhythms. Frequently, synchronous recruitment of the adjacent
neocortical inferotemporal cortex is required to generate ictal theta activity on
scalp EEG recordings in mesial temporal epilepsy.39
Previously, intracranial ictal patterns were defined as sustained and rhythmic
EEG patterns with frequencies greater than 2 Hz that are state independent and
distinct from the background.40 However, several other ictal patterns have been
reported, including low-voltage fast activity, low-frequency high-amplitude
periodic spikes, sharp activity at 13 Hz or less, spike-and-wave activity, bursts of
high-amplitude polyspikes, burst-suppression pattern, and delta brushes.41 Most
intracranial EEG seizure-onset patterns have been found across multiple
epileptogenic lesions, suggesting that different pathologic substrates can affect
similar networks and produce similar mechanisms underlying seizure
generation. Compared with other patterns, low-voltage fast activity is associated
with a larger seizure-onset zone (CASE 7-2).41

CASE 7-2 A 48-year-old woman with a history of uncontrolled seizures was

referred for evaluation. The patient reported seizure onset at 13 years of
age. Her typical seizures started with “a strange feeling in the stomach,”
followed by alteration of consciousness. Her seizures occurred an
average of once per month despite treatment with phenytoin and
levetiracetam.
Her brain MRI showed right-sided mesial temporal sclerosis and a
9-mm meningioma along the paramedian right parietal lobe. Video-EEG
monitoring with scalp electrodes captured interictal epileptiform
discharges over bilateral temporal regions, more frequently on the left.
Typical focal seizures with altered awareness were also recorded with
bilateral ictal onsets that appeared maximal over the temporal regions.
Intracranial EEG was then obtained with depth electrodes placed
stereotactically in bilateral hippocampi, amygdalae, temporal neocortex,
and the orbitofrontal and insular regions, as well as the right posterior
cingulate area. Spikes were recorded by bilateral hippocampal and
amygdalar electrodes (FIGURE 7-18A). Two types of seizures were captured.
One of her seizure types occurred without alteration of awareness and
consisted of auras of “a strange feeling” in her stomach. On stereo-EEG,
repetitive spikes were seen over the left hippocampus, although the
scalp EEG failed to show any ictal changes (FIGURE 7-18B). Her other seizure
type was associated with an alteration of awareness and consisted of
behavioral arrest, staring, and unresponsiveness. At seizure onset, broad
electrodecrement over the right temporal region was appreciated. This
was then quickly followed by high-amplitude spike-and-slow-wave
discharges over the left mesial temporal region before spreading
diffusely (FIGURE 7-18C).
COMMENT This patient had frequent seizures despite two appropriate antiseizure
medications, which met the criteria of refractory epilepsy, thus
necessitating a surgical evaluation. Her inconclusive scalp EEG results led
to bilateral implantation of intraparenchymal electrodes for better
lateralization and localization of the seizure focus. Her intracranial EEG
identified ictal onsets in the left hippocampus when the concomitant scalp
EEG failed to detect any ictal discharges in one of her seizure types. In
her other seizure type, ictal EEG findings were not manifest on the scalp
EEG before major propagation. The patient was offered responsive
neurostimulation of both hippocampi considering her bilateral
hippocampal ictal onsets.
450 APRIL 2019

FIGURE 7-18
EEG of the patient in CASE 7-2. A, Intracranial EEG showing spikes over the left hippocampus as
well as the right hippocampus and amygdala (top panel). Concomitant scalp EEG (bottom panel)
failed to show spike discharges. B, Intracranial EEG showing one type of seizure. The ictal
discharge that is confined in the hippocampus produced no scalp EEG rhythms (not shown).
C, EEG showing another type of seizure. Note the clear ictal onset (arrow) on intracranial
EEG (top panel). The concomitant scalp EEG (bottom panel) showed diffuse ictal onset
(not shown).

KEY POINTS
CONCLUSION
● Intracranial electrodes EEG remains a key diagnostic tool for epilepsy, and interictal or ictal findings are
allow for earlier seizure often identified in the study. However, the absence of interictal epileptiform
detection and spatial discharges or ictal findings cannot rule out the possibility of epilepsy. Seizures are
resolution than do scalp
classified as focal-onset or generalized-onset seizures. They may vary in
recordings.
electrographic patterns, and ictal activity usually evolves during a seizure.
● Intracranial monitoring Continuous EEG monitoring is important in the diagnosis and treatment of
is an invaluable tool in nonconvulsive status epilepticus. Intracranial EEG monitoring often allows for
surgical planning. earlier seizure detection and better spatial resolution than scalp recordings, and it is
invaluable for planning surgery in patients with inconclusive scalp EEG findings.
REFERENCES
1 Buzsáki G, Anastassiou CA, Koch C. The origin of 11 Fisher RS, van Emde Boas W, Blume W, et al.
extracellular fields and currents—EEG, ECoG, LFP Epileptic seizures and epilepsy: definitions
and spikes. Nat Rev Neurosci 2012;13(6):407–420. proposed by the International League Against
doi:10.1038/nrn3241. Epilepsy (ILAE) and the International Bureau for
Epilepsy (IBE). Epilepsia 2005;46(4):470–472.
2 Tao JX, Ray A, Hawes-Ebersole S, Ebersole JS.
doi:10.1111/j.0013-9580.2005.66104.x.
Intracranial EEG substrates of scalp EEG interictal
spikes. Epilepsia 2005;46(5):669–676. doi:10.1111/ 12 Fisher RS, Cross JH, D’Souza C, et al. Instruction
j.1528-1167.2005.11404.x. manual for the ILAE 2017 operational classification
of seizure types. Epilepsia 2017;58(4):531–542.
3 Noachtar S, Binnie C, Ebersole J, et al. A glossary
doi:10.1111/epi.13671.
of terms most commonly used by clinical
electroencephalographers and proposal for 13 Berg AT, Berkovic SF, Brodie MJ, et al. Revised
the report form for the EEG findings. terminology and concepts for organization of
Electroencephalogr Clin Neurophysiol Suppl seizures and epilepsies: report of the ILAE
1999;52:21–41. Commission on Classification and Terminology,
2005-2009. Epilepsia 2010;51(4):676–685.
4 Schulz R, Lüders HO, Hoppe M, et al. Interictal
doi:10.1111/j.1528-1167.2010.02522.x.
EEG and ictal scalp EEG propagation are highly
predictive of surgical outcome in mesial 14 Risinger MW, Engel J Jr, Van Ness PC, et al. Ictal
temporal lobe epilepsy. Epilepsia 2000;41(5): localization of temporal lobe seizures with
564–570. doi:10.1111/j.1528-1157.2000.tb00210.x. scalp/sphenoidal recordings. Neurology 1989;
39(10):1288–1293. doi:10.1212/WNL.39.10.1288.
5 Beleza P, Pinho J. Frontal lobe epilepsy. J Clin
Neurosci 2011;18(5):593–600. doi:10.1016/ 15 Ebersole JS, Pacia SV. Localization of temporal
j.jocn.2010.08.018. lobe foci by ictal EEG patterns. Epilepsia 1996;37(4):
386–399. doi:10.1111/j.1528-1157.1996.tb00577.x.
6 Di Gennaro G, Quarato PP, Onorati P, et al.
Localizing significance of temporal intermittent 16 Foldvary N, Klem G, Hammel J, et al. The localizing
rhythmic delta activity (TIRDA) in drug-resistant value of ictal EEG in focal epilepsy. Neurology
focal epilepsy. Clin Neurophysiol 2003;114(1): 2001;57(11):2022–2028. doi:10.1212/WNL.57.11.2022.
70–78. doi:10.1016/S1388-2457(02)00332-2.
17 Unnwongse K, Wehner T, Foldvary-Schaefer N.
7 Gibbs FA. Petit mal variant revisited. Epilepsia Mesial frontal lobe epilepsy. J Clin Neurophysiol
1971;12(1):89–96. 2012;29(5):371–378. doi:10.1097/WNP.
0b013e31826b3c60.
8 Sagi V, Kim I, Bhatt AB, et al. Generalized
paroxysmal fast activity in EEG: an unrecognized 18 Salanova V, Andermann F, Rasmussen T, et al.
finding in genetic generalized epilepsy. Epilepsy Parietal lobe epilepsy. Clinical manifestations
Behav 2017;76:101–104. doi:10.1016/j.yebeh. and outcome in 82 patients treated surgically
2017.08.019. between 1929 and 1988. Brain 1995;118(pt 3):
607–627. doi:10.1093/brain/118.3.607.
9 Yoo JY, Marcuse LV, Fields MC, et al. Brief
potentially ictal rhythmic discharges [B(I)RDs] in 19 Salanova V. Parietal lobe epilepsy. J Clin
noncritically ill adults. J Clin Neurophysiol 2017; Neurophysiol 2012;29(5):392–396. doi:10.1097/
34(3):222–229. doi:10.1097/WNP.0000000000000357. WNP.0b013e31826c9ebc.
10 Yoo JY, Rampal N, Petroff OA, et al. Brief 20 Ristić AJ, Alexopoulos AV, So N, et al. Parietal
potentially ictal rhythmic discharges in critically lobe epilepsy: the great imitator among focal
ill adults. JAMA Neurol 2014;71(4):454–462. epilepsies. Epileptic Disord 2012;14(1):22–31.
doi:10.1001/jamaneurol.2013.6238. doi:10.1684/epd.2012.0484.
452 APRIL 2019

21 Adcock JE, Panayiotopoulos CP. Occipital lobe 32 Lüders HO. Textbook of epilepsy surgery. Boca
seizures and epilepsies. J Clin Neurophysiol 2012; Raton, FL: Informa Healthcare, 2008.
29(5):397–407. doi:10.1097/WNP.0b013e31826c98fe.
33 Zijlmans M, Jiruska P, Zelmann R, et al.
22 Hirsch LJ, LaRoche SM, Gaspard N, et al. High-frequency oscillations as a new biomarker
American Clinical Neurophysiology Society’s in epilepsy. Ann Neurol 2012;71(2):169–178.
standardized critical care EEG terminology: 2012 doi:10.1002/ana.22548.
version. J Clin Neurophysiol 2013;30(1):1–27.
34 Frauscher B, Bartolomei F, Kobayashi K, et al.
doi:10.1097/WNP.0b013e3182784729.
High-frequency oscillations: the state of clinical
23 Pohlmann-Eden B, Hoch DB, Cochius JI, research. Epilepsia 2017;58(8):1316–1329.
Chiappa KH. Periodic lateralized epileptiform doi:10.1111/epi.13829.
discharges–a critical review. J Clin Neurophysiol
35 Urrestarazu E, Chander R, Dubeau F, Gotman J.
1996;13(6):519–530.
Interictal high-frequency oscillations (100-500 Hz)
24 van Putten MJ, Hofmeijer J. Generalized periodic in the intracerebral EEG of epileptic patients. Brain
discharges: pathophysiology and clinical 2007;130(pt 9):2354–2366. doi:10.1093/brain/awm149.
considerations. Epilepsy Behav 2015;49:228–233.
36 Jacobs J, LeVan P, Chander R, et al. Interictal
doi:10.1016/j.yebeh.2015.04.007.
high-frequency oscillations (80-500 Hz) are an
25 Westhall E, Rossetti AO, van Rootselaar AF. indicator of seizure onset areas independent of
Standardized EEG interpretation accurately spikes in the human epileptic brain. Epilepsia
predicts prognosis after cardiac arrest. 2008;49(11):1893–1907. doi:10.1111/j.1528-1167.
Neurology 2016;86(16):1482–1490. doi:10.1212/ 2008.01656.x.
WNL.0000000000002462.
37 Cho JR, Koo DL, Joo EY, et al. Resection of
26 Seder DB. Management of comatose survivors of individually identified high-rate high-frequency
cardiac arrest. Continuum 2018;24(6):1732–1752. oscillations region is associated with favorable
doi:10.1212/CON.0000000000000669. outcome in neocortical epilepsy. Epilepsia 2014;
55(11):1872–1883. doi:10.1111/epi.12808.
27 Beniczky S, Hirsch LJ, Kaplan PW, et al. Unified
EEG terminology and criteria for nonconvulsive 38 van’t Klooster MA, van Klink NEC, Leijten FSS,
status epilepticus. Epilepsia 2013;54(suppl 6): et al. Residual fast ripples in the intraoperative
28–29. doi:10.1111/epi.12270. corticogram predict epilepsy surgery outcome.
Neurology 2015;85(2):120–128. doi:10.1212/
28 Chong DJ, Hirsch LJ. Which EEG patterns warrant
WNL.0000000000001727.
treatment in the critically ill? Reviewing the
evidence for treatment of periodic epileptiform 39 Pacia SV, Ebersole JS. Intracranial EEG substrates
discharges and related patterns. J Clin of scalp ictal patterns from temporal lobe
Neurophysiol 2005;22(2):79–91. doi:10.1097/ foci. Epilepsia 1997;38(6):642–654. doi:10.1111/
01.WNP.0000158699.78529.AF. j.1528-1157.1997.tb01233.x.
29 Boulanger JM, Deacon C, Lécuyer D, et al. 40 Spencer SS, Guimaraes P, Katz A, et al.
Triphasic waves versus nonconvulsive status Morphological patterns of seizures recorded
epilepticus: EEG distinction. Can J Neurol Sci intracranially. Epilepsia 1992;33(3):537–545.
2006;33(2):175–180. doi:10.1111/j.1528-1157.1992.tb01706.x.
30 Trinka E, Leitinger M. Which EEG patterns in coma are 41 Perucca P, Dubeau F, Gotman J. Intracranial
nonconvulsive status epilepticus? Epilepsy Behav electroencephalographic seizure-onset
2015;49:203–222. doi:10.1016/j.yebeh.2015.05.005. patterns: effect of underlying pathology. Brain
2014;137(pt 1):183–196. doi:10.1093/brain/awt299.
31 Kovac S, Vakharia VN, Scott C, Diehl B. Invasive
epilepsy surgery evaluation. Seizure 2017;44:
125–136. doi:10.1016/j.seizure.2016.10.016.

REVIEW ARTICLE

Epilepsy Emergencies:
ONLINE
Status Epilepticus, Acute
CITE AS:
Repetitive Seizures, and
2019;25(2, EPILEPSY):454–476.
Autoimmune Encephalitis
Address correspondence to By Stephen VanHaerents, MD; Elizabeth E. Gerard, MD
Dr Stephen VanHaerents, 675 N
St Clair, Ste 7-112, Chicago, IL
60611, svanhaer@gmail.com.
Dr VanHaerents has received
ABSTRACT
research/grant support from the PURPOSE OF REVIEW: Thisarticle reviews epilepsy emergencies, including
Citizens United for Research in status epilepticus, acute repetitive seizures, autoimmune encephalitis, and
Epilepsy, National Institute of
Continued on page 476
the current perspective on their diagnosis and treatment.
UNLABELED USE OF
RECENT FINDINGS: Recent guidelines on the treatment of status epilepticus
USE DISCLOSURE: from the Neurocritical Care Society in 2012 and the American Epilepsy
Drs VanHaerents and Gerard Society in 2016 highlight areas of consensus in the treatment of status
discuss the unlabeled/
investigational use of
epilepticus as well as areas of uncertainty. The TRENdS (Treatment of
immunosuppressant Recurrent Electrographic Nonconvulsive Seizures) trial is the first
medications for the treatment prospective randomized clinical trial to evaluate the efficacy of IV
of autoimmune encephalitis
(cyclophosphamide, IV antiseizure medications in controlling nonconvulsive seizures on
immunoglobulin, IV continuous EEG. It demonstrated that IV lacosamide is noninferior to
methylprednisolone/ fosphenytoin in this setting. Autoimmune encephalitis is an increasingly
corticosteroids, plasma
exchange, and rituximab) and recognized cause of new-onset seizures or status epilepticus. Recently
the pharmacologic and described scoring systems, the Antibody Prevalence in Epilepsy score and
nonpharmacologic therapies for
the treatment of seizures and
the Response to Immunotherapy in Epilepsy score, can help in the
status epilepticus (convulsive assessment of autoimmune encephalitis.
and nonconvulsive), which
include diazepam, fosphenytoin/
SUMMARY: Status epilepticus, acute repetitive seizures, and autoimmune
phenytoin, levetiracetam,
lorazepam, midazolam, encephalitis are neurologic emergencies. For all these conditions, rapid
phenobarbital, and valproate and appropriate treatment may influence patient prognosis and mitigate
sodium/valproic acid. Drs
VanHaerents and Gerard discuss neuronal injury. For convulsive status epilepticus, there is reasonable
the unlabeled/investigational consensus on the initial steps that need to be taken. There is less
use of several agents for the
agreement about the management of acute repetitive seizures and
treatment of refractory and super-
refractory status epilepticus, nonconvulsive status epilepticus. An increasingly recognized etiology of
which include corticosteroids/ status epilepticus is autoimmune encephalitis, which may not be as rare as
methylprednisolone,
electroconvulsive therapy,
previously thought.
hypothermia, isoflurane, IV
immunoglobulin, ketamine,
ketogenic diet, midazolam,
pentobarbital, propofol, INTRODUCTION
S
thiopental, transcranial tatus epilepticus is an epilepsy emergency with a clear time-
magnetic stimulation, and dependent relationship to morbidity and risk of mortality. However,
vagal nerve stimulation.
status epilepticus is not a single entity; it has a number of different
forms and a vast array of etiologies. This review will focus on how to
of Neurology. approach status epilepticus, beginning with the most unambiguous
454 APRIL 2019

example of status epilepticus, convulsive status epilepticus, and then moving on KEY POINTS
to less clear-cut forms, with repetitive seizures and nonconvulsive status
● Currently, the most
epilepticus. Etiologies will also be discussed, with a special emphasis on commonly accepted
autoimmune encephalitis. definition for convulsive
status epilepticus is either
DEFINITIONS 5 minutes or more of
continuous seizure activity
A number of definitions exist for status epilepticus, which is, in part, because of
or two or more discrete
the limited understanding of its pathophysiologic mechanisms. One simplified seizures between which
way to think of status epilepticus is convulsive or nonconvulsive. The problem there is incomplete recovery
with this distinction is that motor activity with seizures is evolving and dynamic, of consciousness.
making it challenging to completely separate the two entities. This difficulty has
● Refractory status
led to a number of different definitions and guidelines. epilepticus refers to either
A common definition for convulsive status epilepticus is 30 minutes of clinical or electrographic
continuous seizure activity or two or more sequential seizures without full seizures that persist after
recovery of consciousness between the seizures.1 This duration was based on the adequate doses of an initial
benzodiazepine and an
time required to demonstrate neuronal injury in animal models. The definition acceptable second-line
has more recently been revised to incorporate considerations in patient antiseizure drug.
management. Currently, the most commonly accepted definition for convulsive
status epilepticus is either 5 minutes or more of continuous seizure activity or two ● Super-refractory status
epilepticus is defined as
or more discrete seizures between which there is incomplete recovery
status epilepticus that
of consciousness.2 continues or recurs 24 hours
As opposed to convulsive status epilepticus, definitions of nonconvulsive or more after the initiation of
status epilepticus are more controversial. One common definition of anesthetic therapy.
nonconvulsive status epilepticus is a range of conditions in which greater than
30 minutes of recurrent electrographic seizure activity results in nonconvulsive
clinical symptoms.3 Another definition is when electrographic seizure activity
persists for greater than 30 minutes in the absence of visible convulsion.4 With
these varying definitions, a unified definition of status epilepticus proposed by
the Neurocritical Care Society is defined as 5 minutes or longer of continuous
clinical and/or electrographic seizure activity or recurrent seizure activity
without recovery between seizures.5
More recently, further definitions have arisen based on the response to
antiseizure drugs. Refractory status epilepticus refers to either clinical or
electrographic seizures that persist after adequate doses of an initial
benzodiazepine and an acceptable second-line antiseizure drug.6 In super-refractory
status epilepticus, seizures continue to recur 24hours or more after the onset of
anesthetic therapy.7 Of note, definitions of refractory and super-refractory status
epilepticus can apply to either convulsive or nonconvulsive status epilepticus.
TREATMENT OF STATUS EPILEPTICUS

Because status epilepticus is not a single entity and has numerous forms, a vast
array of etiologies, and multiple definitions, one unified treatment algorithm is
not possible. Consequently, treatment should be tailored to both the type and
etiology of status epilepticus. FIGURE 8-1 introduces a simplified framework for
different forms of status epilepticus. It is important to note that, during
prolonged convulsive status epilepticus, the body is often unable to continue to
produce intense motor activity; seizures may continue without overt clinical
signs, although they may generate more subtle jerks of the face, eyes, and
extremities.8 This is typically referred to as postconvulsive nonconvulsive status
epilepticus and is thought to be a continuation of convulsive status epilepticus and

EPILEPSY EMERGENCIES
KEY POINT
● Benzodiazepine therapy
has been well established
as the first-line treatment
for convulsive status
epilepticus.
FIGURE 8-1
Simplified clinical framework for status epilepticus. The purple box represents all forms of
status epilepticus, which can be subdivided into convulsive status epilepticus and
nonconvulsive status epilepticus. It is generally regarded that postconvulsive nonconvulsive
status epilepticus should be treated similarly to convulsive status epilepticus (in orange
boxes). Blue boxes represent more controversial treatment paradigms for nonconvulsive
status epilepticus in patients who are critically ill and in patients who are ambulatory.
to cause continued neuronal injury.8 The article begins with the clinical approach
to convulsive status epilepticus and postconvulsive nonconvulsive status
epilepticus because this has the most clearly defined treatment approaches (in
orange in FIGURE 8-1). Next, more controversial treatment paradigms for
nonconvulsive status epilepticus (in light blue in FIGURE 8-1) will be discussed.
CONVULSIVE STATUS EPILEPTICUS

Ultimately, the goal of treatment for convulsive status epilepticus is to achieve
seizure control as quickly and safely as possible. Unfortunately, even for
convulsive status epilepticus, there are limited rigorous randomized controlled
clinical trials, and therefore, treatment algorithms are based largely on expert
opinion. FIGURE 8-2 compares two common treatment algorithms from the
Neurocritical Care Society in 2012 and the American Epilepsy Society in 2016.5,9
Initial steps should always be to stabilize the patient with special attention to
airway, breathing, and circulation. The physician should administer oxygen
and secure the airway as needed, initiate ECG monitoring, perform finger-stick
glucose, and screen for any immediate life-threatening causes such as meningitis
and intracranial mass lesions. In addition, the physician should attempt IV
access and send blood for electrolytes, hematology, and toxicology screen.5,9
Benzodiazepine therapy has been well established as the first-line treatment for
convulsive status epilepticus. In the US Department of Veterans Affairs
Cooperative Study, a randomized controlled clinical trial of initial IV abortive
therapies for convulsive status epilepticus, IV lorazepam 0.1 mg/kg given at a
rate of 2 mg/min, was superior to IV phenytoin. Another double-blind study
compared IV diazepam 10 mg versus IV lorazepam 4 mg and demonstrated equal
efficacy in seizure termination (76% versus 89%).11 This is helpful when
456 APRIL 2019

FIGURE 8-2
Comparison of treatment algorithms for convulsive status epilepticus from the Neurocritical
Care Society and the American Epilepsy Society.
AED = antiepileptic drug; EEG = electroencephalogram; IM = intramuscular; IV = intravenous.
Reprinted with permission from Brophy GM, et al, Neurocrit Care,5 © 2012 Springer Science+Business Media;
and Glauser T, et al, Epilepsy Curr,9 © 2016 American Epilepsy Society.

managing seizures in the hospital, but the reality is that often status epilepticus
may occur outside of the hospital or at times when there is no IV access. This
was addressed more recently in RAMPART (Rapid Anticonvulsant Medication
Prior to Arrival Trial). In this trial, 10 mg of IM midazolam was found to be
superior to that of 4 mg of IV lorazepam, presumably because of the rate at which
the medication could be administered because it did not require placement
of an IV catheter.12 Of note, one limitation to this study was that the 4-mg dose
of IV lorazepam was approximately half of the typically recommended dose of
0.1 mg/kg in the US Department of Veterans Affairs Cooperative Study.10
Although initial therapy with a benzodiazepine is generally agreed on, when it
should be given is variable. The Neurocritical Care Society 2012 guidelines on
the treatment of status epilepticus recommend immediate IV access and
administration of a benzodiazepine within 0 to 5 minutes.5 Conversely, the
American Epilepsy Society 2016 guidelines recommend giving first-line
benzodiazepine therapy in the first 5 to 20 minutes (ie, once seizures last
5 minutes or more).9
After first-line benzodiazepine therapy, second-line therapy options have
little evidence that any one option is superior to another, and treatment
recommendations are based largely on expert opinion. Historically, phenytoin
and the prodrug fosphenytoin have been the most commonly used second-line
agents based on consensus and without randomized trials to support their
efficacy. Fosphenytoin is typically preferred over phenytoin because it is better
tolerated and because it can be infused more quickly with a decreased risk of
hypotension or infusion site reactions. However, there is insufficient evidence to
conclude that fosphenytoin is more effective than phenytoin.9 Treatment with
either drug requires cardiac monitoring because of a risk of arrhythmias and QT
prolongation.5 Open-label randomized controlled trials have demonstrated that
IV valproate sodium is as effective or more effective than phenytoin when
used as a second-line antiseizure drug in benzodiazepine-refractory convulsive
status epilepticus.9,13,14 In a meta-analysis of 22 studies, the efficacy of valproate
sodium, phenobarbital, and levetiracetam in benzodiazepine-refractory
convulsive status epilepticus was determined to be 75.7%, 73.6%, and 68.5%,
respectively.15 In contrast, phenytoin was estimated to have an efficacy of 50%,
and the authors concluded that this argues against its common use as the drug of
choice in benzodiazepine-refractory status epilepticus. The ongoing ESETT
(Established Status Epilepticus Treatment Trial) will hopefully provide more
clear guidance in the next few years.16 This prospective randomized double-blind
study will compare the efficacy of phenytoin, valproate sodium, and
levetiracetam in the treatment of benzodiazepine-resistant status epilepticus.
Once again, in comparing the two treatment algorithms in FIGURE 8-2, the
main difference is the recommended time points for second-line therapy. In the
Neurocritical Care Society guidelines, administration of a second-line antiseizure
drug is recommended in 5 to 10 minutes, as opposed to the American Epilepsy
Society, which recommends a second antiseizure drug in 20 to 40 minutes. Both
guidelines state that phenytoin, fosphenytoin, valproate sodium, levetiracetam,
and phenobarbital are reasonable choices at this stage with little evidence to
guide the choice between them. The American Epilepsy Society guidelines
recommend the use of phenobarbital only if the other options are not available
because of adverse effects associated with phenobarbital.9 In keeping with the
more aggressive approach recommended by the Neurocritical Care Society,
458 APRIL 2019

proceeding directly to IV midazolam after the first-line benzodiazepine is also KEY POINTS
presented as an option at this stage in their guidelines.5
● Both guidelines (from the
After administration of a second-line agent, if seizures persist and the patient Neurocritical Care Society
is in refractory status epilepticus, once again, there is very little evidence to guide and American Epilepsy
therapies. In the two presented treatment guidelines, the dosing and medications Society) state that
do not vary greatly, but the timing varies from 20 to 40 minutes to begin this phenytoin, fosphenytoin,
valproate sodium,
third-line therapy. Both guidelines recommend that IV anesthesia be considered
levetiracetam, and
for continued status epilepticus. They also agree that continuous EEG be started phenobarbital are
to titrate the anesthetic agents. The Neurocritical Care Society guidelines reasonable choices for
recommend titrating anesthesia to burst suppression or seizure suppression on second-line antiseizure
drugs with little evidence
EEG (rather than anesthetic drug levels) but acknowledge that there are
to guide the choice
insufficient data to suggest EEG background is predictive of outcome. between them.
Both the 2012 Neurocritical Care Society and 2016 American Epilepsy Society
guidelines offer a reasonable framework for escalation in treatment of continued ● One proposed clinical
convulsive status epilepticus. However, it is important to emphasize that these definition of acute repetitive
seizures is three or more
algorithms/guidelines may not fit all patients and should not replace a clinician’s seizures within 24 hours for
medical judgment. For example, there may be cases in which it is important to patients whose habitual
initiate intubation and anesthetic sedation in patients before the benchmark time seizure frequency is fewer
outlined by these algorithms. For instance, waiting 20 to 40 minutes or for than three seizures per day.
complete infusion of the second antiseizure medications may not be appropriate
for patients who are not supporting their airway or have continued convulsions
despite infusion of the second-line medication. Early treatment is paramount
when dealing with convulsive status epilepticus because the longer the status
epilepticus lasts, the less effective treatments become.10
As previously stated, super-refractory status epilepticus is defined as status
epilepticus that continues or recurs 24 hours or more after the onset of anesthetic
therapy. It also includes cases in which status epilepticus recurs on the reduction
or withdrawal of anesthesia. Although super-refractory status epilepticus is not
common, it has a high mortality and morbidity, with limited proven therapies. A
recent multicenter trial testing brexanolone, an allosteric modulator of
γ-aminobutyric acid A (GABA-A) receptors, for the treatment of super-refractory
status epilepticus was unable to statistically outperform placebo. Although not
officially published, preliminary data suggested that brexanolone helped 43.9%
of patients wean off a medically induced coma without seizures returning for
24 hours, in contrast to 42.4% of patients given placebo and standard-of-care
treatment.17 At this time, therapy for super-refractory status continues to be
based on clinical reports and opinions. TABLE 8-1 presents some potential
treatments for super-refractory status epilepticus.18
ACUTE REPETITIVE SEIZURES

Acute repetitive seizures is a term typically used in an ambulatory patient with
epilepsy or acute symptomatic seizures whose mental status is relatively
preserved between seizures. One proposed clinical definition of acute repetitive
seizures is three or more seizures within 24 hours for patients whose habitual
seizure frequency is fewer than three seizures per day.19 Acute repetitive seizures
may be used to describe convulsive or nonconvulsive seizures and can be
conceived of as seizure clusters. This pattern is often seen in the context of
frontal lobe epilepsy among other conditions (CASE 8-1). In humans, the
consequences and/or degree of potential neuronal injury of intermittent frequent
seizures as compared with continuous seizure activity of status epilepticus is

unclear.20 Using an approach as aggressive as that used in status epilepticus may

be associated with unnecessary morbidity.21 Intubation and anesthetic
management of these patients may obscure the clinical examination that is
important in understanding the patient’s response to treatment. Commonly used
treatments include using IV antiseizure medications and oral benzodiazepines
and/or rectal or parenteral benzodiazepines.19 In practice, seizures may not be
aborted immediately after intervention and often decrease over several hours
to days. Seeing an initial response, such as decreased seizure frequency or
TABLE 8-1 Treatment Intervention Options for Super-refractory Status Epilepticusa
Intervention Studied Doses Adverse Effects Clinical Pearls and Considerations

Ketamine Bolus: 0.5–3 mg/kg Tachycardia; acute elevation in Use only with great caution if the
Infusion: 1–10 mg/kg/h blood pressure; ICP elevation; and patient has an etiology that might
theoretical neurotoxic effects increase ICP (eg, severe brain edema
when used for prolonged periods from anoxic brain injury). Early use of
ketamine may provide better and
faster control of seizures. Consider
in hypotensive patients.
Isoflurane Concentration: 1% to 5% Hypotension requiring IV Likely to stop seizures but not a

vasopressors support; infection; sustained effect. Consider as last-
Infusion: titrate to burst
paralytic ileus; deep vein line therapy.
suppression on EEG
thrombosis; and cognitive
dysfunction with prolonged use
Ketogenic diet 4:1 (the ratio of fat to Hyperlipidemia; weight loss; Compliance is extremely difficult
carbohydrates and contraindicated in pyruvate with long-term use of the diet
protein) carboxylase and beta-oxidation because of social and dietary
deficiency restrictions, cost, and the complexity
involved. Lack of well-designed
trials.
Hypothermia Goal temperature of Coagulation disorders; venous Hypothermia can potentially be used
32°C to 35°C 24 h thrombosis; cardiac arrhythmia; as an alternative to two or more
with rewarming of no electrolyte abnormalities; unsuccessful EEG burst-suppression
more than 0.5°C/h infections; pharmacokinetic and trials. Goal temperature aimed at
pharmacodynamic changes; and appropriate burst-suppression
acute intestinal ischemia/necrosis pattern on EEG.
Electroconvulsive Protocols vary Can induce convulsive and EEG monitoring required. Routine
therapy nonconvulsive status after use not well established. Further
treatment; cognitive impairment; studies are needed.
amnesia; and headache
Transcranial magnetic Can be performed in Rare seizures; headache; Considered a very safe intervention
stimulation the intensive care unit dizziness; and other neurologic and does not require surgery or
setting side effects device implantation. Still
investigational therapy.
Vagal nerve stimulation Surgical implantation Voice hoarseness; infection risk at No strong evidence to support its
the implantation site; and rare use in the acute settings.
bradycardia
EEG = electroencephalogram; ICP = intracranial pressure; IV = intravenous.

a
Modified with permission from Bayrlee A, et al, Curr Neurol Neurosci Rep.18 © 2015 Springer Science+Business Media.
460 APRIL 2019

duration, is important when judging if an intervention is having some effect. KEY POINTS
If seizures were to become prolonged, longer than 5 minutes, or the patient fails
● In a patient who is
to demonstrate significant improvement in mental status between seizures, then critically ill, his or her mental
it is reasonable to revert to the algorithms used for status epilepticus. status may be affected by
Acute repetitive seizures can also occur in a patient who is critically ill. These the underlying cause of
seizures typically are detected on continuous EEG monitoring and may have seizures, so it can be
difficult to determine if the
subtle or no clinical signs. The distinction between repetitive electrographic
patient would return to
seizures, which often occur after a primary neurologic injury (eg, stroke, baseline between seizures.
hemorrhage, tumor, or infection), and prolonged or continuous electrographic
seizure activity of status epilepticus is not well defined. In a patient who is ● In a recent randomized
critically ill, his or her mental status may be affected by the underlying cause controlled trial, lacosamide
was noninferior to
of seizures, so it can be difficult to determine if the patient would return to fosphenytoin in the
baseline between seizures. FIGURE 8-4 places a common consensus between the treatment of nonconvulsive
relationship of urgency of treatment and degree of neuronal injury, with single seizures.
seizures causing little if any neuronal injury and generalized convulsive status
epilepticus clearly resulting in significant neuronal injury, with merging seizures
(ie, seizures that wax and wane in amplitude and frequency characteristics, with
one electrographic seizure merging into another) and nonconvulsive status
epilepticus in the middle.22
There is reasonable consensus and some evidence that acute repetitive
seizures, including repetitive electrographic seizures, can initially be managed
with trials of IV antiseizure drugs as opposed to IV anesthesia.21 However, there
are few data on the efficacy of this approach or on which drugs are most
effective. As in convulsive status epilepticus, phenytoin or fosphenytoin has been
the commonly used first-line treatment. A recent study, TRENdS (Treatment of
Recurrent Electrographic Nonconvulsive Seizures), compared the efficacy of
lacosamide with fosphenytoin in the treatment of nonconvulsive seizures.23 This
is the first prospective randomized study of the treatment of nonconvulsive seizures
and the first interventional study to use continuous EEG as a primary study end
point. In this study, nonconvulsive seizures were defined as electrographic seizures
lasting 10 seconds or more with or without clinical accompaniment. Ictal discharges
lasting 30 minutes or more per hour were not included.
Qualifying patients with nonconvulsive seizures were randomly assigned to
receive an initial load of either fosphenytoin, 20 phenytoin equivalents/kg IV, or
lacosamide 400 mg IV. The patient could receive a re-bolus of the initial drug if
seizures recurred during a 6-hour observation period after treatment with the
initial drug and a 2-hour grace period (ie, between 2 and 8 hours after the initial
treatment). Re-bolus dosing was 5 phenytoin equivalents/kg for fosphenytoin and
200 mg for lacosamide. The primary end point was the absence of electrographic
seizures for 24 hours after the initial treatment with the study drug or after the
re-bolus if needed.
Seventy-four patients were randomly assigned, and 62 had sufficient
continuous EEG for analysis. The mean age of enrolled subjects was 63.6 years.
The majority of patients (62%) had concurrent neurologic injury, including
subdural, parenchymal, or subarachnoid hemorrhage; brain tumor; or
toxic-metabolic encephalopathy. Another 31% of enrolled patients had a history
of epilepsy. Initial treatment with IV lacosamide was associated with a cessation
of seizures without recurrence within 24 hours in 63.3% of patients whereas IV
fosphenytoin terminated seizures in 50% of patients. Although the difference
was not sufficient to show superiority of lacosamide over fosphenytoin, the study

met its primary end point, demonstrating that lacosamide was noninferior to
fosphenytoin in the treatment of nonconvulsive seizures (P=.02). There was no
significant difference in the incidence of treatment-emergent adverse effects or
serious adverse effects between the two groups. Of note, the incidence of
hypotension, arrhythmia, respiratory failure, or multiorgan hypersensitivity was
not different between the two groups (11.4% for fosphenytoin and 13.5% for
lacosamide, P=1.0).
NONCONVULSIVE STATUS EPILEPTICUS

There is no universally accepted definition of nonconvulsive status epilepticus,
and there is no consensus on how to treat it. Not surprisingly, nonconvulsive
status epilepticus is often a challenging diagnosis to make. EEG, although critical
to the evaluation of nonconvulsive status epilepticus, can sometimes be
equivocal. Therefore, in addition to EEG review, an evaluation of the patient’s
clinical state, medical history, and response to antiseizure drugs is often needed
to make the diagnosis. FIGURE 8-5 is an algorithm for the EEG diagnosis of
CASE 8-1 A 20-year-old woman with left frontal lobe epilepsy presented with an
increased frequency of her habitual seizures. Her seizures were well
controlled with levetiracetam until 1 month before when she had a
breakthrough seizure in the setting of sleep deprivation. Two days before
admission, the patient’s mother reported that the patient was sleeping
restlessly, and she started noticing seizures during the day. The patient
would make a facial expression as if she was crying and appear restless
for 20 to 25 seconds. With one of the seizures, she fell to the ground. The
patient was started on clonazepam 0.5 mg 3 times daily, but the
seizures increased in frequency to several times per hour, and the
patient’s mother brought her to the emergency department.
On continuous video-EEG monitoring, the patient was having seizures
every 10 minutes in wakefulness and sleep. Clinically, the seizures were
characterized by an abrupt and involuntary change in the patient’s facial
expression and hypermotor movements. She was aware of the seizures
and was alert throughout. The patient was given lorazepam 2 mg IV and
lacosamide 400 mg IV while she was monitored on cardiac telemetry. She
was continued on lacosamide 200 mg twice daily and levetiracetam
1500 mg twice daily, and clonazepam was increased to 1 mg 3 times
daily. With these interventions, seizures decreased in frequency to once
every 30 minutes and decreased in duration from 15 seconds to 7 to
10 seconds.
Over the next 48 hours, seizures progressively shortened and
decreased in frequency until they resolved. Her seizures were very subtle
on EEG (FIGURE 8-3): They were characterized by rhythmic beta activity
over the left frontocentral region (FIGURE 8-3) and a corresponding
increase in heart rate. An epilepsy-protocol MRI was normal and
unchanged from previous MRIs. The patient was weaned off clonazepam
over the next 2 months. A subsequent inpatient admission confirmed she
was not having clinical or electrographic seizures.
462 APRIL 2019

nonconvulsive status epilepticus and is adapted from previously published
criteria.24–26 The algorithm demonstrates one approach to equivocal EEG
patterns that cannot be definitively classified as ictal or not ictal. In these cases, a
response to treatment (typically IV-administered antiseizure drugs) can help
make the diagnosis. The response must include both a positive response in the
clinical state of the patient and the EEG or complete cessation of electrographic
ictal pattern with return of normal background EEG activity.27
Once the diagnosis of nonconvulsive status epilepticus is made, there is often
disagreement on how aggressively to treat it. Ample animal evidence indicates
that continuous nonconvulsive electrographic seizure activity leads to neuronal
damage and the severity increases with time.28–30 Human data on the degree
of injury associated with nonconvulsive status epilepticus can be difficult to
interpret because the etiology of nonconvulsive status epilepticus often
confounds the outcome data. The serum neuron-specific enolase level has been
noted to be elevated with nonconvulsive status epilepticus without other
neurologic injury, indicating some degree of neuronal damage.31 However,
FIGURE 8-3
EEG of the patient in CASE 8-1 showing very subtle seizures characterized by rhythmic beta
activity over the left frontocentral region (arrows) and a corresponding increase in
heart rate.
This is a case of acute repetitive seizures. Patients with frontal lobe COMMENT
seizures are particularly prone to clusters of repetitive seizures when their
sleep is disturbed. Often, this situation can be managed with standing
benzodiazepines, a new antiseizure medication, and letting the patient
catch up on sleep. Seizures may not completely stop after an intervention,
but a decrease in the frequency and duration of seizures is encouraging,
and seizures may continue to resolve over the next day or two. The
morbidity associated with intubation likely outweighs the benefits in these
cases and limits the ability to follow the patient’s clinical examination,
which was particularly important in this case.

treatment for nonconvulsive

status epilepticus is not benign.21
Initial treatment approaches
range from trials of additional
antiseizure drugs to intubation
and burst suppression with IV
anesthesia, similar to what is
used in convulsive status
epilepticus. No clinical consensus
exists for how aggressively to
treat nonconvulsive status
FIGURE 8-4 epilepticus, but in clinical
Graphic representation of the relationship of the practice, the determination is
urgency of treatment and the morbidity of
seizures, nonconvulsive status epilepticus, and
often based on the patient’s
convulsive status epilepticus. mental status and clinical course
SE = status epilepticus; sz = seizure. (CASES 8-2 and 8-3). In all cases,
Reprinted with permission from Husain AM, Epilepsy
it is important to address the
Behav.22 Published by Elsevier Inc.
underlying etiology.
ETIOLOGY
In adults, status epilepticus occurs in about half of patients as the result of an
acute symptomatic lesion, such as stroke, head trauma, and anoxia. This is then
followed by remote symptomatic lesions and low antiseizure drug levels in
patients with known epilepsy.33–35 In some cases, however, the initial workup
does not reveal a clear etiology. In the management of acute status epilepticus
and recurrent seizures, we often prioritize the control of seizures over the
determination of the cause. However, finding the etiology may have both
therapeutic and prognostic implications. Also, depending on the etiology,
delayed recognition may contribute to a poor outcome in a variety of diseases
that cause status epilepticus.36 Nearly 200 uncommon disorders have been
shown to cause status epilepticus.37
NEW-ONSET REFRACTORY STATUS EPILEPTICUS AND FEBRILE

INFECTION–RELATED EPILEPSY SYNDROME
New-onset refractory status epilepticus (NORSE) is a recently described clinical
presentation that can affect all ages. NORSE occurs in patients without active
epilepsy or other preexisting neurologic disorders and has no clear acute or active
structural, toxic, or metabolic cause.36 Febrile infection–related epilepsy
syndrome (FIRES) is thought to be a subcategory of NORSE that requires a
preceding febrile infection, with fever starting between 2 weeks and 24 hours
before the onset of refractory status epilepticus.36 No known unifying
mechanism exists for cryptogenic NORSE and FIRES, but speculation suggests
that they might be caused by a fulminant inflammatory response in the central
nervous system.38
AUTOIMMUNE ENCEPHALITIS
An increasingly recognized etiology of status epilepticus is autoimmune
encephalitis. In a recent single-center series, among the 570 consecutive patients
with status epilepticus, 2.5% were found to be autoimmune.39 The series found
that, compared with patients with status epilepticus of an infectious cause, the
464 APRIL 2019

KEY POINTS
● No clinical consensus
exists for how aggressively
to treat nonconvulsive
status epilepticus, but in
clinical practice, the
determination is often based
on the patient’s mental
status and clinical course.
● New-onset refractory
status epilepticus (NORSE) is
a recently described clinical
presentation that can affect
all ages and occurs in
patients without active
epilepsy or other preexisting
neurologic disorders and has
no clear acute or active
structural, toxic, or
metabolic cause.
● Febrile infection–related
epilepsy syndrome (FIRES) is
thought to be a subcategory
of new-onset refractory
status epilepticus that
requires a preceding febrile
FIGURE 8-5 infection, with fever starting
Algorithm for the EEG diagnosis of nonconvulsive status epilepticus. Of note, this is a general between 2 weeks and
guideline and does not apply to all scenarios and does not replace clinical judgment. 24 hours before the onset
AED = antiepileptic drug; EEG = electroencephalogram; IV = intravenous; NCSE = nonconvulsive of refractory status
status epilepticus. epilepticus.
Modified with permission from Herman ST, Nonconvulsive Status Elipticus.24 © Spring Publishing
Company, Inc.
patients with autoimmune status epilepticus were younger and had lower
morbidity (return to baseline conditions in 71% versus 32%). However, no
difference in mortality existed between the two groups. Although autoimmune
encephalitis was previously thought of as an extremely rare disease, another
recent population-based comparative study of autoimmune and infectious
encephalitis in Olmsted County, Minnesota, showed that the prevalence and
incidence of autoimmune encephalitis were comparable to infectious
encephalitis, and its detection is increasing over time.40
A small case series of seven patients documented several similarities between
patients with NORSE, including female sex, young age, and a negative
workup.41 In another case series of 11 patients with NORSE, autoantibodies were
identified in seven patients, of which anti–glutamic acid decarboxylase (GAD)
and anti–N-methyl-D-aspartate (NMDA) receptor were most frequent.42 In the
same series of 11 patients, eight were treated with immunotherapy (IV steroids,
immunoglobulins, plasma exchange, or a combination), and four received
chemotherapy. Of the eight patients treated with immunotherapy, six had
favorable outcomes (defined as any outcome other than death, vegetative state,
or inability to take care of oneself ) compared with none of the three patients who
did not receive immunotherapy. Despite the small numbers, this difference in

CASE 8-2 A 66-year-old right-handed man presented after a fall and was found to
have a left hemispheric subdural hematoma involving the tentorium
(FIGURE 8-6). He had been started on phenytoin on admission for seizure
prophylaxis but had not had clinical seizures. Two days after admission,
he developed an acute fluent aphasia. His neuroimaging had not changed.
His speech was nonsensical, and he could not repeat, name, or follow
commands.
Continuous EEG demonstrated frequent left hemispheric
electrographic seizures recurring every 5 to 10 minutes as illustrated by
quantitative EEG (one of the shortest seizures is depicted in the raw EEG
in FIGURE 8-7). Between seizures, EEG demonstrated left hemispheric
lateralized periodic discharges (LPDs). Aphasia was persistent and not
limited to the time when
seizures occurred.
Electrographic seizure
activity initially responded to
lorazepam 2 mg IV, which was
given at 17:00 (5:00 PM) along
with IV levetiracetam 1 g. The
patient’s aphasia, however,
persisted. Seizures recurred
over the next 24 hours but
were shorter in duration and
less frequent. LPDs also
became infrequent. The patient
was given levetiracetam 1 g
3 times over the next 24 hours
and was subsequently continued
on levetiracetam 1500 mg
2 times a day, and phenytoin
was continued. By the following
FIGURE 8-6
day, rare seizures were seen, Axial noncontrast head CT of the patient in CASE 8-2
and the patient had a pure demonstrating left hemispheric subdural hematoma
anomia. By day 3 of continuous involving the tentorium with mass effect.
EEG recording, the aphasia
had resolved as had seizures
and LPDs.
466 APRIL 2019

FIGURE 8-7
Raw EEG of the patient in CASE 8-2 shows one of the patient’s shortest frequent left
hemispheric electrographic seizures.
This case illustrates one form of nonconvulsive status epilepticus. It would COMMENT
be considered nonconvulsive status (as opposed to acute repetitive
seizures) because of the persistent clinical features (ie, the patient did not
return to baseline between seizures). Of course, the distinction between
the two is less clear because it could be argued that the patient’s brain
injury is a factor in his aphasia. The response to treatment, which paralleled
control of seizures, however, suggests that seizures were contributing to
the patient’s mental status. This is a common occurrence in aphasic status;
aphasia often parallels the control of seizures, but the resolution of aphasia
may occur before or after control of electrographic seizures.32 It is also not
uncommon for seizures to wane over 2 to 3 days. In these cases, it is often
not necessary to intubate patients for control of focal status epilepticus.
Demonstrating an initial response to treatment (a temporary cessation of
seizures or a decrease in frequency or duration) offers promise that the
seizures will come under control.
This case contains data from a previously published article.32

CASE 8-3 A 45-year-old woman with no

significant past medical history
presented with hallucinations for
2 weeks and episodes of
behavioral arrest with oral
automatisms. Her initial MRI
demonstrated patchy increased
fluid-attenuated inversion
recovery (FLAIR) signal in the left
more than the right temporal
lobes and left orbitofrontal
cortex (FIGURE 8-8) as well left
parietal and occipital lobes. She
was connected to continuous
video-EEG and found to have
FIGURE 8-8
frequent left temporal seizures, Initial coronal fluid-attenuated inversion recovery
which corresponded to oral (FLAIR) MRI of the patient in CASE 8-3 demonstrates
automatisms (FIGURE 8-9), and patchy increased signal in the left and right
bilateral independent lateralized temporal lobes and left orbitofrontal cortex.
Follow-up imaging after immunotherapy with
periodic discharges (FIGURE 8-10). cyclophosphamide and rituximab demonstrated
In the first 24 hours, she marked improvement (not shown).
was started empirically on
acyclovir and given 4 mg
of lorazepam 2 times and fosphenytoin, levetiracetam, and valproate
sodium IV loads without any clear response to therapy. The patient’s mental
status worsened; she became somnolent and only intermittently followed
commands. The patient was intubated for refractory status epilepticus. She
ultimately required both midazolam and phenobarbital to maintain burst
suppression and control electrographic seizures. CSF analysis was notable
for 2 white blood cells/mm3 (98% lymphocytes) and normal protein.
FIGURE 8-9
Continuous video-EEG of the patient in CASE 8-3 demonstrates left temporal seizures that
corresponded to oral automatisms.
468 APRIL 2019

FIGURE 8-10
Continuous video-EEG of the patient in CASE 8-3 demonstrates that between the seizures the
patient had bilateral independent lateralized periodic discharges.
Infectious studies, including herpes simplex virus, varicella zoster virus, and
cytomegalovirus polymerase chain reaction (PCR), were negative. CT chest,
abdomen, and pelvis and positron emission tomography (PET) scans were
negative for malignancy. The initial PET scan of the body also included the
brain, and sagittal images showed striking hypermetabolism in the limbic
structures (FIGURE 8-11). Two weeks after presentation, the patient’s serum and
CSF returned positive for anti–glutamic
acid decarboxylase (GAD) antibodies.
All other autoantibodies were negative.
In the first week of admission, the
patient was treated with IV steroids and
as well as two courses of plasma
exchange. The patient demonstrated no
improvement with these interventions.
She continued to acquire additional foci
of FLAIR hyperintensity on MRI and new
hypermetabolic lesions on PET. She
remained in burst suppression on EEG,
with seizures returning during weans of IV
pentobarbital. She was ultimately treated
with cyclophosphamide and rituximab.
One week after rituximab treatment, she FIGURE 8-11
was able to be weaned off her IV Sagittal reconstruction of an axial
acquisition brain positron emission
anesthetics without return of seizures. tomography (PET) scan of the patient
She was weaned off most antiepileptic in CASE 8-3 showing striking
drugs over 2 years and has been seizure hypermetabolism in the limbic
free on phenobarbital. structures.
CONTINUED ON
PAGE 470

outcome was significant (P=.026).42 Prospective multicenter studies are

necessary to assess the true efficacy of immunotherapy in patients with NORSE.
Recent predictive models have been created for both neural antibody
positivity and immunotherapy response in epilepsy. These studies were not
specific to status epilepticus or NORSE; they included all new-onset seizures and
status that were cryptogenic. The initial prospective study found 20% of
cryptogenic epilepsy cases to be neural antibody seropositive, indicative of a
possible autoimmune cause. The authors evaluated a scoring system to predict
the detection of these antibodies known as the Antibody Prevalence in Epilepsy
(APE) score (TABLE 8-2).43 An APE score of 4 or greater had a sensitivity of
82.6% and a specificity of 82.0% for detecting a positive antibody. In a follow-up
retrospective study of patients with epilepsy at the Mayo Clinic in whom
autoantibody-testing profiles were requested and who received immunotherapy,
a Response to Immunotherapy in Epilepsy (RITE) score was assigned. A
favorable response to immunotherapy was defined as having a greater than
50% reduction of seizure frequency at the first follow-up (TABLE 8-2). A RITE
score of 7 or greater predicted a favorable seizure outcome and had a sensitivity
of 87.5% and specificity of 83.8%.44 Therefore, the use of APE and RITE scores
can aide diagnosis, treatment, and prognostication of autoimmune etiology in
epilepsy and may potentially be useful in identifying an autoimmune etiology
in NORSE.
If there is a concern for a potential autoimmune etiology for seizures or
NORSE, a standard diagnostic algorithm, shown in FIGURE 8-12, can be followed.
The evaluation should be individualized for each patient, and some patients may
not require all the studies shown. However, it is important to emphasize the
possibility that there could be a coexisting malignancy, and if found, it is
important to treat both the tumor and the paraneoplastic disorder simultaneously.
IMMUNE TREATMENT FOR AUTOIMMUNE ENCEPHALITIS

When choosing a treatment strategy for any antibody-associated central nervous
system disorder, it is key to choose objective markers to monitor treatment
response. For patients with NORSE, we would recommend monitoring seizure
frequency (location and duration of the seizures), imaging (brain MRI and/or
CONTINUED FROM
PAGE 469
COMMENT In contrast to CASE 8-2, this case of nonconvulsive status epilepticus
represents one in which intubation and aggressive management with IV
anesthesia are usually indicated. The patient presented at the beginning
of a likely progressive condition (as indicated by inflammatory MRI changes
and clinical deterioration) and was refractory to medications. This is a
case of new-onset refractory status epilepticus (NORSE), ultimately proven
to be of an autoimmune etiology. The patient would have had an Antibody
Prevalence in Epilepsy (APE) score of 6 and a Response to Immunotherapy
in Epilepsy (RITE) score of 8. This case also illustrates the importance of
patience and persistence in cases of autoimmune refractory status
epilepticus because patients can ultimately do very well despite a
prolonged and arduous course.
470 APRIL 2019

Components of the Antibody Prevalence in Epilepsy and Response to TABLE 8-2
Immunotherapy in Epilepsy Scoresa,b
Antibody Prevalence in Epilepsy of an Unknown Response to Immunotherapy in Epilepsy

Etiology Score Value Score Value
New-onset, rapidly progressive mental status (+1) New-onset, rapidly progressive mental status (+1)
changes that developed over 1–6 weeks or new- changes that developed over 1–6 weeks or new-
onset seizure activity (within 1 year of evaluation) onset seizure activity (within 1 year of evaluation)
Neuropsychiatric changes; agitation, (+1) Neuropsychiatric changes; agitation, (+1)

aggressiveness, emotional liability aggressiveness, emotional liability
Autonomic dysfunction (sustained atrial (+1) Autonomic dysfunction (sustained atrial (+1)
tachycardia or bradycardia, orthostatic tachycardia or bradycardia, orthostatic
hypotension [≥20 mm Hg decrease in systolic hypotension [≥20 mm Hg decrease in systolic
pressure or ≥10 mm Hg decrease in diastolic pressure or ≥10 mm Hg decrease in diastolic
pressure within 3 minutes of standing], pressure within 3 minutes of quiet standing],
hyperhidrosis, persistently labile blood pressure, hyperhidrosis, persistently labile blood pressure,
ventricular tachycardia, or cardiac asystole) ventricular tachycardia, or cardiac asystole)
Viral prodrome (rhinorrhea, sore throat, low- (+2) Viral prodrome (rhinorrhea, sore throat, low- (+2)
grade fever) to be scored in the absence of grade fever) only to be scored in the absence
underlying systemic malignancy of underlying malignancy
Facial dyskinesias or faciobrachial dystonic movements (+2) Facial dyskinesias or faciobrachial dystonic movements (+2)
Seizure refractory to at least to two antiseizure (+2) Seizure refractory to at least to two antiseizure (+2)
medications medications
CSF findings consistent with inflammation (+2) CSF findings consistent with inflammation (+2)
(elevated CSF protein >50 mg/dL and/or (elevated CSF protein >50 mg/dL and/or
lymphocytic pleocytosis >5 cells/mm3, if the total lymphocytic pleocytosis >5 cells/mm3, if the total
CSF red blood cell count is <1000 cells/mm3) CSF red blood cell count is <1000 cells/mm3)
Brain MRI showing signal changes consistent (+2) Brain MRI showing signal changes consistent (+2)
with limbic encephalitis (medial temporal with limbic encephalitis (medial temporal
T2-weighted/FLAIR signal changes) T2-weighted/FLAIR signal changes)
Presence of underlying malignancy (excluding (+2) Presence of underlying malignancy (excluding (+2)
cutaneous squamous cell carcinoma, basal cell cutaneous squamous cell carcinoma, basal
carcinoma) cell carcinoma)
Total Initiation of immunotherapy within 6 months (+2)

(maximum: 15) of symptom onset
Detected neural plasma membrane autoantibody (+2)

(NMDA receptor antibody, GABA-A receptor
antibody, GABA-B receptor antibody, AMPA
receptor antibody, DPPX antibody, mGluR1
antibody, mGluR2 antibody, mGluR5 antibody,
LGI1 antibody, or Caspr2 antibody)
Total
(maximum: 19)
a
Reprinted with permission from Dubey D, et al, Epilepsia.44 © 2017 John Wiley and Sons.
b
A Response to Immunotherapy in Epilepsy (RITE) score includes all the components of the Antibody Prevalence in Epilepsy (APE) score and two additional
variables: initiation of immunotherapy within 6 months of symptom onset and presence of plasma membrane autoantibody. An APE score of 4 or greater was
highly suggestive of the presence of an autoantibody. A RITE score of 7 or greater predicted a favorable seizure outcome in response to immunotherapy in a
patient with known or suspected autoimmune epilepsy.
AMPA = amino-3-hydroxy-5-methyl-4-isoxazolepropionic; Caspr2 = contactin-associated protein 2; CSF = cerebrospinal fluid; DPPX = dipeptidyl-
peptidase–like protein-6; FLAIR = fluid-attenuated inversion recovery; GABA = γ-aminobutyric acid; LGI1 = leucine-rich, glioma-inactivated protein 1;
mGluR = metabotropic glutamate receptor; MRI = magnetic resonance imaging; NMDA = N-methyl-D-aspartate.

KEY POINTS
● In one study, an Antibody

Prevalence in Epilepsy score
of 4 or greater had a
sensitivity of 82.6% and a
specificity of 82.0% for
detecting a positive
antibody.
● In one study, a Response

to Immunotherapy in
Epilepsy score of 7 or
greater predicted favorable
seizure outcome and had a
sensitivity of 87.5% and
specificity of 83.8%.
● When choosing a
treatment strategy for any
FIGURE 8-12
antibody-associated central
Diagnostic algorithm for autoimmune encephalitis.
nervous system disorder, it
ANA = antinuclear antibody; CSF = cerebrospinal fluid; CT = computed tomography; EEG =
is key to choose objective electroencephalography; EGD = esophagogastroduodenoscopy; FLAIR = fluid-attenuated inversion
markers to monitor recovery; GI = gastrointestinal; IgG = immunoglobulin G; MRI = magnetic resonance imaging; PET = positron
treatment response. emission tomography; TPO = thyroid peroxidase antibody.
Modified with permission from Linnoila J, Pittock SJ, Semin Neurol.45 © 2016 Rights managed by Georg
Thieme Verlag KG.
positron emission tomography [PET]), cognitive testing, and, if present, any

movement disorders or autonomic dysfunction. Additionally, the presence of a
neural autoantibody does not guarantee a good response to treatment. This is
especially true when there is a paraneoplastic antibody and/or intracellular
antigen.45 Cell surface autoantibodies and intracellularly targeted antibodies are
highlighted in TABLE 8-3. Therefore, a successful immune treatment may be
marked by stopping of clinical progression and not necessarily a reversal of
clinical symptoms or signs. The initial goal in treatment is to determine whether
there is a response to immunotherapy. If a positive response is found, then the
authors of this article recommend shifting focus to find the minimal/safest dose
of immunotherapy to maintain the response.45 An initial treatment algorithm is
shown in FIGURE 8-13. Typically, patients are treated with first-line therapies: IV
steroids (five daily doses), IV immunoglobulin (five daily doses), and/or plasma
exchange (five sessions total, every other day). These may be potentially
followed by second-line therapies: rituximab and cyclophosphamide. This is a
reasonable approach in hospitalized patients for the majority of cell surface–
targeted neural autoantibodies. It is important to recognize that full recovery
may take months. Additionally, if the initial response to first-line therapies is
small, particularly in a patient whose symptoms are severe, it is reasonable to
start second-line therapies earlier. When treating a patient with an intracellularly
targeted antibody, cyclophosphamide is generally preferred.45
CONCLUSION
Status epilepticus, acute repetitive seizures, and autoimmune encephalitis are
neurologic emergencies. Rapid identification and treatment of these conditions
472 APRIL 2019

Cell Surface and Intracellular Targeted Antibodiesa TABLE 8-3
Intracellularly Targeted Antibodies

◆ Antiglial nuclear antibody (sex determining region Y box 1 transcription factor)
◆ Amphiphysin
◆ Antineuronal nuclear antibody 1 (Anti-Hu)
◆ Antineuronal nuclear antibody 2 (Anti-Ri)
◆ Antineuronal nuclear antibody 3
◆ Collapsin response mediator protein 5 (CRMP-5) (anti-CV2)
◆ Glutamic acid decarboxylase (GAD65)
◆ Ma1
◆ Ma2 (Ta)
◆ Purkinje cell cytoplasmic antibody 1 (Anti-Yo)
◆ Purkinje cell cytoplasmic antibody 2
◆ Retinal (cancer-associated retinopathy, melanoma-associated retinopathy)
◆ Striational
◆ Zinc finger protein
Cell Surface Autoantibodies
◆ Acetylcholine receptor (muscle and ganglionic)
◆ α-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor
◆ Aquaporin-4
◆ Contactin-associated protein-like 2
◆ Dipeptidyl-peptidase–like protein 6
◆ γ-Aminobutyric acid A receptor
◆ γ-Aminobutyric acid B receptor
◆ Glycine α receptor
◆ Immunoglobulinlike family member 5
◆ Leucine-rich, glioma inactivated 1 (LGI1)
◆ Metabotropic glutamate receptor 1
◆ Metabotropic glutamate receptor 5
◆ Myelin oligodendrocyte glycoprotein
◆ N-methyl-D-aspartic acid receptor
◆ Purkinje cell cytoplasmic antibody Tr (anti–delta/notchlike epidermal growth factor–related)
◆ Voltage-gated calcium channel (P/Q, N-type)
◆ Voltage-gated potassium channel complex
a
Reprinted with permission from Linnoila J, Pittock SJ, Semin Neurol.45 © 2016 Rights managed by Georg
Thieme Verlag KG.

FIGURE 8-13
Inpatient treatment workflow for autoimmune central nervous system neurologic disorders.
IVIg = intravenous immunoglobulin.
Adapted with permission from Linnoila J, Pittock SJ, Semin Neurol.45
© 2016 Rights managed by Georg Thieme Verlag KG.
may influence patient prognosis. For convulsive status epilepticus, there is

reasonable consensus on the initial steps that need to be taken, but scarce data
exist to direct treatment after benzodiazepine therapy. Even fewer data exist to
inform management of acute repetitive seizures and nonconvulsive status
epilepticus. In these situations, it is important to use clinical judgment to balance
the risk of ongoing seizure activity with the risks related to potential treatments.
In all seizure emergencies, after initial stabilization of the patient, it is important
to recognize and treat the underlying etiology. Autoimmune encephalitis is an
increasingly recognized and important cause of seizures and status epilepticus,
particularly new-onset seizures and status. When confronted with NORSE,
clinicians should maintain a high-level of suspicion for autoimmune encephalitis
and become familiar with its treatments.
REFERENCES
1 Treatment of convulsive status epilepticus. 6 Bleck TP. Refractory status epilepticus.

Recommendations of the Epilepsy Foundation of Curr Opin Crit Care 2005;11(2):117–120.
America’s Working Group on Status Epilepticus. doi:10.1097/01.ccx.0000157079.72999.87.
JAMA 1993;270(7):854–859.
7 Shorvon S, Ferlisi M. The treatment of super-
2 Lowenstein DH, Bleck T, Macdonald RL. It’s time refractory status epilepticus: a critical review of
to revise the definition of status epilepticus. available therapies and a clinical treatment
Epilepsia 1999;40(1):120–122. doi:10.1111/j. protocol. Brain 2011;134(pt 10):2802–2818.
1528-1157.1999.tb02000.x. doi:10.1093/brain/awr215.
3 Walker M, Cross H, Smith S, et al. Nonconvulsive 8 Treiman DM, Walton NY, Kendrick C. A progressive
status epilepticus: Epilepsy Research Foundation sequence of electroencephalographic changes
workshop reports. Epileptic Disord 2005;7(3): during generalized convulsive status epilepticus.
253–296. Epilepsy Res 2018;5(1):49–60. doi:10.1016/0920-
1211(90)90065-4.
4 Husain AM. Treatment of Recurrent
Electrographic Nonconvulsive Seizures (TRENdS) 9 Glauser T, Shinnar S, Gloss D, et al. Evidence-
Study. Epilepsia 2013;54(suppl 6):84–88. based guideline: treatment of convulsive status
doi:10.1111/epi.12287. epilepticus in children and adults: report of the
guideline committee of the American Epilepsy
5 Brophy GM, Bell R, Claassen J, et al. Guidelines
Society. Epilepsy Curr 2016;16(1):48–61.
for the evaluation and management of status
doi:10.5698/1535-7597-16.1.48.
epilepticus. Neurocrit Care 2012;17(1):3–23.
doi:10.1007/s12028-012-9695-z.
474 APRIL 2019

10 Treiman DM, Meyers PD, Walton NY, et al. A 24 Herman ST. The electroencephalogram of
comparison of four treatments for generalized nonconvulsive status epilepticus. In: Kaplan PW,
convulsive status epilepticus. Veterans Affairs Drislane FW, eds. Nonconvulsive status
Status Epilepticus Cooperative Study Group. epilepticus. New York, NY: Demos Medical
N Engl J Med 1998;339(12):792–798. doi:10.1056/ Publishing, 2009:41–62.
NEJM199809173391202.
25 Young GB, Jordan KG, Doig GS. An assessment
11 Leppik IE, Derivan AT, Homan RW, et al. Double- of nonconvulsive seizures in the intensive
blind study of lorazepam and diazepam in status care unit using continuous EEG monitoring: an
epilepticus. JAMA 1983;249(11):1452–1454. investigation of variables associated with
doi:10.1001/jama.1983.03330350028021. mortality. Neurology 1996;47(1):83–89. doi:10.1212/
WNL.47.1.83.
12 Silbergleit R, Durkalski V, Lowenstein D, et al.
Intramuscular versus intravenous therapy for 26 Chong DJ, Hirsch LJ. Which EEG patterns warrant
prehospital status epilepticus. N Engl J Med 2012; treatment in the critically ill? Reviewing the
366(7):591–600. doi:10.1056/NEJMoa1107494. evidence for treatment of periodic epileptiform
discharges and related patterns. J Clin Neurophysiol
13 Misra UK, Kalita J, Patel R. Sodium valproate vs
2005;22(2):79–91. doi:10.1097/01.WNP.0000158699.
phenytoin in status epilepticus: a pilot study.
78529.AF.
Neurology 2006;67(2):340–342. doi:10.1212/01.
wnl.0000224880.35053.26. 27 Kaplan PW. The EEG of status epilepticus. J Clin
Neurophysiol 2006;23(3):221–229. 10.1097/01.wnp.
14 Agarwal P, Kumar N, Chandra R, et al. Randomized
0000220837.99490.66.
study of intravenous valproate and phenytoin in
status epilepticus. Seizure 2007;16(6):527–532. 28 Olney JW, Collins RC, Sloviter RS. Excitotoxic
doi:10.1016/j.seizure.2007.04.012. mechanisms of epileptic brain damage.
Adv Neurol 1986;44:857–877.
15 Yasiry Z, Shorvon SD. The relative effectiveness
of five antiepileptic drugs in treatment of 29 Meldrum BS. Excitotoxicity and selective
benzodiazepine-resistant convulsive status neuronal loss in epilepsy. Brain Pathol 1993;3(4):
epilepticus: a meta-analysis of published 405–412.
studies. Seizure 2014;23(3):167–174. doi:10.1016/j.
30 Fujikawa DG. The temporal evolution of neuronal
seizure.2013.12.007.
damage from pilocarpine-induced status
16 Cock HR. Established status epilepticus treatment epilepticus. Brain Res 1996;725(1):11–22.
Ttrial (ESETT). Epilepsia 2011;52(suppl 8):50–52. doi:10.1016/0006-8993(96)00203-X.
doi:10.1111/j.1528-1167.2011.03237.x.
31 Rabinowicz AL, Correale JD, Bracht KA, et al.
17 Sage Therapeutics. Sage Therapeutics Neuron-specific enolase is increased after
reports top-line results from phase 3 STATUS nonconvulsive status epilepticus. Epilepsia 1995;
trial of brexanolone in super-refractory 36(5):475–479. doi:10.1111/j.1528-1157.1995.
status epilepticus. investor.sagerx.com/ tb00489.x.
news-releases/news-release-details/sage-
32 Ericson EJ, Gerard EE, MacKen MP, Schuele SU.
therapeutics-reports-top-line-results-phase-
Aphasic status epilepticus: electroclinical correlation.
3-status-trial. Updated September 12, 2017.
Epilepsia 2011;52(8):1452–1458. doi:10.1111/j.1528-1167.
Accessed February 8, 2019.
2011.03084.x.
18 Bayrlee A, Ganeshalingam N, Kurczewski L,
33 Hesdorffer DC, Logroscino G, Cascino G, et al.
Brophy GM. Treatment of super-refractory
Incidence of status epilepticus in Rochester,
status epilepticus. Curr Neurol Neurosci Rep
Minnesota, 1965–1984. Neurology 1998;50(3):
2015;15(10):66. doi:10.1007/s11910-015-0589-2.
735–741. doi:10.1212/WNL.50.3.735.
19 McKee HR, Abou-Khalil B. Outpatient
34 DeLorenzo RJ, Hauser WA, Towne AR, et al. A
pharmacotherapy and modes of administration
prospective, population-based epidemiologic
for acute repetitive and prolonged seizures.
study of status epilepticus in Richmond, Virginia.
CNS Drugs 2014;29(1):55–70. doi:10.1007/
Neurology 1996;46(4):1029–1035. doi:10.1212/
s40263-014-0219-6.
WNL.46.4.1029.
20 Haut SR, Velísková J, Moshé SL. Susceptibility of
35 Coeytaux A, Jallon P, Galobardes B, Morabia A.
immature and adult brains to seizure effects.
Incidence of status epilepticus in French-
Lancet Neurol 2004;3(10):608–617. doi:10.1016/
speaking Switzerland: (EPISTAR). Neurology
S1474-4422(04)00881-6.
2000;55(5):693–697. doi:10.1212/WNL.55.5.693.
21 Hirsch LJ. Finding the lesser of two evils: treating
36 Gaspard N, Hirsch LJ, Sculier C, et al. New-onset
refractory status epilepticus. Epilepsy Curr 2015;
refractory status epilepticus (NORSE) and febrile
15(6):313–316. doi:10.5698/1535-7511-15.6.313.
infection-related epilepsy syndrome (FIRES):
22 Husain AM. Lacosamide in status epilepticus: State of the art and perspectives. Epilepsia 2018;
update on the TRENdS study. Epilepsy Behav 59(4):745–752. doi:10.1111/epi.14022.
2015;49:337–339. doi:10.1016/j.yebeh.2015.06.018.
37 Tan RY, Neligan A, Shorvon SD. The uncommon
23 Husain AM, Lee JW, Kolls BJ, et al. Randomized causes of status epilepticus: a systematic
trial of lacosamide vs fosphenytoin for review. Epilepsy Res 2010;91(2–3):111–122.
nonconvulsive seizures. Ann Neurol 2018. doi:10.1016/j.eplepsyres.2010.07.015.

38 Sakuma H, Tanuma N, Kuki I, et al. Intrathecal 42 Khawaja AM, DeWolfe JL, Miller DW, Szaflarski JP.
overproduction of proinflammatory cytokines New-onset refractory status epilepticus
and chemokines in febrile infection-related (NORSE)—the potential role for immunotherapy.
refractory status epilepticus. J Neurol Neurosurg Epilepsy Behav 2015;47:17–23. doi:10.1016/j.
Psychiatry 2015;86(7):820–822. doi:10.1136/jnnp- yebeh.2015.04.054.
2014-309388.
43 Dubey D, Alqallaf A, Hays R, et al. Neurological
39 Ferlazzo E, Gasparini S, Sueri C, Aguglia U. Status autoantibody prevalence in epilepsy of unknown
epilepticus of inflammatory etiology: a cohort etiology. JAMA Neurol 2017;74(4):397–402.
study. Neurology 2016;86(11):1076. doi:10.1212/ doi:10.1001/jamaneurol.2016.5429.
WNL.0000000000002508.
44 Dubey D, Singh J, Britton JW, et al. Predictive
40 Dubey D, Pittock SJ, Kelly CR, et al. Autoimmune models in the diagnosis and treatment of
encephalitis epidemiology and a comparison to autoimmune epilepsy. Epilepsia 2017;58(7):
infectious encephalitis. Ann Neurol 2018;83(1): 1181–1189. doi:10.1111/epi.13797.
166–177. doi:10.1002/ana.25131.
45 Linnoila J, Pittock SJ. Autoantibody-associated
41 Wilder-Smith EP, Lim EC, Teoh HL, et al. The central nervous system neurologic disorders.
NORSE (new-onset refractory status epilepticus) Semin Neurol 2016;36(4):382–396. doi:10.1055/
syndrome: defining a disease entity. Ann Acad s-0036-1585453.
Med Singapore 2005;34(7):417–420.
DISCLOSURE
Continued from page 454
Mental Health, National Institute on Aging, National received research/grant support from the
Institutes of Health, and SAGE Therapeutics. Eleanor Wood-Prince Grant from the Woman’s
Dr VanHaerents has received travel honoraria from Board of Northwestern Memorial Hospital, the
NeuroPace and SAGE Therapeutics. Dr Gerard has National Institute of Neurological Disorders and
received personal compensation as a lecturer for Stroke/National Institutes of Health, SAGE
the Society for Maternal-Fetal Medicine, Society of Therapeutics, and Sunovion Pharmaceutical Inc.
OB/GYN Hospitalists, and UCB China and has
476 APRIL 2019

Counseling and REVIEW ARTICLE

Management of the Risks C O N T I N U UM A U D I O
ONLINE
of Living With Epilepsy

By Katherine Noe, MD, PhD, FAAN
ABSTRACT
PURPOSE OF REVIEW: For patients living with epilepsy, quality of life is
determined not only by seizure control but by mood, antiepileptic drug
adverse effects, relationships, and access to education, employment, and
transportation. This article reviews some of the most commonly
encountered concerns associated with epilepsy, including mood
disorders, driving, injuries, mortality, bone health, genetic burden, and
impact on relationships.
RECENT FINDINGS:People with epilepsy are at increased risk for anxiety,

depression, and suicide. Depression is underrecognized in patients with
epilepsy, but effective validated screening tools are available for use.
Mortality rates for people with epilepsy are 2 times higher than those of
the general population, but much of this is attributable to underlying
conditions rather than seizures. Sudden unexpected death in epilepsy
(SUDEP) occurs in an estimated 1:1000 adults with epilepsy per year, and
the risk can be reduced by improved observation and seizure control. An
increased risk of injury, including fractures, is also present in patients with
epilepsy. Reduced bone health leading to increased fracture risk is an
important negative consequence of long-term use of antiepileptic
medication. Seizures while driving can also cause accidents and injury. CITE AS:
Despite the importance of driving for people with epilepsy, physicians are CONTINUUM (MINNEAP MINN)
2019;25(2, EPILEPSY):477–491.
underperforming in providing counsel about driving.
SUMMARY: Optimal care of the patient with epilepsy includes addressing Dr Katherine Noe, Department
risks to emotional health, physical health including fractures and SUDEP, of Neurology, Mayo Clinic
Arizona, 5777 E Mayo Blvd,
social health, and an independent lifestyle. Identification of and Phoenix, AZ 85054,
treatments to reduce these risks can do more to improve quality of life noe.katherine@mayo.edu.
than a narrow clinical focus on seizure control alone.
Dr Noe has received
research/grant support from
NeuroPace, Inc.
INTRODUCTION
M
edical care of epilepsy is often focused on reducing the UNLABELED USE OF
frequency and severity of seizures and managing antiepileptic
USE DISCLOSURE:
drugs (AEDs). However, quality of life for people living with Dr Noe reports no disclosure.
epilepsy is determined by many factors beyond seizure
control, as stressed in a 2012 Institute of Medicine report on © 2019 American Academy
epilepsy and public health.1 Comprehensive care of the person with epilepsy of Neurology.

COUNSELING AND MANAGEMENT OF EPILEPSY RISKS
CASE 9-1 A 32-year-old woman with a history of focal epilepsy presented for a
follow-up visit with concerns about her antiepileptic medication. She had
been diagnosed with temporal lobe epilepsy 6 months prior after having
several episodes of transient expressive aphasia, one of which
progressed to loss of awareness with unresponsiveness and oral
automatisms. At that time, she was started on lamotrigine titrated to
100 mg 2 times a day and had subsequently been seizure free.
At her follow-up appointment, she reported significant daytime
fatigue and lack of energy despite routinely getting 10 hours of sleep at
night. At work she struggled to maintain her mental focus and complete
her assignments on time. She admitted to ongoing worries about having
another seizure in public and had cut back her social activities and
work-related travel significantly. She was previously training for a half
marathon with her husband, but she no longer exercised because of loss
of interest and low energy. She reported that these issues were
negatively impacting her relationship with her husband. She suspected
the fatigue and cognitive issues were caused by the lamotrigine and
wondered if she should try a different medication.
Her serum lamotrigine level was in the low therapeutic range. She
scored 15 on the Neurological Disorders Depression Inventory for
Epilepsy, suggesting the possibility of major depression, although her
score on the Adverse Events Profile was only 22, driven mostly by higher
scores for the items in the Mood/Emotion domain of the Adverse Events
Profile. After reviewing her situation, her neurologist advised that these
symptoms were more likely to reflect depression rather than an
antiepileptic drug side effect. She agreed to a trial of citalopram, a
selective serotonin reuptake inhibitor (SSRI), at 20 mg/d and continuation
of her dose of lamotrigine. After 3 months of treatment, her symptoms
significantly improved.
COMMENT This case illustrates the challenges in recognizing depression and anxiety in
patients with epilepsy. Patients with epilepsy and their medical providers
may struggle to distinguish the effects of a mood disorder from medication
side effects. Use of a depression screening tool validated for the epilepsy
population, together with systematic screening for adverse events with the
Adverse Events Profile, can be an effective means of identifying mood
disorders in the outpatient clinic. A study involving 20 epileptologists who
utilized the Neurological Disorders Depression Inventory for Epilepsy and
the Adverse Events Profile systematically in their clinics before patient
visits demonstrated that use of these tools improved both the number of
discussions about these important aspects of daily living with epilepsy, as
well as increasing patient–physician agreement about these issues after
clinic visits.13
478 APRIL 2019

therefore optimally addresses the impact of the diagnosis on other areas of daily KEY POINT
living including mood, AED adverse effects, safety, driving, education,
● Depression risk is
employment, and independence. In current vernacular, this may be described as increased 2 to 3 times in
a focus on living well or wellness rather than a focus on the disease. Wellness people with epilepsy.
incorporates not only seizure management but also overall physical, spiritual, People with epilepsy are
mental, and social health as well as lifestyle factors such as occupation, finances, twice as likely to report
suicidal thoughts and up to 3
and housing. The adverse impact of epilepsy beyond seizures can be measured as
times more likely to die of
a financial cost to society. In the United States alone, the total cost related to suicide compared to the
epilepsy in 1995 was $12.5 billion, 85% of which was the indirect cost.2 For general population.
individuals with epilepsy, these costs are real and measurable: missed days of
work and school, underemployment, unemployment, social isolation, strains on
relationships, and the emotional and financial burden of care placed on family.
Costs, direct and indirect, are higher in people with refractory epilepsy.2
Addressing wellness and the risks associated with epilepsy is not a “one size
fits all” endeavor. People with epilepsy experience widely varying types, timing,
frequency, and severity of their seizures. Further, epileptic seizures by their
nature are often unpredictable and sporadic. The challenge is how to best balance
restrictions on activity designed to maintain health and safety during seizures
with the sometimes-competing goal of maintaining a full and active lifestyle
during periods of seizure freedom. This review aims to help medical care
providers impart accurate information about the risks associated with seizures
while still empowering people with epilepsy to achieve optimal quality of life.
EMOTIONAL HEALTH: MOOD DISORDERS, DEPRESSION, AND ANXIETY

People with epilepsy are at increased risk for psychiatric disorders including
depression, anxiety, and attention deficit disorder.3–5 Meta-analyses of
population-based studies indicates a pooled lifetime prevalence of 23% for
depression and 20% for anxiety in people with epilepsy, representing a twofold
to threefold increased lifetime risk.4,5 Rates are even higher in people with
intractable epilepsy.6 Furthermore, people with epilepsy are twice as likely to
report suicidal thoughts and up to 3 times more likely to die of suicide compared
to the general population.7–9
Mood disorders and anxiety are significant determinates of quality of life for
children and adults with epilepsy.10 Depression can occur at any time or any age
in people with epilepsy and may precede the onset of seizures.11 Furthermore,
increased depression severity is associated with an increased likelihood of
uncontrolled seizures and has been correlated with lower adherence to AED
therapy.11,12 The relationship between epilepsy and mood is complex and likely
reflects shared biological and neurochemical pathways related to etiology,
seizures, and medication as well as psychosocial factors.10
Despite the potentially serious implications for quality of life, depression in
epilepsy is underrecognized.10 One reason for this may be the challenge of
distinguishing the physical symptoms of depression such as fatigue, altered sleep,
and cognitive slowing from the adverse side effects of AEDs, as illustrated in
CASE 9-1. Patients and family members may also attribute depression to a
“normal” response to the limitations from seizures and thus not seek care.
Furthermore, depression in people with epilepsy may present in atypical ways or
may fluctuate in temporal relation to seizure activity.10 Recognition can be
improved through use of brief screening tools for depression that have been
validated for people with epilepsy and are available for free, public use: the

Neurological Disorders Depression Inventory for Epilepsy, the Patient Health

Questionnaire-9, and the Patient Health Questionnaire-2.14 These tools
intentionally focus less on physical symptoms of depression that could overlap
with adverse AED side effects.14 Of course, screening tools are not a replacement
for a more detailed evaluation by a physician but can help identify depression as
a concern requiring further attention.
Distinguishing the symptoms of a mood disorder from the common problem
of AED adverse effects in a person with epilepsy is a related challenge for the
epilepsy clinician. Fortunately, a readily available tool, the Liverpool Adverse
Events Profile, can assist with discriminating adverse effects from AED
treatment, and use of this tool has been shown to improve clinicians’ ability to
effectively screen for side effects of treatment to make therapeutic changes in
AEDs that may benefit patients’ quality of life.15
Mood disorders in people with epilepsy are also undertreated.10 As many as
70% of people with depression seen in epilepsy clinics are not receiving
TABLE 9-1 Studies of Treatment of Depression in Patients With Epilepsy
Intervention Study Sample Outcome
Amitriptyline versus nomifensine n = 42, ages 18–60 Equivalent improvement in Hamilton Depression Rating
versus placebo19 Scale score for both active treatments and placebo at
6 weeks, nomifensine more effective than amitriptyline
at 12 weeks
Venlafaxine versus no treatment20 n = 64, ages 7–60 Improved Hamilton Depression Rating Scale score at
8 weeks with active treatment
Paroxetine versus doxepin21 n = 67, ages 14–62 Improved Hamilton Depression Rating Scale score in
both groups at 8 weeks; lower side effects with
paroxetine
Cognitive-behavioral therapy versus n = 15, adults with Improved Beck Depression Inventory score in both
selective serotonin reuptake temporal lobe epilepsy groups at 12 weeks
inhibitor (SSRI)22
Cognitive-behavioral therapy versus n = 37, age 60 and older, Geriatric Depression Scale score improved in both
control23 community dwelling groups, no difference in treatment versus control after
6-week intervention
PEARLS randomized trial: Home-based n = 80, age 18 and older Improved Hopkins Symptom Checklist-20 score at 6, 12,
collaborative care intervention and 18 months versus control
including psychiatric consultation,
versus treatment as usual24
Cognitive-behavioral therapy versus n = 45, ages 18–65 Improved Neurological Disorders Depression Inventory
control25 for Epilepsy score after 9 weeks active treatment; no
difference seen at 3-months posttreatment
Project UPLIFT: Web/telephone- n = 128, adults Decreased incidence of new or relapsed major
delivered cognitive-behavioral depressive episodes and Beck Depression Inventory
therapy–based mindfulness program score at 10 weeks in treatment group
versus treatment as usual26
PEARLS = Program to Encourage Active, Rewarding Lives; UPLIFT = Using Practice and Learning to Increase Favorable Thoughts.
480 APRIL 2019

treatment for the mood disorder.16 One often cited barrier to pharmacologic KEY POINTS
treatment is fear that antidepressant medications exacerbate seizures.10 In fact,
● Depression in patients
this fear appears largely unfounded and is not supported by meta-analysis of with epilepsy is
trials submitted to the US Food and Drug Administration (FDA).16 One underrecognized and
exception is bupropion, which has been shown to increase the risk of seizures and undertreated.
is generally contraindicated in patients with epilepsy.17,18
● The often-cited fear that
Another barrier may be the lack of high-quality evidence to guide treatment
a selective serotonin
decisions specific to depression in patients with epilepsy. Few randomized reuptake inhibitor will
controlled trials have been conducted on the treatment of depression in epilepsy worsen seizure control is not
(TABLE 9-1).19–26 Class III and IV evidence supports the effectiveness of supported by evidence and
citalopram, mirtazapine, reboxetine (not available in the United States), and should not be a barrier to
treatment of depression in
sertraline.27–30 Consensus expert recommendation is to use a selective serotonin the patient with epilepsy.
reuptake inhibitor (SSRI) as first-line therapy.10,31 Psychotherapy and counseling
are also appropriate interventions, either alone or in combination with
pharmacologic treatment.31 Cognitive-behavioral therapy–based treatments, in
particular, administered directly or via telecommunication programs, have
evidence as effective interventions for depression in epilepsy (TABLE 9-1).22–26
Referral to a psychiatrist is often appropriate, particularly for depression that is
moderate to severe or that does not respond to first-line interventions.
Choice of AED therapy can also impact mood for better or worse. Expert
consensus recommendations for treatment of mild depression in people with
epilepsy include use of an AED with mood-stabilizing properties such as
lamotrigine or valproate, as well as avoidance of medications with an increased
likelihood for adverse emotional side effects.10 Lamotrigine was shown in two
randomized, double-blind trials to improve mood in participants with epilepsy
after 7 to 8 weeks of treatment.32
In 2008 the FDA issued a safety alert for increased risk of suicidal behavior
and suicidal ideation with AEDs as a class, leading to a change in drug labeling.33
This was based on a meta-analysis of placebo-controlled AED trials for any
indication (25% epilepsy, 27% psychiatric, 48% other, including pain) showing
an odds ratio of 1.8 (1.24 to 2.66) for suicidal behavior or ideation in individuals
exposed to AEDs.33 This analysis was limited by having been based on studies
that were not designed to assess psychiatric outcomes, studies with short
observation times, and studies that often excluded subjects with a significant
psychiatric history. Previously released expert consensus recommendations for
treatment of mild depression in patients with epilepsy had recommended use of
an AED with mood-stabilizing properties such as lamotrigine, valproate, or
carbamazepine and avoidance of those AEDs with an increased likelihood for
adverse mood side effects, reflecting the long-established clinical observations
that impact on mood varied by AED.10 A 2010 case-control study from the
United Kingdom Practice Research Database did find increased risk of suicidal
behavior in users of newer AEDs with a “high potential to cause depression”
(levetiracetam, tiagabine, topiramate, vigabatrin) as compared to “low potential”
AEDs (lamotrigine, gabapentin, pregabalin, oxcarbazepine).34 When the impact
of individual drugs was analyzed, only current use of levetiracetam was associated
with increased risk of suicidal thoughts and behavior.35 Providers should be aware
of the potential impact of a given AED to influence depression and anxiety when
making therapeutic decisions for epilepsy. The author of this article has found that
discussing these issues, including the FDA warning, when prescribing a new AED is
beneficial to reducing patient concerns and may help improve medication compliance.

IMPACT OF LIFESTYLE: FOCUS ON DRIVING

In a survey of people with epilepsy, driving was rated as their number one
concern impacting quality of life.34 Ability to drive can influence ability to work,
maintain relationships, and to live independently. It is not difficult to conceive
that a seizure with loss of awareness while driving could have catastrophic
consequences; however, a recent systematic review found insufficient evidence
to conclude that people with epilepsy are at greater risk for motor vehicle
crashes.36
One challenge in ascertainment of risk is that single-vehicle crashes without
major injury may not be reported to law enforcement agencies or to health care
providers. Self-reporting surveys provided to drivers with epilepsy have the
potential to capture the frequency of such events. One such survey in Canada
found no difference in the rates of motor vehicle crashes in drivers with and
without epilepsy.37 A self-reporting survey of adults with epilepsy living in
Arizona and New Jersey found 5% to 11% reported having a crash related to a
seizure.38 Studies on publicly reported crashes also paint an inconsistent picture
of the true risk of driving with epilepsy. A recent study from Maryland
evaluating crash reports for 254 patients with epilepsy approved to drive by the
Maryland Motor Vehicle Administration found only two seizure-related crashes
in 7 years, suggesting a very low risk for drivers meeting local legal licensing
requirements.39 In contrast, a Canadian population-based study of motor
vehicle crashes leading to medical care found an increased risk for drivers with
epilepsy (odds ratio 1.62, 95% confidence interval 0.95–2.76).40
Driving is also highly regulated for people with epilepsy compared to other
medical conditions that could impact driving, although epilepsy may not
represent the condition of greatest threat to on-the-road safety. A study of fatal
US car crashes found that 0.2% were related to epilepsy as compared to 4%
related to cardiovascular disease and 30% related to alcohol.41,42 The risk of
crashes for drivers with epilepsy should also be placed in perspective to other
categories of drivers. For example, in the United States, the fatal crash rate per
100,000 population is estimated at 8.6 for those with epilepsy as compared to
28.1 in all drivers younger than 25 years of age.41 Regardless of absolute risk, every
state has laws regulating driving for people with seizures. Specific requirements
and processes vary between states.43 Driving privileges in the United States are
primarily determined by the patient’s duration of seizure freedom. The optimal
duration of seizure freedom is not established, and standards typically range
between 3 and 12 months.43 An analysis of motor vehicle crash reports in Arizona
before and after the required seizure-free interval was changed from 12 to
3 months found no significant change in the rate of seizure-related crashes and a
decrease in the number of such crashes resulting in fatality.44 In a 2007 position
statement, the American Academy of Neurology (AAN) endorsed a 3-month
seizure-free period.45 Some state laws also consider modification of driving
restrictions based on special circumstances such as a consistent prolonged aura,
seizures that are exclusively nocturnal, seizures provoked by a prescribed
medication change, or seizure symptomatology that is unlikely to impact safe
driving; these exceptions are supported in the AAN position statement.43,45
Patients with epilepsy require accurate information about driving if they are to
be protected from physical and legal harm. Physicians are generally poor at
counseling patients with epilepsy regarding driving laws.38,46 In one survey,
one-third to one-half of adults with epilepsy stated they had never been
482 APRIL 2019

counseled by a medical provider about driving.38 Chart review of a single US KEY POINTS
tertiary care hospital found that among people seen by an emergency department
● People living with
provider for a seizure or syncope, less than 10% were documented to have epilepsy rated their ability to
received counseling on driving.46 Providers may be challenged to balance drive as the most important
advocating for the needs of an individual patient with promoting the safety of the factor determining quality
individual and the public, a dilemma that is illustrated in CASE 9-2. Physicians are of life.
strongly encouraged to know their relevant state driving laws and to follow them.
● Physicians are
This includes completing necessary forms as required by the Department of underperforming in
Motor Vehicles and reporting as mandated by law in selected states (TABLE 9-2). counseling patients with
For medicolegal reasons, it is also important to document counseling of epilepsy about driving.
the patient regarding driving and the reasons for individual driving
● Fracture risk is 2 to 6
recommendations. Refer to the article “Driving and Epilepsy: Ethical, Legal, and times higher in people with
Health Care Policy Challenges” by Joseph S. Kass, MD, JD, FAAN, and Rachel V. epilepsy than in the general
Rose, JD, MBA,47 in this issue of Continuum for further information about the population, and fracture risk
medicolegal issues for the provider with regard to driving and epilepsy. increases with duration of
antiepileptic drug therapy.
PHYSICAL HEALTH
People with epilepsy are at increased risk of accidents and injuries,48 which, to a
large degree, is directly attributable to seizures causing immediate harm.45
Adverse effects of AEDs on motor coordination and alertness and comorbid
cognitive or physical disabilities can also contribute. Seizures often cause injuries
due to falls. Additional risk factors for seizure-related injury include uncontrolled
epilepsy, seizures with altered awareness, medication noncompliance, and
seizures occurring when a person is alone. Generalized tonic-clonic seizures can
result in shoulder dislocation, vertebral compression fractures, and tongue
lacerations. Seizures with altered awareness of the environmental dangers can
lead to burns, drowning, and motor vehicle crashes. Most seizure-related injuries
are mild to moderate in severity and commonly include lacerations, fractures,
dental injuries, concussions, and burns.49 Severe injuries such as subdural
hematoma or death due to drowning do occur, but infrequently.49
Among many suggested strategies to reduce injury risk, the most important
intervention is improved seizure control. Other methods to reduce injuries are
summarized in TABLE 9-3. Risk prevention strategies should be individualized
and include consideration of epilepsy-related factors such as seizure type,
frequency, timing (sleep versus awake), and triggers. It is further important to
consider the age, independence, and cultural and social norms of the patient and
family. For example, recommending a parent sleep with a young child with
frequent nocturnal convulsions might be an acceptable and reasonable safety
precaution but is an unnecessary and unacceptable recommendation if the
patient is a college-age young adult with rare seizures living in the family home.
Excessive safety precautions may also discourage people with epilepsy from
participating in activities that are part of a healthy lifestyle, such as physical
exercise, and may keep the patient from achieving independence.
Fractures are of particular concern in patients with epilepsy. The rate of
fractures in both children and adults with epilepsy is estimated to be 2 to 6 times
higher than in the general population.50–52 Fractures are partially attributable to
falls, but use of an AED also plays a role. Fracture risk increases with duration of
AED exposure.51,53 By one estimate, for every decade of AED use, risk of any
fracture increased by 40% and risk of seizure-related fracture increased by
60%.53 Women on AEDs are at higher risk for both falls and fractures as

CASE 9-2 A 21-year-old man with juvenile myoclonic epilepsy presented for an
outpatient visit after a seizure-related motor vehicle crash. He had been
diagnosed with juvenile myoclonic epilepsy at age 15 and was initially
seizure free on valproic acid. During college he had several episodes of
morning myoclonus provoked by binge alcohol use and sleep deprivation.
He related that in the last 3 months he had experienced three generalized
tonic-clonic seizures triggered by alcohol use and minimal sleep. He
had often stayed up late because of schoolwork and socializing. In
addition, he admitted to frequently missing doses of his medication.
The most recent seizure occurred while driving to a 7:00 AM college
class and resulted in major damage to the vehicle but no personal injury.
His serum valproate level was found to be subtherapeutic in the
emergency department, where he had been transported after the
collision.
At the current presentation, he brought a letter from the Department of
Motor Vehicles requesting a medical report from his physician regarding
his fitness to drive. He stressed the importance of his being able to drive
to get to school and work. The neurologist explained that according to
the state law where this patient resided, he must be seizure free for a
period of at least 90 days before he can legally operate a motor vehicle.
The patient and neurologist reviewed the importance of compliance
with the law both for his safety and the safety of the general public. They
discussed strategies to improve seizure control including medication
compliance, improved sleep schedule, and avoidance of alcohol. In
addition, the neurologist communicated with the patient’s university,
requesting accommodation with class sessions starting later in the day.
She explained that if he was seizure free in 3 months and was compliant
with medication and lifestyle changes, that she would work with him at
that time to file an updated medical report for the Department of Motor
Vehicles. The neurologist documented in the electronic medical record
the recommendations and counseling that were provided regarding
driving status.
COMMENT This case illustrates the provider’s roles in caring for the driver with
epilepsy and exemplifies how responsibilities toward the patient and
public may sometimes be in conflict. In the context of the doctor-patient
relationship, the provider has an obligation to optimize seizure control and
the patient’s quality of life. As a patient advocate, the physician should
promote the patient’s independence, including facilitating attendance at
school and work, but also promote personal safety. Finally, the physician
has an obligation to maintain public safety and has a responsibility and
liability under the law, including a duty to accurately inform and counsel the
patient about relevant driving laws and to file required reports to
government agencies. The physician can also help protect herself from
liability by carefully documenting these steps in the medical record.
484 APRIL 2019

compared to men.51,53,54 In appropriately selected cases in which seizure freedom
is likely without medication, timely discontinuation of chronic AED therapy may
help lower future fracture risk. In addition to concern about impact of chronic
AED use on bone, people with epilepsy commonly have concerns about the
adverse effects of AEDs on employment, education, sex life, driving, emotion,
and cognitive function.55
The association between AED use and fracture is largely attributed to adverse
drug effects on bone density and quality.54,56 One of the most commonly cited
explanations for this relationship is alteration of vitamin D metabolism by
induction of hepatic cytochrome P450 enzymes, resulting in reduced calcium
absorption. Supporting this theory, a large cohort study of more than 60,000
people with epilepsy on active treatment found the rate of fractures was
significantly higher in those taking hepatic enzyme–inducing AEDs compared to
non–enzyme-inducing AEDs.54 However, other studies have shown increased
rates of bone density loss and fracture compared to the general population in
people who used either enzyme-inducing AEDs or non–enzyme-inducing
AEDs.51,56,57 In a tertiary epilepsy center, more than half of patients on hepatic
enzyme–inducing AEDs and one-third on non–enzyme-inducing AEDs had
vitamin D deficiency.58 Published evidence evaluating the level of risk to bone
health associated with an individual AED is very limited, particularly for
newer-generation agents.56
Presently, no evidence-based guidelines address prevention, screening, or
treatment of low bone density in people with epilepsy. One prospective
treatment trial was conducted in men on chronic AED therapy and compared use
of calcium and vitamin D supplementation alone or in combination with
risedronate.59 Calcium and vitamin D alone were effective in improving bone
mineral density but less so than with the addition of risedronate. The group
treated with risedronate also had a lower rate of vertebral fracture.59 Common
practice is to recommend calcium and vitamin D supplementation to patients
taking AEDs as a simple, low-cost intervention for prevention and potential
treatment of AED-related bone disease, with higher doses of vitamin D (such
as 2000 IU per day) recommended for those taking enzyme-inducing AEDs.
Measurement of serum vitamin D levels and parathyroid hormone can be
considered, but again there is no consensus about when or how often.
Additionally, no consensus or evidence indicates when and in whom to
consider screening with bone mineral density analysis. Given the complexity of
States Requiring Mandatory Reporting of People With Seizures to the TABLE 9-2
Motor Vehicle Administration
◆ California
◆ Delaware
◆ Nevada
◆ New Jersey
◆ Oregon
◆ Pennsylvania

diagnostic and management decisions, involvement of an endocrinologist may

be considered.
Mortality is also increased for people with epilepsy. Population-based studies
have found standardized mortality ratios of 1.6 to 3.0 for adults with epilepsy and
6.4 to 7.5 for the subgroup with childhood-onset epilepsy.60 Mortality is often
related to medical conditions, of which seizures are symptomatic. A 30-year
population-based study of childhood-onset epilepsy in Olmsted County,
Minnesota, found that death was primarily caused by complications of disease
underlying epilepsy or from causes unrelated to epilepsy.61 Mortality due to a
seizure was rare and similar to that expected in the general population.61 In
children and adults, death can occur from seizure-related injuries and accidents,
including drowning, falls, and motor vehicle crashes.60 As noted earlier, the risk
of suicide is also increased in people with epilepsy.7–9 Noncompliance with
prescribed AED therapy has been linked to increased risk of mortality.62
The most common cause of mortality directly associated with a seizure is
sudden unexpected death in epilepsy (SUDEP).63 SUDEP is a sudden,
unexpected, nontraumatic, nondrowning death in a person with epilepsy where
autopsy reveals no other anatomic or toxicologic cause of death. Most
occurrences of SUDEP are unwitnessed, but victims are often reported to have
had a known or suspected generalized tonic-clonic seizure shortly prior to death.
Based on a recent evidence-based review, the incidence of SUDEP in childhood
is 0.22/1000 patient-years (95% confidence interval 0.16–0.31) and in adults is
1.2/1000 patient-years (95% confidence interval 0.64–2.32).64 This corresponds
to 1:4500 children with epilepsy in 1 year and 1:1000 adults with epilepsy in 1 year.
Several risk factors for SUDEP have been identified (TABLE 9-4).64 The strongest
association is with uncontrolled generalized tonic-clonic seizures. Having one to
two convulsions per year carries a fivefold increased risk for SUDEP, and three or
more convulsions per year carries a 15-fold increased risk.65
Physicians do not necessarily agree on how, when, and with whom to discuss
SUDEP. Chart review indicates that documentation of SUDEP discussion by
providers is rare.66 Reasons cited by providers to avoid this discussion include the
TABLE 9-3 Selected Strategies to Reduce Risk of Seizure-related Injury
◆ Wear a helmet when biking or horseback riding

◆ No unsupervised swimming
◆ Use a shower rather than a bathtub
◆ Adjust the temperature on the hot water heater to lower the risk of scalding
◆ Use a microwave versus a stovetop
◆ Avoid locking the bathroom or bedroom door
◆ Place bed mattress on the floor
◆ Use an epilepsy safety pillow (designed to reduce risk of suffocation if lying face down)
◆ Use soft or padded furniture
◆ Avoid high ladders
◆ Follow driving regulations for people with epilepsy
◆ Take medication as prescribed
486 APRIL 2019

generally low risk and the potential to cause unnecessary anxiety and stress.67 In
a survey of families who had experienced SUDEP, only 18% recalled being
informed about SUDEP, but 72% wished they had received this information from
their physician.68 Only 10% preferred no counseling.68 In surveys of adults with
epilepsy and parents of children with epilepsy, the majority of respondents
indicated they want this information from their neurologist.69,70 In 2017, the
AAN issued a practice guideline on SUDEP, which includes recommendations
about how to counsel patients and families.64 Risk of SUDEP can be reduced
through nocturnal supervision, either direct or via a listening device, and
through improved medication compliance.64 As with other safety restrictions,
recommendations should take into consideration the specifics of the individual’s
epilepsy as well as social and cultural norms.
SOCIAL HEALTH: RELATIONSHIPS AND GENETIC BURDEN

Epilepsy can negatively impact relationships with family, caregivers, friends,
schoolmates, and coworkers. In a survey of more than 1000 people with epilepsy
living in the community, the diagnosis was reported to negatively affect
relationships with others in 32%, relationship with spouses or partners in 28%,
and relationship between patients and their children in 25%55; 21% of patients
reported that epilepsy negatively affected their sex lives.55 Men and women with
epilepsy are less likely to be married and less likely to have children regardless of
marital status.71–73 The lower birth rate for people with epilepsy likely reflects
both social and biological factors. Potential barriers include financial restrictions
due to underemployment, social isolation, adverse effects of seizures and AEDs
on sexual desire and reproductive endocrine function, and concerns about AED
use during pregnancy causing birth defects. Some may also worry about
epilepsy heredity.
Surveys of adults with epilepsy in both the United States and Korea showed
that patients perceive the risk of passing on epilepsy to a child as much higher
than the actual risk, and that this concern contributed to decisions to have fewer
children.74,75 A population-based study in Olmsted County, Minnesota, found
that first-degree relatives of people with epilepsy have a threefold increased risk
of having seizures.76 Epilepsy is, of course, not a single disease and encompasses
many etiologies ranging from those with little to no genetic risk to catastrophic
genetic disorders with serious implications for family planning. Even among
Risk Factors for Sudden Unexpected Death in Epilepsya TABLE 9-4
Factor Strength of Evidence
Presence of generalized tonic-clonic seizures (present versus not Moderate

present)
Frequency of generalized tonic-clonic seizures >3 per year High
Uncontrolled seizures Moderate
Failure to adjust medication for treatment-refractory seizures Moderate
a
Modified with permission from Harden C, et al, Neurology.64 © 2017 American Academy of Neurology.

KEY POINTS hereditary forms of epilepsy, inheritance is often complex and multifactorial.
The task of appropriately counseling individuals who seek to know the risk of
● Evidence-based or
expert consensus
passing on epilepsy to a child can thus be highly complex. Genetic testing for
recommendations on epilepsy diagnosis is reviewed in the article “Evaluation of Seizure Etiology From
prevention, screening, and Routine Testing to Genetic Evaluation” by Stephan U. Schuele, MD, MPH,
treatment of bone disease FAAN,77 in this issue of Continuum. For some individuals and their families,
and fractures in people with
genetic testing can provide important information about etiology, treatment, and
epilepsy are lacking.
Medical providers must rely prognosis and thus improve overall life quality. However, genetic testing
on common sense or can also have negative financial, emotional, and social consequences. The
strategies used in the International League Against Epilepsy endorsed guidelines on genetic testing
general population.
in 2010 that address these concerns.78 Before proceeding with a genetic
● Mortality in patients with investigation, clinicians should consider the test accuracy, clinical validity,
epilepsy is primarily clinical utility, and utility to the individual with epilepsy and must be prepared to
attributable to underlying provide counsel on the implications of both a positive and negative result.79
medical disorders, rather The task of counseling and test selection can be daunting to many neurologists,
than to seizure-related
injury or sudden unexpected
and referral to a geneticist or epileptologist with expertise in this area may
death in epilepsy. be warranted.
● Uncontrolled generalized
tonic-clonic seizures are an
CONCLUSION
important risk factor for
sudden unexpected death The repercussions of epilepsy can extend well beyond the isolated moments of
in epilepsy. time in which seizures occur. Negative implications for emotional, physical, and
social health can significantly adversely affect the quality of life and “wellness” of
● The majority of people the person with epilepsy, as well as that of their family, and has a high cost to
with epilepsy want to be
informed by their physician society. Physicians are committed to promoting the health of patients under their
about sudden unexpected care and should advocate for their well-being and independence, but to do so
death in epilepsy. successfully, providers must be well versed in the risks of epilepsy beyond
seizures to improve detection, treatment, and provide appropriate counsel. For
● People with epilepsy tend
to overestimate the risk of
many areas including depression treatment, SUDEP, bone health, and fractures,
passing on epilepsy to a additional investigation and evidence to guide management is greatly needed.
child, which may falsely
influence decisions about
having children.
USEFUL WEBSITES
CENTERS FOR DISEASE CONTROL AND PREVENTION EPILEPSY FOUNDATION STATE DRIVING LAWS DATABASE
MANAGING EPILEPSY WELL NETWORK A searchable database of state driving laws for
The Managing Epilepsy Well Network website has epilepsy.
information for providers and patients about epilepsy.com/driving-laws
several self-management interventions including
PEARLS (Program to Encourage Active, Rewarding WEBEASE (EPILEPSY, AWARENESS, SUPPORT, AND
Lives) and Project UPLIFT (Using Practice and EDUCATION)
Learning to Increase Favorable Thoughts). WebEase (Epilepsy, Awareness, Support, and
cdc.gov/epilepsy/research/MEW-network.htm Education) is a free online self-management
program for adults with epilepsy designed to help
set and achieve goals.
webease.org/
REFERENCES
1 England MJ. Institute of Medicine committee of 2 Begley CE, Famulari M, Annegers JF, et al. The
the public health dimensions of the epilepsies. cost of epilepsy in the United States: an estimate
Epilepsy across the spectrum: promoting health from population-based clinical and survey
and understanding. Washington DC: National data. Epilepsia 2000;41:342–351. doi:10.1111/
Academies Press, 2012. j.1528-1157.2000.tb00166.x.
488 APRIL 2019

3 McDermott S, Mani S, Krishnaswami S. A 16 Fiest KM, Patten SB, Altura KC, et al. Patterns and
population-based analysis of specific behavior frequency of the treatment of depression in
problems associated with childhood seizure persons with epilepsy. Epilepsy Behav 2014;39:
disorders. J Epilepsy 1995;8:110–118. 54–64. doi:10.1016/j.yebeh.2014.08.023.
doi:10.1016/0896-6974(95)00019-A.
17 Alper K, Schwartz KA, Kolts RL, Khan A. Seizure
4 Fiest KM, Dykeman J, Patten SB, et al. Depression incidence in psychopharmacological clinical
in epilepsy: a systematic review and meta- trials: an analysis of Food and Drug Administration
analysis. Neurology 2013;80:590–599. doi:10.1212/ (FDA) summary basis of approval reports. Biol
WNL.0b013e31827b1ae0. Psychiatry 2007;62(4):345–354. doi:10.1016/
j.biopsych.2006.09.023.
5 Scott AJ, Sharpe L, Hunt C, Gandy M. Anxiety and
depressive disorders in people with epilepsy: a 18 Johnston JA, Lineberry CG, Ascher JA, et al.
meta-analysis. Epilepsia 2017;58(6):973–982. A 102-center prospective study of seizure in
doi:10.1111/epi.13769. association with bupropion. J Clin Psychiatry 1991;
52(11):450–456.
6 Pham T, Sauro KM, Patten SB, et al. The
prevalence of anxiety and associated factors in 19 Robertson MM, Trimble MR. The treatment
persons with epilepsy. Epilepia 2017;58(8): of depression in patients with epilepsy.
e107–e110. doi:10.1111/epi.13817. A double-blind trial. J Affect Disord 1985;9(2):
127–136. doi:10.1016/0165-0327(85)90091-6.
7 Christensen J, Vestergaard M, Mortensen PB,
et al. Epilepsy and risk of suicide: a population- 20 Zhu S, Luo L, Gui Y, et al. Short-term efficacy of
based case-control study. Lancet Neurol 2007; venlafaxine treating the depression in epilepsy
6(8):693–698. doi:10.1016/S1474-4422(07)70175-8. patients. Chin J Rehabil 2004:19:101.
8 Karouni M, Arulthas S, Larsson PG, et al. 21 Li W, Ma D. A randomized controlled trial to
Psychiatric comorbidity in patients with epilepsy: evaluate the efficacy of paroxetine and doxepin
a population-based study. Eur J Clin Pharmacol in treating epileptic patients with depression.
2010;66(11):1151–1160. doi:10.1007/s00228-010-0861-y. Chin J Clin Rehabil 2005;9(12):20–21.
9 Tian N, Cui W, Zack M, et al. Suicide among 22 Orjuela-Rojas JM, Martínez-Juárez IE, Ruiz-Chow
people with epilepsy: a population-based A, Crail-Melendez D. Treatment of depression in
analysis of data from the U.S. National Violent patients with temporal lobe epilepsy: a pilot
Death Reporting System, 17 states, 2003-2011. study of cognitive behavioral therapy vs.
Epilepsy Behav 2016;61:210–217. doi:10.1016/ selective serotonin reuptake inhibitors.
j.yebeh.2016.05.028. Epilepsy Behav 2015;51:176–181. doi:10.1016/
j.yebeh.2015.07.033.
10 Barry JJ, Ettinger AB, Friel P, et al. Consensus
statement: the evaluation and treatment of 23 McLaughlin DP, McFarland K. A randomized trial
people with epilepsy and affective disorders. of a group based cognitive behavior therapy
Epilepsy Behav 2008;13(suppl 1):S1–S29. program for older adults with epilepsy: the
doi:10.1016/j.yebeh.2008.04.005. impact on seizure frequency, depression and
psychosocial well-being. J Behav Med 2011;34(3):
11 Josephson CB, Lowerison M, Vallerand I, et al.
201–207. doi:10.1007/s10865-010-9299-z.
Association of depression and treated depression
with epilepsy and seizure outcomes: a multicohort 24 Chaytor N, Ciechanowski P, Miller JW, et al.
analysis. JAMA Neurol 2017;74(5):533–539. Long-term outcomes from the PEARLS
doi:10.1001/jamaneurol.2016.5042. randomized trial for the treatment of depression
in patients with epilepsy. Epilepsy Behav 2011;
12 Ettinger AB, Good MB, Manjunath R, et al. The
20(3):545–549. doi:10.1016/j.yebeh.2011.01.017.
relationship of depression to antiepileptic drug
adherence and quality of life in epilepsy. 25 Gandy M, Sharpe L, Nicholson Perry K, et al.
Epilepsy Behav 2014;36:138–143. doi:10.1016/ Cognitive behaviour therapy to improve mood in
j.yebeh.2014.05.011. people with epilepsy: a randomised controlled
trial. Cognitive Behav Therapy 2014;43(2):153–166.
13 Stern JM, Labiner DM, Gilliam FG, et al. More
doi:10.1080/16506073.2014.892530.
effective assessment of adverse effects and
comorbidities in epilepsy: results of a Phase II 26 Thompson NJ, Patel AH, Selwa LM, et al.
communication study. Epilepsy Behav 2011;22(3): Expanding the efficacy of Project UPLIFT:
552–556. doi:10.1016/j.yebeh.2011.06.007. distance delivery of mindfulness-based
depression prevention to people with epilepsy.
14 Gill SJ, Lukmanji S, Fiest KM, et al. Depression
J Consult Clin Psychol 2015;83(2):304–313.
screening tools in persons with epilepsy: a
doi:10.1037/a0038404.
systematic review of validated tools. Epilepsia
2017;58(5):695–705. doi:10.1111/epi.13651. 27 Kühn KU, Quednow BB, Thiel M, et al.
Antidepressive treatment in patients with
15 Gilliam FG, Fessler AJ, Baker G, et al. Systematic
temporal lobe epilepsy and major depression:
screening allows reduction of adverse antiepileptic
a prospective study with three different
drug effects: a randomized trial. Neurology 2004;
antidepressants. Epilepsy Behav 2003;4(6):
62(1):23–27.
674–679. doi:10.1016/j.yebeh.2003.08.009.

28 Kanner AM, Kozak AM, Frey M. The use of 41 Sheth SG, Krauss G, Krumholz A, Li G. Mortality
sertraline in patients with epilepsy: is it safe? in epilepsy: driving fatalities vs other causes
Epilepsy Behav 2000;1(2):100–105. doi:10.1006/ of death in patients with epilepsy. Neurology
ebeh.2000.0050. 2004;63(6):1002–1007. doi:10.1212/01.WNL.
0000138590.00074.9A.
29 Hovorka J, Herman E, Nemcova I. Treatment of
interictal depression with citalopram in patients 42 Sauber-Schatz EK, Ederer DJ, Dellinger AM,
with epilepsy. Epilepsy Behav 2000;1(6):444–447. Baldwin GT. Vital signs: motor vehicle injury
doi:10.1006/ebeh.2000.0123. prevention – United States and 19 comparison
countries. MMWR Morb Mortal Wkly Rep 2016;
30 Specchio LM, Iudice A, Specchio N, et al.
65(26):672–677. doi:10.15585/mmwr.mm6526e1.
Citalopram as treatment of depression in
patients with epilepsy. Clin Neuropharmacol 43 Krumholz A, Hopp JL, Sanchez AM. Counseling
2004;27(3):133–136. epilepsy patients on driving and employment.
Neurol Clin 2016;34(2):427–442, ix. doi:10.1016/
31 Kerr MP, Mensah S, Besag F, et al. International
j.ncl.2015.11.005.
consensus clinical practice statements for the
treatment of neuropsychiatric conditions 44 Drazkowski JF, Fisher RS, Sirven JI, et al.
associated with epilepsy. Epilepsia 2011;52(11): Seizure-related motor vehicle crashes in Arizona
2133–2138. doi:10.1111/j.1528-1167.2011.03276.x. before and after reducing the driving restriction
from 12 to 3 months. Mayo Clin Proc 2003;78(7):
32 Miller JM, Kustra RP, Vuong A, et al. Depressive
819–825. doi:10.4065/78.7.819.
symptoms in epilepsy: prevalence, impact,
aetiology, biological correlates and effect of 45 Bacon D, Fisher RS, Morris JC, et al. American
treatment with antiepileptic drugs. Drugs 2008; Academy of Neurology position statement on
68(11):1493–1509. physician reporting of medical conditions that
may affect driving competence. Neurology 2007;
33 Statistical review and evaluation: antiepileptic
68(15):1174–1177. doi:10.1212/01.
drugs and suicidality. fda.gov/downloads/
wnl.0000259514.85579.e0.
drugs/drugsafety/postmarketdrugsafety
informationforpatientsandproviders/ucm192556. 46 Shareef YS, McKinnon JH, Gauthier SM, et al.
pdf. Published May 23, 2008. Accessed Counseling for driving restrictions in epilepsy and
January 28, 2019. other causes of temporary impairment of
consciousness: how are we doing? Epilepsy
34 Gilliam F, Kuzniecky R, Faught E, et al.
Behav 2009;14(3):550–552. doi:10.1016/
Patient-validated content of epilepsy-specific
j.yebeh.2008.12.020.
quality-of-life measurement. Epilepsia 1997;38(2):
233–236. doi:10.1111/j.1528-1157.1997.tb01102.x. 47 Kass JS, Rose RV. Driving and epilepsy: ethical,
legal, and health care policy challenges.
35 Andersohn F, Schade R, Willich SN, Garbe E. Use
Continuum 2019;25(2, Epilepsy):537–542.
of antiepileptic drugs in epilepsy and the risk of
doi:10.1212/CON.0000000000000714.
self-harm or suicidal behavior. Neurology 2010;
75(4):335–340. doi:10.1212/WNL.0b013e3181ea157e. 48 Beghi E, Cornaggia C, RESt-1 Group. Morbidity
and accidents in patients with epilepsy: results
36 Naik PA, Fleming ME, Bhatia P, Harden CL. Do
of a European cohort study. Epilepsia 2002;43(9):
drivers with epilepsy have higher rates of motor
1076–1083. doi:10.1046/j.1528-1157.2002.18701.x.
vehicle accidents than those without epilepsy?
Epilepsy Behav 2015;47:111–114. doi:10.1016/ 49 Camfield C, Camfield P. Injuries from seizures are
j.yebeh.2015.04.016. a serious, persistent problem in childhood onset
epilepsy: a population-based study. Seizure
37 McLachlan RS, Starreveld E, Lee MA. Impact of
2015;27:80–83. doi:10.1016/j.seizure.2015.02.031.
mandatory physician reporting on accident risk
in epilepsy. Epilepsia 2007;48(8):1500–1505. 50 Souverein PC, Webb DJ, Petri H, et al. Incidence
doi:10.1111/j.1528-1167.2007.01051.x. of fractures among epilepsy patients: a
population-based retrospective cohort study in
38 Drazkowski JF, Neiman ES, Sirven JI, et al.
the General Practice Research Database.
Frequency of physician counseling and attitudes
Epilepsia 2005;46(2):304–310. doi:10.1111/j.0013-
toward driving motor vehicles in people with
9580.2005.23804.x.
epilepsy: comparing a mandatory-reporting with
a voluntary-reporting state. Epilepsy Behav 2010; 51 Souverein PC, Webb DJ, Weil JG, et al. Use of
19(1):52–54. doi:10.1016/j.yebeh.2010.06.041. antiepileptic drugs and risk of fractures:
case-control study among patients with
39 Ma BB, Bloch J, Krumholz A, et al. Regulating
epilepsy. Neurology 2006;66(9):1318–1324.
drivers with epilepsy in Maryland: results of the
doi:10.1212/01.wnl.0000210503.89488.88.
application of a United States consensus
guideline. Epilepsia 2017;58(8):1389–1397. 52 Gniatkowska-Nowakowska A. Fractures in
doi:10.1111/epi.13804. epilepsy children. Seizure 2010;19(6):324–325.
doi:10.1016/j.seizure.2010.04.013.
40 Kwon C, Liu M, Quan H, et al. Motor vehicle
accidents, suicides, and assaults in epilepsy: a 53 Shiek Ahmad B, Hill KD, O’Brien TJ, et al. Falls and
population-based study. Neurology 2011;76(9): fractures in patients chronically treated with
801–806. doi:10.1212/WNL.0b013e31820e7b3b. antiepileptic drugs. Neurology 2012;79(2):145–151.
doi:10.1212/WNL.0b013e31825f0466.
490 APRIL 2019

54 Nicholas JM, Ridsale L, Richardson MP, et al. 66 Waddell B, McColl K, Turner C, et al. Are we
Fracture risk with use of liver enzyme inducing discussing SUDEP?–A retrospective case note
antiepileptic drugs in people with active analysis. Seizure 2013;22(1):74–76. doi:10.1016/
epilepsy: cohort study using the general practice j.seizure.2012.09.017.
research database. Seizure 2013;22(1):37–42.
67 Miller WR, Young N, Friedman D, et al. Discussing
doi:10.1016/j.seizure.2012.10.002.
sudden unexpected death in epilepsy (SUDEP)
55 Fisher RS, Vickrey BG, Gibson P, et al. The impact with patients: practices of health-care providers.
of epilepsy from the patient’s perspective I. Epilepsy Behav 2014;32:38–41. doi:10.1016/
Descriptions and subjective perceptions. j.yebeh.2013.12.020.
Epilepsy Res 2000;41(1):39–51. doi:10.1016/
68 Louik J, Doumlele K, Hussain F, et al. Experiences
S0920-1211(00)00126-1.
with premorbid SUDEP discussion among
56 Vestergaard P, Rejnmark L, Mosekilde L. Fracture participants in the North American SUDEP
risk associated with use of antiepileptic drugs. Registry (NASR). Epilepsy Behav 2017;70(pt A):
Epilepsia 2004;45(11):1330–1337. doi:10.1111/ 131–134. doi:10.1016/j.yebeh.2017.02.027.
j.0013-9580.2004.18804.x.
69 Ramachandrannair R, Jack SM, Meaney BF,
57 Ensrud KE, Walczak TS, Blackwell TL, et al. Ronen GM. SUDEP: what do parents want to
Antiepileptic drug use and rates of hip bone loss know? Epilepsy Behav 2013;29(3):560–564.
in older men: a prospective study. Neurology doi:10.1016/j.yebeh.2013.09.040.
2008;71(10):723–730. doi:10.1212/01.
70 RamachandranNair R, Jack SM. SUDEP: what
wnl.0000324919.86696.a9.
do adult patients want to know? Epilepsy
58 Teagarden DL, Meador KJ, Loring DW. Low Behav 2016;64(pt A):195–199. doi:10.1016/
vitamin D levels are common in patients with j.yebeh.2016.09.022.
epilepsy. Epilepsy Res 2014;108(8):1352–1356.
71 Schupf N, Ottman R. Likelihood of pregnancy in
doi:10.1016/j.eplepsyres.2014.06.008.
individuals with idiopathic/cryptogenic epilepsy:
59 Lazzari AA, Dussault PM, Thakore-James M, et al. social and biologic influences. Epilepsia 1994;
Prevention of bone loss and vertebral fractures 35(4):750–756. doi:10.1111/j.1528-1157.1994.tb02506.x.
in patients with chronic epilepsy–antiepileptic
72 Artama M, Isojärvi J, Raitanen J, Auvinen A. Birth
drug and osteoporosis prevention trial. Epilepsia
rate among patients with epilepsy: a nationwide
2013;54(11):1997–2004. doi:10.1111/epi.12351.
population-based cohort study in Finland. Am J
60 Thurman DJ, Logroscino G, Beghi E, et al. The Epidemiol 2004;159(11):1057–1063. doi:10.1093/
burden of premature mortality of epilepsy in aje/kwh140.
high-income countries: a systematic review from
73 Kobau R, Zahran H, Thurman DJ, et al. Epilepsy
the mortality task force of the International
surveillance among adults–19 states, behavioral
League Against Epilepsy. Epilepsia 2017;58(1):
risk factor surveillance system, 2005. MMWR
17–26. doi:10.1111/epi.13604.
Surveil Summ 2008;57(6):1–20.
61 Nickels KC, Grossardt BR, Wirrell EC.
74 Lee SM, Nam HW, Kim EN, et al.
Epilepsy-related mortality is low in children: a
Pregnancy-related knowledge, risk perception,
30-year population-based study in Olmsted
and reproductive decision making of women
County, MN. Epilepsia 2012;53(12):2164–2171.
with epilepsy in Korea. Seizure 2013;22(10):
doi:10.1111/j.1528-1167.2012.03661.x.
834–839. doi:10.1016/j.seizure.2013.07.002.
62 Faught E, Duh MS, Weiner JR, et al.
75 Helbig KL, Bernhardt BA, Conway LJ, et al.
Nonadherence to antiepileptic drugs and
Genetic risk perception and reproductive
increased mortality: findings from the RANSOM
decision making among people with epilepsy.
study. Neurology 2008;71(20):1572–1578.
Epilepsia 2010;51(9):1864–1867. doi:10.1111/
doi:10.1212/01.wnl.0000319693.10338.b9.
j.1528-1167.2009.02507.x.
63 Tomson T, Beghi E, Sundqvist A, Johannessen SI.
76 Peljto AL, Barker-Cummings C, Vasoli VM, et al.
Medical risks in epilepsy: a review with focus on
Familial risk of epilepsy: a population-based
physical injuries, mortality, traffic accidents and
study. Brain 2004;137(pt 3):795–805. doi:10.1093/
their prevention. Epilepsy Res 2004;60(1):1–16.
brain/awt368.
doi:10.1016/j.eplepsyres.2004.05.004.
77 Schuele SU. Evaluation of seizure etiology from
64 Harden C, Tomson T, Gloss D, et al. Practice
routine testing to genetic evaluation. Continuum
guideline summary: sudden unexpected death in
2019;25(2, Epilepsy):322–342. doi:10.1212/
epilepsy incidence rates and risk factors: report
CON.0000000000000723.
of the guideline development, dissemination, and
implementation Subcommittee of the American 78 Ottman R, Hirose S, Jain S, et al. Genetic testing in
Academy of Neurology and the American the epilepsies–report of the ILAE genetics
Epilepsy Society. Neurology 2017;88(17):1674–1680. commission. Epilepsia 2010;51:655–670.
doi:10.1212/WNL.0000000000003685. doi:10.1111/j.1528-1167.2009.02429.x.
65 Hesdorffer DC, Tomson T, Benn E, et al. Combined 79 Poduri A, Shieldley BR, Shostak S, Ottman R.
analysis of risk factors for SUDEP. Epilepsia 2011; Genetic testing in the epilepsies-developments
52(6):1150–1159. doi:10.1111/j.1528-1167.2010.02952.x. and dilemmas. Nature Rev Neurol 2014;10(5):
293–299. doi:10.1038/nrneurol.2014.60.

REVIEW ARTICLE

Nonepileptic Episodic
ONLINE
Events
By Jennifer L. Hopp, MD, FAAN

VIDEO CONTENT
A VA I L A B L E O N L I N E
ABSTRACT
PURPOSE OF REVIEW: This review addresses the scope, evaluation, treatments,
and outcomes of patients with nonepileptic episodic events with a focus
on psychogenic nonepileptic seizures. Differentiation of the types of
events, including a review of terminology, is included, as well as a brief
review of special patient populations with these disorders.
CITE AS: RECENT FINDINGS: There are continued efforts to develop tools to improve the
CONTINUUM (MINNEAP MINN) diagnosis of these disorders. A thorough evaluation with trained personnel
2019;25(2, EPILEPSY):492–507.
and physicians knowledgeable in the assessment and treatment of these
disorders is important. Although inpatient video-EEG monitoring in an
Dr Jennifer L. Hopp, Department epilepsy monitoring unit remains the gold standard for diagnosis, the
of Neurology, University of assessment of clinical and historical factors is critical and can be useful in
Maryland School of Medicine,
110 S Paca St, 3S-131, Baltimore, expediting the process and improving diagnostic certainty. International
MD 21201, JHopp@som. efforts have recently assisted in providing guidelines for the evaluation of
umaryland.edu.
the psychogenic disorders and may help target educational and other
RELATIONSHIP DISCLOSURE: resources to underserved areas.
Dr Hopp has received grant/
research support as a site
SUMMARY: The prompt and accurate diagnosis of nonepileptic episodic
principal investigator of the
Established Status Epilepticus events and psychogenic nonepileptic seizures is possible with current
Treatment Trial from the National technology, and the appropriate and targeted use of evidence-based
Institute of Neurological
Disorders and Stroke and from treatments may help improve patient quality of life and avoid unnecessary
SAGE Therapeutics. Dr Hopp disability in patients with these disorders.
compensation as a speaker
for J. Kiffin Penry Epilepsy
MiniFellow Network’s Epilepsy
MiniFellowship and Residents INTRODUCTION
N
Epilepsy Program. Dr Hopp onepileptic episodic events are a group of relatively common disorders
receives publishing royalties
from UpToDate, Inc and has
that present a significant problem in the field of neurology regarding
given expert medical testimony challenges in diagnosis and treatment. Nonepileptic episodic
for Venable LLP. events may resemble epileptic seizures but are distinguished by
UNLABELED USE OF both differences in symptomatology as well as the lack of abnormal
PRODUCTS/INVESTIGATIONAL epileptiform activity on EEG. In the broadest sense, nonepileptic episodic
USE DISCLOSURE:
events may include any paroxysmal events characterized by changes in behavior,
Dr Hopp discusses the
unlabeled/investigational use of experience, sensation, or movement that resemble seizures and are of either
sertraline for the treatment of psychogenic or physiologic origin.
psychogenic nonepileptic
seizures.
It is important to note that the term nonepileptic episodic events is also often
used interchangeably with nonepileptic seizures or psychogenic nonepileptic seizures
(PNES), so caution should be taken to use precise terminology.
© 2019 American Academy Nonepileptic episodic events with a physiologic cause are common, and
of Neurology. prevalence is dependent on the specific diagnosis. PNES were traditionally
492 APRIL 2019

characterized as a behavioral or physical manifestation of a somatoform,
conversion, or dissociative disorder but are now considered to have a more
complex, multifactorial etiology, as in functional neurologic disorders in general
(FIGURE 10-11).
There are commonly delays of several years2,3 in the identification and
treatment of these disorders.4 Accurate diagnosis is key because many patients
are treated for years as if they have epilepsy, when in fact, they may have a
psychological or other physiologic disorder. Delays can lead to inappropriate
treatment, worsened quality of life,5 and overuse of health care resources.6
TERMINOLOGY
Although nonepileptic episodic events is intended to be a term with a broad
definition, it is also a commonly used term to describe PNES. There remains a
great deal of variability in the vocabulary used to describe this group of disorders.
Although some terms used to describe events that resemble epileptic seizures
have become obsolete, there is still a lack of uniformity in the language
physicians and patients use to describe these conditions. Some commonly used
terms, such as pseudoseizure,7 may have a negative connotation and should likely
be avoided. Some variability in language may reflect cultural and national
differences, but there is still not an internationally accepted term used to describe
this set of conditions. It is important to establish consistency of language to
ensure accuracy of diagnosis, good communication with patients, and clarity
regarding the etiology of the conditions. PNES is one of the more common terms
used to describe seizurelike events with a psychological etiology because many
FIGURE 10-1
Possible mechanisms underlying functional neurologic disorders.
Reprinted with permission from Voon V, et al, J Neuropsychiatry Clin Neurosci.1
© 2016 American Psychiatric Foundation.

NONEPILEPTIC EPISODIC EVENTS
experts feel that it is important to convey the psychological nature of these

events rather than the pure neurologic cause associated with epilepsy. The
term nonepileptic episodic events will be used here for the purposes of a broader
discussion and will include both physiologic nonepileptic events/seizures as
well as psychogenic nonepileptic events/seizures.
Specific terms will be used when needed to indicate different disorders. That
said, it is important to be aware that the term nonepileptic episodic events is often
considered synonymous with physiologic nonepileptic seizures in the literature.8
Much of this review will center on nonepileptic episodic events that are psychogenic
in origin, and the term PNES will be used to describe this disorder in this article.
HISTORICAL AND CLINICAL DIAGNOSIS OF NONEPILEPTIC

EPISODIC EVENTS
The diagnosis of nonepileptic episodic events can sometimes be made with a
thorough history and neurologic assessment, although often additional tools,
including video-EEG monitoring, cardiac monitoring, or other testing, may be
needed to confirm the diagnosis.
Physiologic Causes
Accurate and early diagnosis of all types of nonepileptic episodic events is critical
for the establishment of an appropriate and effective targeted treatment plan.
The broad term nonepileptic episodic events may include events of both
psychogenic as well as physiologic origin, and it is important to differentiate
between these groups. Nonepileptic episodic events of physiologic origin may
include both neurologic and non-neurologic conditions that could be mistaken
for epileptic seizures. Neurologic conditions may include conditions such as
migraine, sleep disorders (parasomnias), cerebrovascular disorders such as
transient ischemic attack, and movement disorders (eg, tremor, nonepileptic
myoclonus).9 Non-neurologic conditions that may be mistaken for epileptic
seizures include metabolic abnormalities, toxic ingestions, and cardiac
arrhythmias. Syncope may be of a neurologic or non-neurologic origin and may
be commonly mistaken for epileptic seizures, particularly when accompanied by
tonic or clonic movements.10
To differentiate physiologic causes of nonepileptic episodic events from those
of epileptic or psychogenic origin, it is key to take a thorough history with an
emphasis on a description of prodromal symptoms (diaphoresis and a feeling of
the world “closing in” may suggest syncope, although some cardiac causes of
syncope may have no prodrome; preserved consciousness may suggest migraine
or transient ischemic attack). A description of the postictal period is useful
because a rapid return to consciousness is not typical in epileptic events, although
it can sometimes be seen with frontal lobe epilepsy. Careful attention to the
presence or absence of epilepsy risk factors is important, including complex
febrile seizures, history of significant head trauma, history of brain infections,
and family history of epilepsy. An eyewitness account of the event can be
particularly useful, although it should be considered with caution. It is very
important when analyzing historical and clinical features of nonepileptic episodic
events to remember that no single clinical sign or examination finding can
distinguish between the types of nonepileptic episodic events, including
differentiation between nonepileptic episodic events/PNES and
epileptic seizures.
494 APRIL 2019

Although there is no single, widely used, effective screening measure, there KEY POINTS
are efforts to create both historical11 as well as clinical tools12 to assist in the
● The term nonepileptic
prehospital assessment of nonepileptic episodic events to further improve episodic events is broad and
accuracy and reduce delays in diagnosis. includes disorders of both
physiologic and psychogenic
Psychogenic Causes origin. Nonepileptic events
of a psychogenic origin
Nonepileptic episodic events that are of psychogenic origin are commonly
are often referred to as
referred to as psychogenic nonepileptic events or PNES. These are paroxysmal psychogenic nonepileptic
episodes of altered awareness, movement, or sensation that mimic epileptic seizures.
seizures but are not associated with concomitant epileptiform abnormalities on
an EEG. There are many features of the medical history of patients with PNES ● Syncope may be mistaken
for an epileptic seizure when
that can be suggestive of this disorder. Most features described here have followed by tonic-clonic
important exceptions, and great caution should be taken not to assume a patient movements (convulsive
has PNES based on one or many of these features of their history. In addition, syncope), and a thorough
patients who may be thought to have comorbid epileptic seizures and PNES may evaluation should be
performed to exclude
have components of their history or description of events that could be physiologic causes of loss
suggestive of either disorder. of consciousness.
PNES are fairly common, although rates vary widely across populations.
Prevalence is 5 in 100,000 to 30 in 100,000,13,14 but data should be interpreted ● No single sign can
differentiate between
with caution because most studies occur in tertiary epilepsy centers. In these
types of physiologic or
settings, patients with PNES may account for approximately 20% of the patients psychogenic nonepileptic
seen in an outpatient epilepsy center15 and at least one-third of the patients episodic events or between
evaluated in inpatient epilepsy monitoring units.16,17 nonepileptic episodic
events and epilepsy.
Epidemiologic studies of PNES have shown higher rates of this condition in
women.18 Other historical factors that have been associated with the diagnosis of ● Patients with
PNES include older age at onset, a report of sexual abuse, and a history of mild psychogenic nonepileptic
traumatic brain injury.11 A higher seizure frequency, a longer seizure duration, seizures may account for
and the lack of response to antiepileptic drugs (AEDs) are common reasons that 20% of those seen in an
outpatient setting and up to
physicians may question the diagnosis of epilepsy and also result in referral to one-third of patients in an
a tertiary care epilepsy center for further evaluation. The response to AEDs epilepsy monitoring unit.
should be interpreted with caution because the lack of or incomplete response
to AEDs may also be indicative of drug-resistant epilepsy, and an apparent ● Psychiatric disorders,
such as depression and
response to AEDs could represent a placebo response in patients with PNES.19
anxiety, are common in
Reports of stress, including both external and emotional events, can be patients with nonepileptic
triggers identified in a patient with nonepileptic episodic events including PNES, episodic events and
although these features can also be true of epileptic seizures. Stimulus-specific epileptic seizures, and their
presence should not be
factors that elicit events may suggest other neurologic problems, such as
used to discriminate
movement disorders, including paroxysmal kinesigenic dyskinesia. Comorbid between the two disorders
psychiatric disorders, medically unexplained symptoms, and a higher number of when making a diagnosis.
responses in the review of systems questionnaire have also been associated with a
greater association with PNES than epileptic seizures.20 Particular caution should
be advised to not assume that the presence of comorbid psychiatric disorders is
suggestive of a nonepileptic episodic event, however, because psychiatric
problems are also quite prevalent in patients with epileptic seizures.21
Clinical features suggestive of nonepileptic episodic events, including PNES,
can be useful in making a diagnosis and have been well described. The clinical
signs and symptoms may be quite variable, and despite the common teaching
that patients with PNES have nonstereotyped behaviors, this is not always the
case. There are many clinical signs that have been associated with PNES, and they
include waxing and waning movements or fluctuating course, a long duration of

events, eye closure, ictal crying, gradual onset, asynchronous movements, pelvic
thrusting, recall during the period of apparent unresponsiveness, and
hyperventilation.22,23 If a tongue bite occurs, it is more typical to be on the tip or
middle of the tongue in a person with PNES compared with the side of the tongue
in someone with epileptic seizures.24 Peri-ictal headache can also be seen with
both epileptic seizures as well as PNES but tends to be common in patients with
PNES.25 Preictal pseudosleep, described as a period of apparent sleep with eyes
closed and lack of motion prior to seizures, can also be seen in patients with
PNES.26 It is important to note that none of these signs is diagnostic of PNES or
epileptic seizures, and clinical assessment of symptoms should be an analysis of
the constellation of symptoms rather than interpretation of isolated findings.27
Although the gold standard for diagnosis remains video-EEG monitoring,
and this is a useful way to visualize the clinical events, not all patients have easy
or immediate access to this testing. Thus, each clinical factor should be
interpreted within the context of a thorough history and appropriate testing.
Clinical features should be compared and contrasted with those seen in patients
with epileptic seizures (TABLE 10-1).28
Our understanding of the etiology and psychopathology of PNES has
developed significantly over the past 15 to 20 years, although much work needs to
be done in this area to further our understanding of these problems. Although
traditionally classified under a single diagnosis of conversion, somatoform, or
TABLE 10-1 Clinical Characteristics of Epileptic Seizures Versus Psychogenic

Nonepileptic Seizuresa
Clinical Feature Epileptic Characteristics Psychogenic Characteristics
Age of onset All ages and common in children, onset in All ages, 15–25 most common
twenties and after age 50
Sex No clear sex difference Female more common, 3:1 ratio
Psychiatric history Occasionally present Commonly noted
Motor Generalized convulsions, bilateral Flailing, thrashing, and asynchronous movements more
movements are often synchronous common, side-to-side movements, head thrusts, pelvic
thrusting
Vocalization Ictal cry (vocalization) at onset for Weeping or screaming more common
generalized convulsions
Incontinence Frequent with some seizure types Occasional/rare
Duration <2–3 minutes Often prolonged, >3 minutes, waxing and waning
Injury Common; tongue biting with generalized Uncommon, tongue bite usually midline
convulsions (lateral aspect of tongue)
Amnesia Common Variable, sometimes conscious during seizure, rapid return

to consciousness
Suggestion Uncommon/no Often

provokes seizure
a
Modified with permission from Krumholz A, Hopp J, Semin Neurol.28 © 2006 Thieme Medical Publishers, Inc.
496 APRIL 2019

dissociative disorder, PNES is thought to have a multifactorial, and sometimes KEY POINTS
complex, etiology that may incorporate biological, psychological, and social
● Common clinical signs in
factors.29,30 This may suggest that the cause varies more than previously thought patients with psychogenic
within this patient population and emphasizes the need for a comprehensive nonepileptic seizures
psychosocial assessment. include waxing and waning
movements or a fluctuating
course, a long duration of
TOOLS TO CONFIRM THE DIAGNOSIS
events, eye closure, ictal
Although a careful history and neurologic examination may suggest the diagnosis crying, gradual onset,
of nonepileptic episodic events, it is common that additional testing may be asynchronous movements,
needed to confirm the diagnosis. pelvic thrusting, recall
during the period of
apparent unresponsiveness,
Guidelines and hyperventilation.
The diagnosis of PNES and other nonepileptic episodic events can be challenging.
An efficient and thorough evaluation, as well as access to resources, can play a ● Although a psychogenic
key role in these patients. Most neurologists and epilepsy subspecialists recognize nonepileptic seizure was
traditionally considered a
that it is often possible to make a reasonably accurate diagnosis of PNES manifestation of a
without video-EEG monitoring in many patients; at a minimum, a detailed conversion, somatization, or
and complete history and clinical assessment is the first step. dissociative disorder, it is
The International League Against Epilepsy published a report in 2013 to now considered to have a
multifactorial etiology that
establish diagnostic guidelines based on a stepwise and comprehensive also comprises biological
process of clinical and historical assessment and testing.31 A task force was and social factors.
organized to outline the process in this patient population and establish levels
of certainty. They posit that, in many cases, it may be possible to make a ● The diagnosis of
nonepileptic episodic
diagnosis without inpatient video-EEG monitoring. The group reiterated the
events, including
importance of taking a detailed clinical history and identification of comorbid psychogenic nonepileptic
conditions as the first steps in making a “possible” diagnosis of PNES in seizures, should be a
conjunction with the lack of epileptiform activity on an interictal EEG. This stepwise process that
approach is consistent with current Diagnostic and Statistical Manual of Mental includes clinical and
historical assessment and
Disorders, Fifth Edition, criteria for conversion disorders, including PNES, video and EEG monitoring
because it is now a diagnosis of inclusion rather than exclusion.32 A “probable” capturing typical events
diagnosis can be made if further review of the clinical event is performed by for the patient.
video or in person and is considered suggestive of PNES when seen with a
normal interictal EEG. There is additional support of the conclusion when the
clinician is experienced in the evaluation of seizure disorders and concludes
an ictal EEG is normal during a typical event seen with video-EEG monitoring.
Finally, a diagnosis of PNES is considered “documented” if there is an expert
review of video-EEG monitoring that includes normal interictal EEG as well as
normal ictal EEG during the typical events with phenomenology suggestive
of PNES.31
Inpatient Video-Electroencephalographic Monitoring

Although an extensive history and clinical signs can be extremely useful in making
the diagnosis of nonepileptic episodic events, often video-EEG monitoring is
necessary for confirmation. This test is considered the gold standard for the
diagnosis of epileptic and nonepileptic events, and when performed in an
epilepsy monitoring unit, offers added safety and clinical testing over outpatient
testing.33 Inpatient video-EEG should capture typical events that are analyzed
by trained personnel. The historical and clinical reports from patients and
witnesses may not always be accurate, and direct visualization of the events
and in-person assessment may be necessary. The video component of

video-EEG monitoring is key, and it is possible that the observation of clinical

events alone can be used to predict which patients have nonepileptic episodic
events and specifically PNES.22 Video without EEG may be more useful with
events that have motor manifestations and when interpreted by trained
personnel.23 This may be important, particularly in areas where easy and
inexpensive access to inpatient video-EEG monitoring may be limited.
Ideally, a typical seizure is captured and recorded with simultaneous video
and EEG with concomitant ECG recording. In patients with generalized
convulsive epileptic seizures, there is typically an ictal correlate during the
seizure, although caution should be taken when diagnosing frontal lobe seizures,
which may not be associated with clear EEG changes. In focal seizures with
alteration of awareness, there are typically changes seen on ictal EEG. All
EEGs and ECGs should be examined carefully by physicians trained in the
interpretation of long-term monitoring, and evaluation for cardiac etiology
should be pursued as necessary (CASE 10-1).
Key in the differentiation of various types of nonepileptic episodic events is
the state of the patient. Nonepileptic episodic events of physiologic causes may
CASE 10-1 A 49-year-old man was referred for evaluation of seizures. He reported
the onset of generalized convulsions approximately 9 months before
referral for video-EEG monitoring. His wife described that each episode
consisted of a loud vocalization followed by stiffening and generalized
shaking. He felt nauseated and sometimes had a feeling of déjà vu before
the events but otherwise did not remember much before the episodes.
He would be confused briefly upon awakening. His episodes continued
despite treatment with three anticonvulsant medications with an average
of two seizures per month. His outpatient EEG and MRI were normal. He
had a history of aortic stenosis and sleep apnea but no known epilepsy
risk factors.
He was admitted for inpatient video-EEG monitoring with concomitant
ECG. He had typical events captured that were associated with cardiac
asystole and diffuse suppression and slowing on the EEG (VIDEO 10-1, links.
lww.com/CONT/A273). Cardiology was immediately consulted. The
patient was transferred to the cardiac intensive care unit, and a
pacemaker was implanted.
COMMENT It is critical that patients who have an onset of apparent epileptic seizures
with normal outpatient EEG and MRI, as well as lack of epilepsy risk factors,
be considered for referral for video-EEG monitoring to make an accurate
diagnosis. Video-EEG monitoring should include continuous ECG because
cardiac arrhythmias and asystole can lead to convulsive activity that can be
mistaken for epilepsy. Clues in this patient also included an atypical age of
onset of epilepsy and a lack of prolonged postictal confusion. Emergent
referral for appropriate treatment was key in this case, and the patient was
able to stop all anticonvulsant medications and receive appropriate
cardiac care.
498 APRIL 2019

occur in wakefulness and sleep, depending primarily on the specific diagnosis. KEY POINTS
The wake and sleep states may be more useful in differentiating the diagnosis
● Inpatient video-EEG
when considering nonepileptic episodic events, including differentiating PNES monitoring will demonstrate
from epileptic seizures. Patients with PNES virtually always have events that wakefulness during events
occur from wakefulness. The state is important to verify by EEG because many for patients with psychogenic
patients will report that the events arise from sleep, but careful review of the nonepileptic seizures
compared with some events
EEG typically shows that there is a period of wakefulness before the onset of
in patients with physiologic
PNES events. nonepileptic episodic events
The length of video-EEG monitoring typically depends on the frequency of or epileptic seizures.
events, as well as the indication for the study. If there is suspicion for both
nonepileptic episodic events and epileptic seizures, then there should be an ● Prolactin levels may help
distinguish epileptic
attempt to capture all typical seizure types for diagnosis. Adequate clinical seizures from psychogenic
assessment, including testing of patients during seizures, is important to making nonepileptic seizures by
an accurate diagnosis.34,35 Detailed guidelines on the technologic criteria and demonstrating a twofold
clinical protocols are available through the National Association of Epilepsy rise from baseline in the 10 to
20 minutes after a seizure
Centers.33 It is very important to remember that, although video-EEG monitoring but should be interpreted
is considered important for the diagnosis of nonepileptic episodic events and, with caution.
in particular, PNES, this test should be considered a part of a process of making
an accurate diagnosis and has limitations despite high sensitivity.
Outpatient/Ambulatory Video-Electroencephalogram Monitoring

Outpatient or ambulatory video-EEG monitoring may be used in lieu of or in
addition to inpatient monitoring. This test can be useful when events are
frequent and are more likely to be captured during a shorter period of
monitoring. Limitations may include the lack of ability to ensure that patients are
on camera, to fix electrodes and troubleshoot poor EEG recordings, and to
perform clinical testing of patients during the study. In addition, most epilepsy
specialists do not advise withdrawal or tapering of anticonvulsant medications
during an outpatient video-EEG study for safety reasons.
Prolactin Levels and Other Laboratory Testing

Although less widely used now than in the past, serum prolactin levels have some
utility in the diagnosis of PNES. The increase in prolactin is more typical after
convulsive epileptic seizures and may help distinguish nonepileptic episodic
events that are PNES versus epileptic seizures. The maximal increase in prolactin
typically occurs in the first 10 to 20 minutes after the seizure and is considered
significant if the increase is at least twice the baseline level.36 An increase in
prolactin would not be expected in most physiologic nonepileptic episodic events
or in PNES. Prolactin testing will not help distinguish certain types of physiologic
nonepileptic episodic events, such as syncope.36 False positives can occur in
patients taking dopamine antagonists or with breast stimulation. There may also
be false negatives with frontal lobe seizures, with status epilepticus due to the
release of prolactin at the beginning of the event, or in patients taking dopamine
agonists because dopamine regulates prolactin secretion.36
Other laboratory markers have been studied to determine their utility in
diagnosing and distinguishing nonepileptic episodic events compared with
PNES from epileptic seizures. White blood cell counts, neurotrophic factors,
cortisol, and creatine kinase are among those that have shown limited utility
but are not used on a routine basis to make a diagnosis in an individual
patient.37,38

Neuropsychological Testing
Neuropsychological testing should be used adjunctively with clinical assessment
and video-EEG monitoring to assist in the diagnosis and to guide future testing.
It should not be used in lieu of video-EEG monitoring to make a psychological
diagnosis of PNES. In addition, the findings should not be reliably used to
differentiate nonepileptic episodic events from epileptic seizures because higher
levels of depression are seen in both patient groups.39 Ideally, this testing is
performed by a mental health professional with experience in the assessment
and treatment of patients with psychogenic disorders. It is often more helpful
when performed after video-EEG monitoring because the results can be
integrated into a clinical and diagnostic context. It may then help to target
treatment more directly to the patient. When personality inventories are used
within the context of neuropsychological testing, they typically show that
patients with PNES endorse conversion, somatic, dissociative, anxious, and
depressive symptoms.40
Neuroimaging
Neuroimaging studies are typically performed as part of the initial assessment of
patients with nonepileptic episodic events because this is a standard component
of the evaluation of seizures. Abnormalities seen on imaging should be
interpreted with caution because they do not necessarily differentiate between
nonepileptic episodic events and epileptic seizures or between types of
nonepileptic episodic events.
More recent work has focused on the search for a surrogate marker for PNES
through brain imaging. Initial data may suggest that there are abnormalities
within the brain connectivity of regions associated with motor activity, emotional
processing, and executive functions in patients with PNES.41 Findings remain
preliminary and suggest heterogeneity in this patient group as well as the need for
future work with rigorous physiologic measures.
TREATMENT
The management and treatment of patients with nonepileptic episodic events
depend on the specific diagnosis. The treatment of nonepileptic episodic events
of physiologic causes will not be discussed here because they should be directed
to the specific diagnosis.
Nonepileptic episodic events of a psychogenic etiology (ie, PNES) remain a
challenging disorder to treat, and delays to diagnosis may contribute to this
problem.42 Many patients undergo evaluation and treatment for presumed
epilepsy or other neurologic disorders for many years. They often are seen by
physicians and health care providers in many disciplines and experience disability
and a decreased quality of life.43 They also, unfortunately, may continue to
experience disability despite appropriate treatment. Despite these challenges,
there has been significant progress in the study of treatment strategies for PNES
in recent years.
The first consideration is the way the diagnosis of PNES is presented to the
patient, family, and caregivers. An honest, positive, and encouraging approach is
important. Many experts suggest that the physician should emphasize that the
diagnosis means that the patient does not have epilepsy, that the disorder is
“real” and should be taken seriously, and that there will be a comprehensive and
unified approach to treatment.44 Unless the patient has concomitant epileptic
500 APRIL 2019

seizures, the physician should emphasize that this problem does not require KEY POINTS
treatment with anticonvulsant medications and that there is a potential for
● The diagnosis of
improvement with the appropriate, targeted treatment strategies. The use of psychogenic nonepileptic
clear and consistent terminology is important in the initiation of treatment. seizures should be
This is helpful for patient education and understanding, as well as to avoid presented to the patient in
negative connotations associated with the disorder. Many suggest the use an honest, positive manner
with an outline of plans
of the term seizure, as opposed to attack or event, to provide clarity and to
for further evaluation
avoid pejorative terminology in patients with PNES who may have had and treatment.
attacks of an emotional or sexual nature that preceded the diagnosis of this
condition.45 ● Barriers to the treatment
There are many significant obstacles to the treatment of PNES. In addition of psychogenic nonepileptic
seizures include a lack of
to the delays to referral for appropriate evaluation and testing, there are many consistent follow-up with
issues that occur after the diagnosis is made. Supportive measures should be neurologists, shortage of
initiated after diagnosis. Some patients may not readily accept the diagnosis trained treatment providers,
and may seek other opinions. The patient should be encouraged to return for and stigma related to
psychological and
evaluation and care as desired, and regular follow-up may be beneficial in the psychiatric care.
long term.46 There may be problems with adherence to treatment. Although
many patients will attend the first session of treatment, there is a significant
rate of recidivism regardless of the patient’s apparent level of acceptance of
the disorder.47
There is also often a lack of a defined treatment provider, and in many
instances, there is a shortage of providers trained to deal with these
conditions. Although neurologists and, specifically, epileptologists tend to
make the diagnosis of PNES, patients often do not return for follow-up with
the diagnosing physician and may then not have referrals for appropriate
targeted treatment. At times, follow-up appointments with the neurologist
may not be offered, and this may lead to feelings of abandonment and
disengagement with treatment. Other barriers include the lack of treatment
providers with expertise in the psychological treatment of PNES and lack of
resources to pay for these services. There are, however, many good resources
available both for providers and for patients.48,49 In many countries, the
stigma associated with psychological care may prohibit many from getting
much-needed treatment.50
Psychotherapy
If evaluation has suggested the presence of a mental health disorder, then a
referral should be made to the appropriate provider. Particular attention should
be given to a rapid referral to address urgent psychiatric issues, such as suicidal
ideation, and a subsequent focus may also be centered on previous stressors
that may be identified through evaluation and therapy. There are several types
of psychotherapy that may be useful for the management of this disorder.
In addition to targeted therapy to address previous trauma or abuse,
psychodynamic interpersonal psychotherapy and group therapy and education
may be useful.51,52
Cognitive-Behavioral Therapy
Cognitive-behavioral therapy (CBT), group and family therapy, and other forms
of rehabilitation may all play a role in the management of PNES. CBT can be
administered by trained personnel using defined protocols. The basis of this
treatment is that patients learn to increase awareness of their dysfunctional

thoughts and learn to develop new behavioral responses. There are reports of
significant reductions in seizure frequency shown in several controlled trials of
CBT in this patient population.53 An initial pilot study showed a reduction in
seizure frequency when patients were randomly assigned to CBT versus
standard medical care, although the effect did not persist at 3-month follow-up.54
A subsequent multicenter randomized trial demonstrated a significant seizure
reduction of 51.4% in patients receiving CBT and 59.3% seizure reduction in
patients receiving CBT with sertraline (59%). There were also notable
improvements in quality of life, global functioning, and mood in the CBT-only
CASE 10-2 A 32-year-old woman presented with daily seizures that consisted of an
initial report from her family that her “eyes glazed over,” her muscles
would “droop,” and her voice changed during events. She typically would
then have trouble lifting her arms and described that she would “go
limp.” The length of the event could vary from minutes to hours and
occurred approximately 4 times per week. She felt she could stop the
events by taking medication. A second, less common event involved
spontaneous shaking of her left leg.
She was referred for inpatient video-EEG monitoring with a 5-year
history of these events after taking five different antiepileptic drugs. She
had inpatient video-EEG monitoring that captured her typical event, and
the EEG was normal during the event as well as throughout her 3-day
study. The findings on video-EEG monitoring, in conjunction with
neuropsychological testing, led her neurologist to make a diagnosis of
psychogenic nonepileptic seizures. The diagnosis was discussed in a
positive manner, and her neurologist emphasized that she did not have
epilepsy and that she would not need to take anticonvulsant medications.
She was not initially accepting of the diagnosis but agreed to follow-up
with the neurologist as an outpatient. She had been encouraged to attend
cognitive-behavioral therapy to manage the events.
Despite follow-up and the initiation of cognitive-behavioral therapy,
she continues to have seizures on a weekly basis, although they have
reduced in frequency by approximately 25%.
COMMENT It is critical that patients are referred for early evaluation and diagnosis
of seizures of an unknown etiology that are refractory to treatment.
Although an accurate diagnosis can allow the treating physician to stop
unnecessary medications and this may reduce morbidity, it is common
for patients to have continued seizures even after the diagnosis of
psychogenic nonepileptic seizures is made. It is important to note that,
despite appropriate testing and a positive approach to the diagnosis,
many patients will continue to have seizures, and continued efforts
should be made to have regular follow-up and referral for therapy
targeted to the individual.
502 APRIL 2019

group.55 Further studies are underway to continue to provide evidence-based KEY POINTS
data for CBT.56
● Cognitive-behavioral
therapy is an evidence-
OUTCOMES
based treatment shown to
There are several factors that may assist in determining outcomes in this reduce seizure frequency
patient population. Within such a broad group of disorders such as nonepileptic and improve the quality of
episodic events, the prognosis is heavily dependent on the type of nonepileptic life in patients with
psychogenic nonepileptic
episodic event. The outcome of nonepileptic episodic events of a physiologic
seizures.
origin will not be discussed here because the outcomes are dependent on the
diagnosis, clinical context, and treatments specific to the cause of the patient’s ● Although up to 70% of
disorder. Nonepileptic episodic events of a psychogenic origin (ie, PNES) can patients with psychogenic
be highly refractory to treatment with more than 70% of patients continuing to nonepileptic seizures may
continue to have seizures
have seizures after diagnosis and reporting high rates of disability (CASE 10-2). after diagnosis, higher
Factors that can be associated with a better prognosis for PNES include higher education, shorter time to
education, shorter time to diagnosis from symptom onset, and lower scores on diagnosis, and lower
assessments of somatoform and dissociative scales.57 somatoform and
dissociative scores may
predict better outcomes.
SPECIAL POPULATIONS
Special considerations should be kept in mind for patients with concomitant ● Up to 10% of patients with
PNES and epilepsy and for nonepileptic episodic events occurring in the pediatric psychogenic nonepileptic
seizures may also have
patient population.
epileptic seizures, and
careful attention is needed
Comorbid Psychogenic Nonepileptic Seizures and Epileptic Seizures during evaluation and
At least 10% of patients with PNES may also have epileptic seizures, so an treatment to identify and
manage both disorders.
accurate diagnosis can be challenging in this group.58 A particular challenge is
the definition of epilepsy in this population because the criteria for the ● Physiologic causes of
diagnosis of concomitant PNES and epileptic seizures may range from interictal nonepileptic episodic
abnormalities seen on an EEG with documented PNES to video-EEG confirmation events in children that may
mimic epilepsy include
of both epileptic seizures and PNES in one hospital admission.59 These variations
reflux, sleep disorders, and
may account for the wide range of reported epileptic seizures in patients with breath-holding spells.
PNES, and great care should be taken to make this complex diagnosis. Treatment
strategies need to be selected carefully and often integrated to address both
seizure types. It can be useful to share the video from video-EEG monitoring
with patients, families, and caregivers to assist in identifying and discriminating
epileptic seizures from PNES when reporting continued seizures. This can be
critical in further care because it is important to try to target AED treatment to
epileptic seizures and psychological treatment for recurrent PNES.
Pediatrics
The diagnosis of nonepileptic episodic events, including PNES, in children can be
particularly challenging. There are many stereotyped events in the pediatric
population that differ from adults and should be considered when a child
presents with paroxysmal events. Physiologic events that may mimic epileptic
seizures may include gastroesophageal reflux, night terrors, breath-holding
spells, or shuddering.60,61
PNES are well documented in the pediatric population, although the
epidemiology and characteristics of this disorder may be distinct from those seen
in adults and typically may involve apparent alteration of consciousness.62
There may be several factors that play a role in the development of this problem,
and school difficulties are often cited as precipitating factors.63

CONCLUSION
Nonepileptic episodic events remain a diagnostic and therapeutic challenge
in the field of neurology. A careful and thorough evaluation to determine whether
events have a physiologic or psychological cause is critical to guiding patient
education and treatment strategies. This begins with a careful history-taking
process, including a detailed assessment of the clinical presentation of the events,
the social history, and evaluation of psychological function. Further study with
video-EEG is often required to distinguish these types of events and should be
undertaken early in the disorder to avoid additional disability and to target
treatment initiatives. Once the diagnosis is made, the discussion should be
positive and thoughtful, and coordination between neurologists and mental
health experts is key in coordinating efforts for both diagnosis and treatment.
The management of PNES remains a challenge. A recent study1 suggests that
psychogenic disorders may be caused by an interaction of biological,
psychological, and social factors, and this new understanding may help improve
treatment and potentially individualize management in the future. Psychogenic
events may lead to continued disability, but psychological and behavioral
therapy may reduce seizures and improve the quality of life for these patients.
Coordinated international efforts have brought more attention to these common
problems and are likely to assist in targeting work to improve education and
expertise for physicians who diagnose and manage these disorders.
VIDEO LEGEND
VIDEO 10-1
Nonepileptic episodic event of cardiac
etiology. A 49-year-old man presented with a
history of generalized convulsions of sudden onset
preceded by nausea. In this captured episode, he is
initially asleep (confirmed by EEG [not shown] at the
onset of the event). He has an arousal on EEG (not
shown) at the time of asystole seen on ECG (not
shown), and after a brief pause, he has body
stiffening (tonic activity) followed by clonic activity.
His eyes are open during the event. Clinical testing
by staff in the epilepsy monitoring unit
demonstrates a relatively rapid return of
consciousness. He is able to show two fingers,
point to his nose, and raise his arm, as well as
answer questions correctly. After awakening, he
reported mild nausea preceding the event.
links.lww.com/CONT/A273
© 2019 American Academy of Neurology
REFERENCES
1 Voon V, Cavanna AE, Coburn K, et al. Functional 3 Reuber M, Fernández G, Bauer J, et al. Diagnostic
neuroanatomy and neurophysiology of functional delay in psychogenic nonepileptic seizures.
neurological disorders (conversion disorder). Neurology 2002;58(3):493–495. doi:10.1212/
J Neuropsychiatry Clin Neurosci 2016;28(3): WNL.58.3.493.
168–190. doi:10.1176/appi.neuropsych.14090217. 4 Gates JR, Mercer K. Nonepileptic events. Semin
2 Asadi-Pooya AA, Tinker J. Delay in diagnosis of Neurol 1995;15(2):167–174.
psychogenic nonepileptic seizures in adults: a 5 Robson C, Myers L, Pretorius C, et al. Health related
post hoc study. Epilepsy Behav 2017;75:143–145. quality of life of people with non-epileptic seizures:
doi:10.1016/j.yebeh.2017.08.005. the role of socio-demographic characteristics
and stigma. Seizure 2018;55:93–99. doi:10.1016/
j.seizure.2018.01.001.
504 APRIL 2019

6 Begley CE, Famulari M, Annegers JF, et al. The 21 Griffith N, Szaflarski J. Epidemiology and
cost of epilepsy in the united states: an estimate classification of psychogenic nonepileptic
from population-based clinical and survey seizures. In: Schachter SC, LaFrance WC Jr, editors.
data. Epilepsia 2000;41(3):342–351. doi:10.1111/ Gates and Rowan’s nonepileptic seizures. 3rd ed.
j.1528-1157.2000.tb00166.x. Cambridge, UK: Cambridge University Press,
2010:3–16.
7 Brigo F, Igwe SC, Ausserer H, et al. Terminology of
psychogenic nonepileptic seizures. Epilepsia 22 Syed TU, LaFrance WC Jr, Kahriman ES, et al. Can
2015;56(3):e21–e25. doi:10.1111/epi.12911. semiology predict psychogenic nonepileptic
seizures? A prospective study. Ann Neurol 2011;
8 Gates J. Diagnosis and treatment of non-epileptic
69(6):997–1004. doi:10.1002/ana.22345.
seizures. In: Shorvon S, Perucca E, Fish D, Dodson
E, editors. The treatment of epilepsy. Malden, 23 Erba G, Giussani G, Juersivich A, et al. The
MA: Blackwell Science, 2004:307–313. semiology of psychogenic nonepileptic seizures
revisited: can video alone predict the diagnosis?
9 Mellers JD. The approach to patients with “non-
Preliminary data from a prospective feasibility
epileptic seizures.” Postgrad Med J 2005;81(958):
study. Epilepsia 2016;57(5):777–785. doi:10.1111/
498–504. doi:10.1136/pgmj.2004.029785.
epi.13351.
10 Smith D, Defalla B, Chadwick DW. The misdiagnosis
24 Ettinger AB, Devinsky O, Weisbrot DM,
of epilepsy and the management of refractory
Ramakrishna RK, Goyal A. A comprehensive
epilepsy in a specialist clinic. QJM 1999;92(1):
profile of clinical, psychiatric, and psychosocial
15–23. doi:10.1093/qjmed/92.1.15.
characteristics of patients with psychogenic
11 Kerr WT, Janio EA, Braesch CT, et al. An objective nonepileptic seizures. Epilepsia 1999;40(9):
score to identify psychogenic seizures based on 1292–1298.
age of onset and history. Epilepsy Behav 2018;80:
25 Ebner A. Long-term EEG and video monitoring in
75–83. doi:10.1016/j.yebeh.2017.11.035.
the differential diagnosis of epilepsy [abstract].
12 De Paola L, Terra VC, Silvado CE, et al. Improving Epilepsia 1998;39(suppl 2):2.
first responders' psychogenic nonepileptic
26 Benbadis SR, Lancman ME, King LM, Swanson SJ.
seizures diagnosis accuracy: development and
Preictal pseudosleep a new finding in psychogenic
validation of a 6-item bedside diagnostic tool.
seizures. Neurology 1996;47(1):63–67.
j.yebeh.2015.10.025. 27 Gröppel G, Kapitany T, Baumgartner C. Cluster
analysis of clinical seizure semiology of
13 Kanemoto K, LaFrance WC, Duncan R, et al. PNES
psychogenic nonepileptic seizures. Epilepsia
around the world: where we are now and how we
2000;41(5):610–614.
can close the diagnosis and treatment gaps-an
ILAE PNES task force report. Epilepsia Open 2017; 28 Krumholz A, Hopp J. Psychogenic (nonepileptic)
2(3):307–316. doi:10.1002/epi4.12060. seizures. Semin Neurol 2006;26(3):341–350.
doi:10.1055/s-2006-945520.
14 Asadi-Pooya AA, Sperling MR. Epidemiology of
psychogenic nonepileptic seizures. Epilepsy 29 Perez DL, LaFrance WC Jr. Nonepileptic seizures:
Behav 2015;46:60–65. doi:10.1016/j.yebeh.2015. an updated review. CNS Spectr 2016;21(3):
03.015. 239–246. doi:10.1017/S109285291600002X.
15 Lesser RP. Psychogenic seizures. Neurology 1996; 30 Brown RJ, Reuber M. Psychological and
46(6):1499–1507. doi:10.1212/WNL.46.6.1499. psychiatric aspects of psychogenic non-
epileptic seizures (PNES): a systematic
16 Reuber M. Psychogenic nonepileptic seizures:
review. Clin Psychol Rev 2016;45:157–182.
answers and questions. Epilepsy Behav 2008;
doi:10.1016/j.cpr.2016.01.003.
12(4):622–635. doi:10.1016/j.yebeh.2007.11.006.
31 LaFrance WC Jr, Baker GA, Duncan R, et al.
17 Benbadis SR, O'Neill E, Tatum WO, Heriaud L.
Minimum requirements for the diagnosis of
Outcome of prolonged video-EEG monitoring at a
psychogenic nonepileptic seizures: a staged
typical referral epilepsy center. Epilepsia 2004;
approach: a report from the International
45(9):1150–1153. doi:10.1111/j.0013-9580.2004.14504.x.
League Against Epilepsy Nonepileptic Seizures
18 Thomas AA, Preston J, Scott RC, Bujarski KA. Task Force. Epilepsia 2013;54(11):2005–2018.
Diagnosis of probable psychogenic nonepileptic doi:10.1111/epi.12356.
seizures in the outpatient clinic: does gender
32 American Psychiatric Association. Diagnostic
matter? Epilepsy Behav 2013;29(2):295–297.
and statistical manual of mental disorders.
doi:10.1016/j.yebeh.2013.08.006.
5th edition. Washington, DC: American
19 Alessi R, Valente KD. Psychogenic nonepileptic Psychiatric Association, 2013.
seizures: should we use response to AEDS as a
33 Shih JJ, Fountain NB, Herman ST, et al. Indications and
red flag for the diagnosis? Seizure 2014;23(10):
methodology for video-electroencephalographic
906–908. doi:10.1016/j.seizure.2014.07.016.
studies in the epilepsy monitoring unit. Epilepsia
20 Asadi-Pooya AA, Rabiei AH, Tinker J, Tracy J. 2018;59(1):27–36. doi:10.1111/epi.13938.
Review of systems questionnaire helps
differentiate psychogenic nonepileptic seizures
from epilepsy. J Clin Neurosci 2016;34:105–107.
doi:10.1016/j.jocn.2016.05.037.

34 Beniczky S, Neufeld M, Diehl B, et al. Testing 46 Aboukasm A, Mahr G, Gahry BR, et al. Retrospective
patients during seizures: a European consensus analysis of the effects of psychotherapeutic
procedure developed by a joint taskforce of the interventions on outcomes of psychogenic
ILAE - Commission on European Affairs and the nonepileptic seizures. Epilepsia 1998;39(5):
European epilepsy monitoring unit association. 470–473. doi:10.1111/j.1528-1157.1998.tb01407.x.
Epilepsia 2016;57(9):1363–1368. doi:10.1111/
47 Tolchin B, Dworetzky BA, Baslet G. Long-term
epi.13472.
adherence with psychiatric treatment among
35 Chabolla DR, Shih JJ. Postictal behaviors patients with psychogenic nonepileptic seizures.
associated with psychogenic nonepileptic Epilepsia 2018;59(1):e18–e22. doi:10.1111/epi.13969.
seizures. Epilepsy Behav 2006;9(2):307–311.
48 LaFrance WC Jr, Wincze JP. Treating nonepileptic
doi:10.1016/j.yebeh.2006.06.009.
seizures: therapist guide. New York, NY: Oxford
36 Chen DK, So YT, Fisher RS, Therapeutics and University Press, 2015.
Technology Assessment Subcommittee of the
49 Reiter J, Andrews D, Reiter C, LaFrance WC Jr.
American Academy of Neurology. Use of serum
Taking control of your seizures: workbook. New
prolactin in diagnosing epileptic seizures: report
York, NY: Oxford University Press, 2015.
of the therapeutics and technology assessment
subcommittee of the American Academy of 50 Hingray C, El-Hage W, Duncan R, et al. Access
Neurology. Neurology 2005;65(5):668–675. to diagnostic and therapeutic facilities for
doi:10.1212/01.wnl.0000178391.96957.d0. psychogenic nonepileptic seizures: an
international survey by the ILAE PNES task force.
37 Bakvis P, Spinhoven P, Giltay EJ, et al. Basal
Epilepsia 2018;59(1):203–214. doi:10.1111/epi.13952.
hypercortisolism and trauma in patients with
psychogenic nonepileptic seizures. Epilepsia 51 Mayor R, Howlett S, Grünewald R, Reuber M.
2010;51(5):752–759. doi:10.1111/j.1528-1167. Long-term outcome of brief augmented
2009.02394.x. psychodynamic interpersonal therapy for
psychogenic nonepileptic seizures: seizure
38 LaFrance WC Jr, Leaver K, Stopa EG, et al.
control and health care utilization. Epilepsia 2010;
Decreased serum BDNF levels in patients with
51(7):1169–1176. doi:10.1111/j.1528-1167.2010.02656.x.
epileptic and psychogenic nonepileptic seizures.
Neurology 2010;75(14):1285–1291. doi:10.1212/WNL. 52 Duncan R, Anderson J, Cullen B, Meldrum S.
0b013e3181f612bb. Predictors of 6-month and 3-year outcomes
after psychological intervention for psychogenic
39 Asmussen SB, Kirlin KA, Gale SD, Chung SS.
non epileptic seizures. Seizure 2016;36:22–26.
Differences in self-reported depressive
doi:10.1016/j.seizure.2015.12.016.
symptoms between patients with epileptic and
psychogenic nonepileptic seizures. Seizure 53 Carlson P, Nicholson Perry K. Psychological
2009;18(8):564–566. doi:10.1016/j.seizure. interventions for psychogenic non-epileptic
2009.05.006. seizures: a meta-analysis. Seizure 2017;45:
142–150. doi:10.1016/j.seizure.2016.12.007.
40 Thompson AW, Hantke N, Phatak V, Chaytor N.
The personality assessment inventory as a tool 54 Goldstein LH, Chalder T, Chigwedere C, et al.
for diagnosing psychogenic nonepileptic seizures. Cognitive-behavioral therapy for psychogenic
Epilepsia 2010;51(1):161–164. doi:10.1111/j.1528-1167. nonepileptic seizures: a pilot RCT. Neurology
2009.02151.x. 2010;74(24):1986–1994. doi:10.1212/
WNL.0b013e3181e39658.
41 Mcsweeney M, Reuber M, Levita L. Neuroimaging
studies in patients with psychogenic non- 55 LaFrance WC Jr, Baird GL, Barry JJ, et al.
epileptic seizures: a systematic meta-review. Multicenter pilot treatment trial for psychogenic
Neuroimage Clin 2017;16:210–221. doi:10.1016/ nonepileptic seizures: a randomized clinical
j.nicl.2017.07.025. trial. JAMA Psychiatry 2014;71(9):997–1005.
doi:10.1001/jamapsychiatry.2014.817.
42 LaFrance WC Jr, Benbadis SR. Avoiding the costs
of unrecognized psychological nonepileptic 56 Goldstein LH, Mellers JD, Landau S, et al.
seizures. Neurology 2006;66(11):1620–1621. Cognitive behavioural therapy vs standardised
doi:10.1212/01.wnl.0000224953.94807.be. medical care for adults with dissociative
non-epileptic seizures (CODES): a multicentre
43 Szaflarski JP, Szaflarski M, Hughes C, et al.
randomised controlled trial protocol. BMC Neurol
Psychopathology and quality of life: Psychogenic
2015;15:98. doi:10.1186/s12883-015-0350-0.
non-epileptic seizures versus epilepsy. Med Sci
Monit 2003;9(4):CR113–8. 57 Reuber M, Pukrop R, Bauer J, et al. Outcome in
psychogenic nonepileptic seizures: 1 to 10-year
44 Friedman JH, LaFrance WC Jr. Psychogenic
follow-up in 164 patients. Ann Neurol 2003;53(3):
disorders: the need to speak plainly.
305–311. doi:10.1002/ana.3000.
Arch Neurol 2010;67(6):753–755. doi:10.1001/
archneurol.2010.91. 58 Benbadis S, Agrawal V, Tatum WO 4th. How
many patients with psychogenic nonepileptic
45 LaFrance WC Jr. Psychogenic nonepileptic
seizures also have epilepsy? Neurology 2001;
"seizures" or "attacks"? It's not just semantics:
57(5):915–917.
seizures. Neurology 2010;75(1):87–88.
doi:10.1212/WNL.0b013e3181e62181.
506 APRIL 2019

59 Reuber M, Fernández G, Bauer J, et al. Interictal 62 Szabó L, Siegler Z, Zubek L, et al. A detailed
EEG abnormalities in patients with psychogenic semiologic analysis of childhood psychogenic
nonepileptic seizures. Epilepsia 2002;43(9): nonepileptic seizures. Epilepsia 2012;53(3):
1013–1020. doi:10.1046/j.1528-1157.2002.52301.x. 565–570. doi:10.1111/j.1528-1167.2012.03404.x.
60 Kotagal P, Costa M, Wyllie E, Wolgamuth B. 63 Reilly C, Menlove L, Fenton V, Das KB. Psychogenic
Paroxysmal nonepileptic events in children and nonepileptic seizures in children: a review.
adolescents. Pediatrics 2002;110(4):e46. Epilepsia 2013;54(10):1715–1724. doi:10.1111/
epi.12336.
61 Sawchuk T, Buchhalter J. Psychogenic
nonepileptic seizures in children - psychological
presentation, treatment, and short-term outcomes.
Epilepsy Behav 2015;52(pt A):49–56. doi:10.1016/
j.yebeh.2015.08.032.

REVIEW ARTICLE

Update on Antiepileptic
ONLINE
Drugs 2019
By Bassel W. Abou-Khalil, MD, FAAN
EDITOR’S NOTE
The article “Update on Antiepileptic Drugs 2019” by Dr Abou-Khalil
was first published in the February 2016 Epilepsy issue of Continuum:
Lifelong Learning in Neurology as ”Antiepileptic Drugs“ and has
been updated by Dr Abou-Khalil for this issue.
CITE AS: ABSTRACT

PURPOSE OF REVIEW: This article is an update from the article on antiepileptic
2019;25(2, EPILEPSY):508–536.
drug (AED) therapy published in the last Continuum issue on epilepsy and
Address correspondence to is intended to cover the vast majority of agents currently available to the
Dr Bassel W. Abou-Khalil, neurologist in the management of patients with epilepsy. Treatment of
Vanderbilt University, A-0118
Medical Center North, epilepsy starts with AED monotherapy. Knowledge of the spectrum of
Neurology Department, efficacy, clinical pharmacology, and modes of use for individual AEDs is
Nashville, TN 37232, Bassel.
essential for optimal treatment for epilepsy. This article addresses AEDs
abou-khalil@vanderbilt.edu.
individually, focusing on key pharmacokinetic characteristics, indications,
RELATIONSHIP DISCLOSURE: and modes of use.
Dr Abou-Khalil has served on the
editorial board of Clinical
Neurophysiology and has RECENT FINDINGS: Since the previous version of this article was published,
received research/grant support three new AEDs, brivaracetam, cannabidiol, and stiripentol, have been
from Biogen, the National
Institute of Neurological approved by the US Food and Drug Administration (FDA), and ezogabine
Disorders and Stroke, SK-Pharma, was removed from the market because of decreased use as a result of
Sunovion Pharmaceuticals Inc, bluish skin pigmentation and concern over potential retinal toxicity.
and UCB SA.
Older AEDs are effective but have tolerability and pharmacokinetic
UNLABELED USE OF disadvantages. Several newer AEDs have undergone comparative trials
USE DISCLOSURE:
demonstrating efficacy equal to and tolerability at least equal to or better
Dr Abou-Khalil discusses the than older AEDs as first-line therapy. The list includes lamotrigine,
unlabeled/investigational use oxcarbazepine, levetiracetam, topiramate, zonisamide, and lacosamide.
of primidone for the treatment
of essential tremor, valproate
Pregabalin was found to be less effective than lamotrigine. Lacosamide,
for the treatment of pregabalin, and eslicarbazepine have undergone successful trials of
generalized myoclonic and conversion to monotherapy. Other newer AEDs with a variety of mechanisms
generalized tonic-clonic
seizures, gabapentin for the of action are suitable for adjunctive therapy. Most recently, the FDA adopted
treatment of headache and a policy that a drug’s efficacy as adjunctive therapy in adults can be
sleep disorders, lamotrigine
extrapolated to efficacy in monotherapy. In addition, efficacy in adults can
as a first-line treatment for
epilepsy, zonisamide as initial be extrapolated for efficacy in children 4 years of age and older. Both
monotherapy for epilepsy, extrapolations require data demonstrating that an AED has equivalent
and cannabidiol and clobazam
for the treatment of focal
pharmacokinetics between its original approved use and its extrapolated
seizures. use. In addition, the safety of the drug in pediatric patients has to be
demonstrated in clinical studies that can be open label. Rational AED
© 2019 American Academy combinations should avoid AEDs with unfavorable pharmacokinetic interactions
of Neurology. or pharmacodynamic interactions related to mechanism of action.
508 APRIL 2019

SUMMARY: Knowledge of AED pharmacokinetics, efficacy, and tolerability KEY POINTS
profiles facilitates the choice of appropriate AED therapy for patients with
● Phenobarbital,
epilepsy. primidone, phenytoin, and
carbamazepine are potent
inducers of liver enzymes,
INTRODUCTION reducing the efficacy of
A
drugs metabolized by
ntiepileptic drugs (AEDs) are the mainstay of epilepsy therapy.
the cytochrome P450
Until 1993, the choice of AED was limited to seven or eight major enzyme system.
agents. However, more than 17 new AEDs have been approved and
marketed since then. With such a large choice of AEDs, much ● Long-term phenobarbital
guidance is needed in the choice of AEDs for initial therapy, later use is associated with
decreased bone density,
replacement monotherapy, or adjunctive therapy. Considerations in AED choice Dupuytren contractures,
must include the spectrum of efficacy of the AED (TABLE 11-1), its pharmacokinetic plantar fibromatosis, and
properties (TABLE 11-2), its safety and tolerability profile, and its efficacy against frozen shoulder.
comorbidities, as relevant to the patient’s specific circumstances. This article
addresses each AED, focusing on indications, tolerability, and clinical use.
Relevant pharmacokinetic properties are also discussed. This article focuses on
AED use in adults; however, salient features related to use in children are
highlighted throughout the text. TABLE 11-3 summarizes AED dosing in children,
and TABLE 11-4 summarizes teratogenicity data for AEDs.1,2 The order in which
AEDs are presented is roughly based on the order in which AEDs were marketed,
although related AEDs will be discussed together with their oldest relative.
PHENOBARBITAL
Phenobarbital has been in clinical use since 1912, although initially used as a
sedative and sleep aid. Its main mechanism of action is through binding the
γ-aminobutyric acid (GABA)-A receptor, prolonging the opening of the associated
chloride channel. It is available as an oral preparation as well as a parenteral
solution. It has excellent oral bioavailability and relatively low protein binding. It is
mostly metabolized in the liver, but approximately one-quarter of the dose is
eliminated unchanged in the urine. It has a long half-life of approximately 80 to
100 hours. Phenobarbital is a potent hepatic P450 enzyme inducer, accelerating
the metabolism of medications processed by this enzyme system and reducing
their plasma concentration. This affects its use in combination therapy because
it may render concomitant AEDs less effective if they are metabolized by
the liver.
Phenobarbital is effective against focal seizures and generalized tonic-clonic
seizures but is not effective against generalized absence seizures. The parenteral
solution has been used effectively for status epilepticus.
The suggested maintenance dose is 1 mg/kg/d to 2.5 mg/kg/d,3 but a much
lower starting dose is recommended, such as 30 mg to 60 mg at bedtime. The
dose can be increased by 30 mg to 60 mg every 2 weeks as needed, depending on
seizure control and tolerability. A once-daily dose at bedtime may reduce
sedation and is adequate because of its long half-life. The recommended serum
concentration is 15 mg/L to 40 mg/L.
Phenobarbital’s main adverse effects are sedation, decreased concentration,
and mood changes, particularly depression. In children, it can cause
hyperactivity. Long-term use is associated with decreased bone density,
Dupuytren contractures, plantar fibromatosis, and frozen shoulder. It is not
recommended in pregnancy because of teratogenicity with increased risk of

ANTIEPILEPTIC DRUGS
cardiac malformations in the exposed fetus. Evidence also exists of decreased

cognitive abilities in males exposed in utero. Phenobarbital is a controlled
substance.
Place in Therapy
Because of its adverse effect on cognitive function and its enzyme induction,
phenobarbital is used very infrequently as first-line therapy in developed
countries. However, its low cost and wide availability make it the only affordable
AED in much of the developing world. In addition, there has been some debate
about adverse cognitive effects; one study in rural China reported no major
negative cognitive effects, and some cognitive gains, likely related to improved
seizure control.4
TABLE 11-1 Spectrum of Efficacy of Select Antiepileptic Drugs
Lennox-Gastaut
Generalized Generalized Generalized Syndrome/Infantile
Antiepileptic Tonic-Clonic Absence Myoclonic Spasms/Dravet
Drug Focal Seizures Seizures Seizures Seizures Syndromea
Phenobarbital Class I trials Suggested, but Not effective Class IV evidence

not proven in
Class I trials
Phenytoin Class I trials Suggested, but Not effective Not effective

not proven in
Class I trials
Carbamazepine Class I trials Suggested, but Not effective Not effective

not proven in
Class I trials
Oxcarbazepine Class I trials Unknown Not effective Not effective
Eslicarbazepine Class I trials Unknown Not effective Not effective

acetate
Valproate Class I trials Suggested, but Class I trials Suggested, but Suggested, but
not proven in not proven in not proven in
Class I trials Class I trials Class I trials
Ethosuximide Not effective Not effective Class I trials Not effective
Clobazam Suggested, but Suggested, but Suggested, but Suggested, but not Class I trials
not proven in not proven in not proven in proven in Class I Lennox-Gastaut
Class I trials Class I trials Class I trials trials syndrome
Felbamate Class I trials Suggested, but Unknown Unknown Class I trials

not proven in Lennox-Gastaut
Class I trials syndrome
Gabapentin Class I trials Not effective Not effective Not effective
Pregabalin Class I trials Not effective Not effective Not effective
CONTINUED ON PAGE 511
510 APRIL 2019

PRIMIDONE
Primidone is converted in the liver to phenobarbital and phenylethylmalonamide,
which is also an active metabolite. It is available only as an oral preparation.
When used in monotherapy, about 25% of oral primidone is converted to
phenobarbital. The half-life of primidone is 10 to 15 hours in monotherapy and
6.5 to 8.3 hours with enzyme inducers. Primidone is a potent enzyme inducer.
Primidone is effective against focal seizures and generalized tonic-clonic
seizures. Anecdotal evidence also exists of efficacy against myoclonic seizures.
Primidone is also effective in controlling essential tremor.
In addition to sedation and other adverse effects of phenobarbital, primidone
use is associated with an acute toxic reaction unrelated to phenobarbital, with
potentially debilitating drowsiness, dizziness, ataxia, nausea, and vomiting.
CONTINUED FROM PAGE 510
Lennox-Gastaut
Generalized Generalized Generalized Syndrome/Infantile
Antiepileptic Tonic-Clonic Absence Myoclonic Spasms/Dravet
Drug Focal Seizures Seizures Seizures Seizures Syndromea
Lamotrigine Class I trials Class I trials Suggested, but Variable Class I trials Lennox-
not proven in Gastaut syndrome
Class I trials
Topiramate Class I trials Class I trials Not effective in Unknown Class I trials Lennox-
one Class I trial Gastaut syndrome
Tiagabine Class I trials Not effective Not effective Not effective
Levetiracetam Class I trials Class I trials Suggested, but Class I trials

not proven in
Class I trials
Brivaracetam Class I trials Unknown Unknown Unknown
Zonisamide Class I trials Suggested, but Suggested, but Suggested, but not
not proven in not proven in proven in Class I
Class I trials Class I trials trials
Lacosamide Class I trials Unknown Not effective Not effective
Vigabatrin Class I trials Not effective Not effective Not effective Class I trials infantile
spasms
Rufinamide Class I trials, but Suggested, but Unknown Unknown Class I trials Lennox-
not FDA approved not proven in Gastaut syndrome
Class I trials
Perampanel Class I trials Class I trials Unknown Class IV evidence
Cannabidiol Class IV evidence Unknown Unknown Unknown Class I trials in

Lennox-Gastaut
syndrome and
Dravet syndrome
FDA = US Food and Drug Administration.

a
Blank cells in this column represent no convincing or Class I data.

ANTIEPILEPTIC DRUGS
TABLE 11-2 Select Pharmacokinetic Parameters of Antiepileptic Drugs
Potential for
Oral Protein Pharmacokinetic
Antiepileptic Drug Bioavailability Bindinga Metabolism Half-lifeb Interactions
Phenobarbital Good Low >70% Long High
Phenytoin Variable High Extensive, Intermediate (long High

nonlinear with toxicity)
Carbamazepine Good Intermediate Extensive Intermediate High
Oxcarbazepine Good Low Extensive Short Moderate
Eslicarbazepine Good Low ~40% Intermediate Moderate

acetate
Valproate Good High Extensive Intermediate High
Ethosuximide Good Low Extensive Long Moderate
Clobazam Good High Extensive Intermediate High
Felbamate Good Low ~50% Intermediate High
Gabapentin Low Low None Short No/minimal
Pregabalin Good Low None Short No/minimal
Lamotrigine Good Intermediate Extensive Intermediate Moderate
Topiramate Good Low ~30% Intermediate No/minimal
Tiagabine Good High Extensive Short High
Levetiracetam Good Low ~30% nonhepatic Short No/minimal
Brivaracetam Good Low Extensive Short Moderate
Zonisamide Good Low ~65% Long Moderate
Lacosamide Good Low ~60% Intermediate No/minimal
Vigabatrin Good Low None Intermediate No/minimal
Rufinamide Good Intermediate Extensive Short Moderate
Perampanel Good High Extensive Long Moderate
Cannabidiol Low High Extensive Long High
a
Low: <50%; intermediate: 50% to 85%; high: >85%.
b
Short: <10 hours; intermediate: 10 to 30 hours; long: >30 hours.
512 APRIL 2019

Place in Therapy KEY POINTS
Primidone was the least-tolerated agent in the large cooperative US Department
● In addition to sedation
of Veterans Affairs trial comparing the efficacy and tolerability of and other adverse effects of
carbamazepine, phenobarbital, phenytoin, and primidone.5 As a result, it is phenobarbital, primidone
infrequently used. In view of the acute toxic adverse effects, primidone should be use is associated with an
started at a low dose, for example 50 mg to 125 mg at bedtime, then increased acute toxic reaction
unrelated to phenobarbital,
gradually by 50 mg to 125 mg every 3 to 7 days to 250 mg 3 times a day.
with potentially debilitating
drowsiness, dizziness,
PHENYTOIN ataxia, nausea, and vomiting.
Phenytoin has been in clinical use since 1938 when its efficacy in the maximum
electroshock animal model was discovered. Phenytoin binds to the active state of ● Phenytoin has saturable
nonlinear kinetics. Beyond a
the sodium channel to prolong its fast inactivated state, thus reducing high-frequency certain serum
firing as might occur during a seizure, while allowing normal action potentials to concentration, usually
occur. It is available as an oral preparation and a parenteral solution, and a within the accepted
phenytoin prodrug, fosphenytoin, is available for IV and IM administration. therapeutic range,
phenytoin concentration
Phenytoin bioavailability is reduced with coadministration of calcium, increases disproportionately
antacids, and nasogastric feedings. It is highly protein bound at approximately with an increase in the dose.
90%. It is metabolized in the liver, mostly by cytochrome P450 (CYP) 2C9 and, to Small increments are
a lesser extent, CYP 2C19. Phenytoin’s metabolism is saturable, resulting in necessary when increasing
the dose at a serum
nonlinear kinetics. As the serum concentration increases, it reaches a point concentration in the
within the recommended therapeutic range after which the half-life starts therapeutic range.
increasing. Beyond that point, the phenytoin plasma level increases
disproportionately with an increase in the dose (FIGURE 11-1). ● The traditional sodium
channel blockers phenytoin,
Phenytoin is a potent enzyme inducer that reduces the efficacy of drugs
carbamazepine, and
metabolized by the P450 enzyme system. Phenytoin is also affected by a number oxcarbazepine may
of agents that reduce its metabolism and cause it to accumulate. These include exacerbate generalized
amiodarone, fluoxetine, fluvoxamine, isoniazid, and azole antifungal agents. The absence and myoclonic
phenytoin protein-free fraction may increase with hepatic and renal failure, in seizures and should be
avoided in idiopathic
low-protein states, during pregnancy, in old age, and in the presence of highly generalized epilepsy. Other
protein-bound drugs, such as valproate, that compete for protein binding. This antiepileptic drugs that have
is of clinical relevance when decisions are made based on total phenytoin similar potential are
serum concentration. gabapentin, pregabalin,
tiagabine, and vigabatrin.
Phenytoin is effective against focal seizures and generalized tonic-clonic
seizures. Phenytoin is not effective against generalized myoclonic or generalized
absence seizures and may even exacerbate these seizures; hence, it is not a drug
of choice in idiopathic generalized epilepsy.
The usual phenytoin initiation dose is 200 mg/d to 400 mg/d, initially given as
a bedtime dose. Titration is primarily based on clinical response but also takes
into consideration the serum concentration. The recommended “therapeutic”
serum concentration is 10 mg/L to 20 mg/L; the protein-free recommended
“therapeutic” serum concentration is 1 mg/L to 2 mg/L. Protein-free phenytoin
levels should be checked in clinical situations where the protein-free fraction is
expected to be increased. In view of nonlinear kinetics, small increments (eg,
30 mg to 60 mg) should be used when the phenytoin level is in the “therapeutic
range” but the clinical situation warrants optimization of therapy. Extended-release
capsules are preferred. Dosing 2 times a day may be needed when seizures are
drug resistant. Phenytoin can be loaded orally at 18 mg/kg divided into 3 doses
given 2 to 3 hours apart (or even as a single dose if needed).
The IV preparation of phenytoin is associated with local reactions, including
burning pain, phlebitis, cellulitis, and, rarely, the purple glove syndrome. IM

ANTIEPILEPTIC DRUGS
administration is contraindicated because of erratic absorption and sterile

abscess formation. The phenytoin water-soluble prodrug fosphenytoin is
preferred for parenteral use. It has a lower incidence of local reactions with IV
administration. It is also well absorbed after IM administration, which can be
considered in the absence of IV access. When administered intravenously in an
awake individual, it may be associated with paresthesia and pruritis, most often
in the groin region. IV administration of either phenytoin or fosphenytoin can be
associated with hypotension and arrhythmias, so ECG and blood pressure
TABLE 11-3 Suggested Pediatric Antiepileptic Drug Dosinga
Target Total Daily Dose; Usual

Antiepileptic Drug Starting Total Daily Dose Titration Maximal Effective Dose
Phenobarbital 1–3 mg/kg/d 1 mg/kg/d every 1–2 weeks 3 mg/kg/d; up to 8 mg/kg/d
Phenytoin 5–7 mg/kg/d No titration needed 6–8 mg/kg/d; up to 10 mg/kg/d

(may be guided by serum
concentration)
Carbamazepine 10–20 mg/kg/d Increase weekly using 100 mg 20 mg/kg/d; usually <35 mg/kg/d
increments
Oxcarbazepine 8–10 mg/kg/d 5–10 mg/kg/d every 3–7 days as 30–50 mg/kg/d; usually
needed <60 mg/kg/d
Eslicarbazepine 10-20 mg/kg/d (200–400 mg/d 200–400 mg/wk as needed 20–60 mg/kg/d (400–1200 mg/d
acetate depending on weight) depending on weight)
Valproate 15 mg/kg/d 5–10 mg/kg/d every week 30 mg/kg/d; up to 60 mg/kg/d

with enzyme-inducing
antiepileptic drugs
Ethosuximide 10–15 mg/kg/d 5 mg/kg/d every week as 20–30 mg/kg/d; up to 40 mg/kg/d

needed
Clobazam 0.1 mg/kg/d 0.1 mg/kg/d every week as 1.0 mg/kg/d

needed
Felbamate 15 mg/kg/d 15 mg/kg/d every week 45 mg/kg/d
Gabapentin 10–15 mg/kg/d 10 mg/kg/d every day 40 mg/kg/d; up to 50 mg/kg/d
Lamotrigine Monotherapy for weeks 1 and 2: Monotherapy for weeks 3 and 4: Maintenance dose
0.3 mg/kg/d 0.6 mg/kg/d; week 5 and on: for monotherapy:
increase by 0.6 mg/kg/d every 4.5–7.5 mg/kg/d
1–2 weeks
With valproate: 0.15 mg/kg/d With valproate for weeks 3 and 4: With valproate:
0.3 mg/kg/d; week 5 and on: 1–5 mg/kg/d
increase by 0.3 mg/kg/d every
1–2 weeks
With enzyme inducer: With enzyme inducer for weeks 3 With enzyme inducers:
0.6 mg/kg/d and 4: 1.2 mg/kg/d; week 5 and 5–15 mg/kg/d
on: increase by 1.2 mg/kg/d
every 1–2 weeks
CONTINUED ON PAGE 515
514 APRIL 2019

monitoring are recommended, and the rate of IV administration should not
exceed 50 mg/min for phenytoin and 150 mg/min for fosphenytoin.
Phenytoin is less sedating than phenobarbital but nevertheless may have
cognitive adverse effects in some individuals, even within the therapeutic range.
Adverse effects that occur with high concentrations include ataxia,
incoordination, dysarthria, nystagmus, and diplopia. A paradoxical increase in
seizures has been documented with concentrations exceeding 30 mg/L.
Idiosyncratic reactions include allergic rash (almost 6% in a study based on
CONTINUED FROM PAGE 514
Target Total Daily Dose; Usual

Antiepileptic Drug Starting Total Daily Dose Titration Maximal Effective Dose
Topiramate 1–3 mg/kg/d 1–3 mg/kg/d every 1–2 weeks 5–9 mg/kg/d
Levetiracetam 20 mg/kg/d (infants 1 month 10 mg/kg/d every 1–2 weeks Children 4 years to <16 years:
to <6 months of age: 60 mg/kg/d; children 6 months to
14 mg/kg/d) <4 years: 50 mg/kg/d; infants
1 month to <6 months: 42 mg/kg/d
Brivaracetam 1–2 mg/kg/d Dose adjustment based on 1–5 mg/kg/d (pediatric patients
response weighing >50 kg [110 lb]: initial
dose of 50–100 mg/d with
maximum dose of 200 mg/d)
Zonisamideb 1 mg/kg/d 2 mg/kg/d every 2 weeks as Usual dose of 4–8 mg/kg/d with
needed maximum dose of 12 mg/kg/d
Vigabatrin 20 mg/kg/d 20 mg/kg/d every week as 40–60 mg/kg/d

needed
Infantile spasms: 50 mg/kg/d Infantile spasms: Increase to Infantile spasms: 100 mg/kg/d;
100 mg/kg/d after 5 days maximum dose is 150 mg/kg/d
Rufinamide 10 mg/kg/d; in the presence of Increase by 10 mg/kg/d every 45 mg/kg/d; in the presence of
valproate, the starting dose other day; in the presence of valproate, target dose should be
should be ~5 mg/kg/d valproate, titration rate should 20–30 mg/kg/d
be ~5 mg/kg/d every other day
Perampanel 2 mg/d Increase by 2 mg as needed, no 8–12 mg/d

more frequently than weekly
Cannabidiol 5 mg/kg/d Increase by 5 mg/kg/d every 20 mg/kg/d

week as needed
a
Generally applicable to children younger than 12 years of age. The dosing is provided for antiepileptic drugs that have at least been tested in
children.
b
Not US Food and Drug Administration–approved for children.

ANTIEPILEPTIC DRUGS
clinical practice)5 and, rarely, Stevens-Johnson syndrome, toxic epidermal

necrolysis, or hypersensitivity syndrome with fever, rash, lymphadenopathy,
eosinophilia, and liver and renal impairment. Adverse effects associated with
long-term use include gingival hyperplasia, acne, hirsutism, cerebellar atrophy,
decreased bone density, anemia, and peripheral neuropathy.
Place in Therapy
Phenytoin was the most frequently used AED for many years, but its use has
declined considerably since the appearance of newer AEDs with improved
TABLE 11-4 Antiepileptic Drug Teratogenicitya
Antiepileptic Drug Number of Monotherapy Exposures Major Malformation Rateb
Phenobarbitalc >200 High
Phenytoin >500 Intermediated
Carbamazepine >2000 Intermediated
Oxcarbazepine >500 Low
Eslicarbazepine acetate Insufficient exposure Unknown

e
Valproate >1000 Very high
Ethosuximide Insufficient exposure Unknown
Clobazam Insufficient exposure Unknown
Felbamate Insufficient exposure Unknown
Gabapentin >100 Low
Pregabalin Insufficient exposure Unknown
Lamotrigine >4000 Low
Topiramate >500 Intermediate
Tiagabine Insufficient exposure Unknown
Levetiracetam >1000 Low
Brivaracetam Insufficient exposure Unknown
Zonisamide >100 Low
Lacosamide Insufficient exposure Unknown
Vigabatrin Insufficient exposure Unknown
Rufinamide Insufficient exposure Unknown
Perampanel Insufficient exposure Unknown
a
Data are extracted from North American and European registries.1,2 When the two registries differed in
malformation rate, a weighted average was used.
b
Low: <3%; intermediate: 3.1% to 6%; high: 6.1% to 9%; very high: >9%.
c
An additional negative effect is decreased IQ in male offspring.
d
The two registries had different results.
e
Additional negative effects are decreased verbal IQ and autism.
516 APRIL 2019

tolerability and pharmacokinetic KEY POINTS
profiles. Other factors in its
● Unlike phenytoin, the
declining use are its narrow phenytoin prodrug
therapeutic window and the fosphenytoin may
challenge in maintaining the be administered
recommended therapeutic intramuscularly, with
reliable absorption, in the
concentration range without
absence of IV access.
producing toxicity or
underdosing because of ● Carbamazepine induces
nonlinear kinetics as well as its own metabolism, so it has
variable absorption. to be titrated gradually to
the target dose.
CARBAMAZEPINE
Carbamazepine’s mechanism of
action is similar to that of
phenytoin. It blocks the sodium
channel in a voltage-dependent
FIGURE 11-1
and use-dependent fashion, Example of the nonlinear kinetics of phenytoin. An
reducing high-frequency increase in the daily dose beyond 300 mg is
neuronal firing. associated with a disproportionate increase in
serum concentration. At a dose of 400 mg/d, the
Carbamazepine was only
serum concentration is about 13 mg/L. If seizures
available as an oral preparation are still not controlled at this dose, an increment
until a parenteral preparation of 100 mg pushes the serum concentration beyond
was approved in 2016 as 30 mg/L, with clinical toxicity. An increment of
temporary replacement therapy 30 mg would be more appropriate.
when oral administration is not
feasible. Carbamazepine has good oral bioavailability. Its protein binding of
about 75% is not of clinical importance. It is metabolized in the liver, mainly by
CYP 3A4; the most important metabolite is carbamazepine-10,11-epoxide. It is an
active metabolite also responsible for some adverse effects. Carbamazepine is a
potent enzyme inducer, reducing the levels of drugs as well as endogenous
substances metabolized by the CYP enzyme system. Carbamazepine also induces
its own metabolism, a process known as autoinduction, which results in
increased clearance over 2 to 4 weeks, with shortened half-life and lower serum
concentration. Carbamazepine may accumulate when coadministered with
inhibitors of CYP 3A4, such as erythromycin and other macrolide antibiotics
(except azithromycin), fluoxetine, propoxyphene, and grapefruit juice.
Carbamazepine epoxide levels increase with concomitant use of some inhibitors,
such as valproate and felbamate.
Carbamazepine is effective against focal seizures and generalized tonic-clonic
seizures. However, it may exacerbate absence, myoclonic, and atonic seizures.
Hence, it is not a good choice in idiopathic generalized epilepsy. It has US Food
and Drug Administration (FDA) indications for trigeminal neuralgia and for
acute mania and bipolar disorder. The starting dose is 100 mg 2 times a day or
200 mg at bedtime when the extended-release preparation is used. The dose can
be increased by 200 mg every 3 days to a target total daily dosage of 400 mg to
800 mg in two divided doses, and the dose can be increased further, if needed,
for persistent seizures. When immediate-release formulations of carbamazepine
are used, administration in 3 divided doses is recommended, although patients
may have difficulty adhering to this more complex dosing schedule. The

ANTIEPILEPTIC DRUGS
extended-release preparation, indicated for dosing 2 times a day, provides

steadier levels with evidence for improved tolerability as well as efficacy. The
recommended therapeutic range of carbamazepine concentration is 4 mg/L
to 12 mg/L.
Adverse effects noted with carbamazepine include nausea, headache,
dizziness, sedation, and tiredness. Cognitive impairment has been reported on
neuropsychological testing. With elevated levels, there may be blurred vision,
diplopia, nystagmus, unsteadiness, incoordination, and tremor. Hyponatremia
may occur. Weight gain and decreased bone density are reported with long-term
use. Mild leukopenia seen in 10% to 20% of patients is usually benign, although it
may be persistent; the more serious aplastic anemia is rare (estimated at 1 in
200,000). It is advisable to check a complete blood cell count and liver enzymes
before initiating therapy, after 2 to 3 months of treatment, then every 6 to
12 months as needed depending on the clinical setting. Idiosyncratic adverse
experiences include rash, which may be less common than with phenytoin.
Stevens-Johnson syndrome and toxic epidermal necrolysis are rare but more
likely with the HLA-B1502 allele in individuals of Asian descent, for whom
genetic testing of HLA-B polymorphisms is indicated prior to initiation. Other
rare idiosyncratic adverse effects include a lupuslike syndrome, hepatotoxicity,
and hypersensitivity syndrome with fever, rash, and organ involvement.
Carbamazepine use in polytherapy has been associated with increased risk of
spina bifida in infants exposed during gestation. Abrupt withdrawal may be
associated with severe rebound seizures.
Place in Therapy
Carbamazepine had the best balance of efficacy and tolerability in the large
cooperative US Department of Veterans Affairs study that also included
phenytoin, phenobarbital, and primidone.6 As a result, it became the standard
treatment for focal seizures. No drug has been demonstrated to be more effective
than carbamazepine, but its use has declined with the marketing of new AEDs
that have pharmacokinetic advantages. Lamotrigine, oxcarbazepine, and
gabapentin have better tolerability than immediate-release carbamazepine.7–13
However, comparative trials using extended-release carbamazepine have failed
to show superior tolerability of lamotrigine, levetiracetam, zonisamide, or
lacosamide.14–17 Nevertheless, enzyme induction and pharmacokinetic
interactions have been issues favoring newer AEDs. On the other hand, economic
considerations favor the less-expensive carbamazepine.
OXCARBAZEPINE
Oxcarbazepine is a structural analogue of carbamazepine, but the minor
structural differences have resulted in major differences in metabolism and
induction of metabolic pathways. Like carbamazepine and phenytoin,
oxcarbazepine binds to the sodium channel, inhibiting high-frequency repetitive
neuronal firing. Oxcarbazepine is only available as an oral preparation.
Oxcarbazepine has excellent oral bioavailability. It is very rapidly converted to
the monohydroxy derivative, which has two enantiomers, the active
S-licarbazepine, responsible for most of oxcarbazepine’s antiseizure activity
(80%), and R-licarbazepine (less active but contributes to adverse effects).
Its protein binding is not clinically important. The half-life of oxcarbazepine is
only 1 to 3.7 hours, and that of the monohydroxy derivatives is 8 to 10 hours.
518 APRIL 2019

Oxcarbazepine is a weak inducer of CYP 3A4, which is responsible for KEY POINTS
estrogen metabolism, and reduces the efficacy of the oral contraceptive pill at
● The HLA-B1502 allele
high doses, usually greater than 900 mg/d. It is a weak inhibitor of CYP 2C19, is predictive of a
thus raising the phenytoin level when used at high doses. It does not induce carbamazepine-induced
its own metabolism. Unlike carbamazepine, it is not affected by CYP 3A4 severe rash in individuals
inhibitors, such as erythromycin, fluoxetine, propoxyphene, and grapefruit of Asian descent.
juice.
● Oxcarbazepine is more
Oxcarbazepine is effective against focal seizures. It may exacerbate absence likely to cause hyponatremia
and myoclonic seizures and should be avoided in patients with generalized than carbamazepine. Older
epilepsy. It can be started at the dose of 300 mg 2 times a day, but in the absence individuals taking a diuretic
of urgency, it is better to start at 150 mg 2 times a day. The dose can be titrated are at particularly high risk.
by 300 mg per week as needed. The highest dose used in clinical trials was
1200 mg 2 times a day. An extended-release preparation is available, allowing
for once-daily dosing. The recommended therapeutic range for the
monohydroxy derivative is 15 mg/L to 35 mg/L. Conversion from carbamazepine
can be made overnight by using 300 mg of oxcarbazepine for every 200 mg of
carbamazepine when the carbamazepine dose is 800 mg or less. A slower
conversion and lower ratio are advisable with higher carbamazepine doses.
Conversion from carbamazepine may be accompanied by reduction in sodium
concentration and increased levels of concomitant medications metabolized by
the CYP enzyme system.
Oxcarbazepine may cause drowsiness, headache, and fatigue. Higher doses
can cause dizziness, blurred vision, diplopia, nausea, vomiting, and ataxia. Rash
may occur in 2% to 4% of individuals; oxcarbazepine has 25% cross-reactivity
with carbamazepine. Oxcarbazepine is more likely to cause hyponatremia than
carbamazepine is18,19; symptomatic hyponatremia is more likely in older
individuals and those taking a diuretic. Abrupt withdrawal may be associated
with severe rebound seizures.20
Place in Therapy
Oxcarbazepine is approved as a first-line monotherapy for focal seizures.
Multiple comparative monotherapy trials for new-onset focal epilepsy have
demonstrated that oxcarbazepine is equal in efficacy to phenytoin and
immediate-release carbamazepine but with possibly superior tolerability.21,22
Combining oxcarbazepine with other classic sodium channel blockers, such as
carbamazepine, lamotrigine, and phenytoin, may limit tolerability because of
dizziness, diplopia, and ataxia.
ESLICARBAZEPINE ACETATE
Eslicarbazepine acetate was approved for marketing in the United States in
2014, but it is listed here because it represents a third-generation relative of
carbamazepine and oxcarbazepine. It is a prodrug rapidly converted to the active
metabolite S-licarbazepine, also known as eslicarbazepine, the active enantiomer
of the monohydroxy derivative of oxcarbazepine. Eslicarbazepine acts by
blocking sodium channels and stabilizing the inactive state of the voltage-gated
sodium channel. A 2015 study suggested that, unlike carbamazepine, it may
enhance slow inactivation of voltage-gated sodium channels.23 It is available
only as an oral preparation.
Eslicarbazepine is metabolized to inactive compounds, but more than
50% is excreted in the urine as unchanged eslicarbazepine. The half-life of

ANTIEPILEPTIC DRUGS
eslicarbazepine is 13 to 20 hours in plasma and 20 to 24 hours in CSF, justifying

once-daily dosing. Unlike oxcarbazepine, eslicarbazepine acetate is not followed
by a CSF spike, which is suspected to be responsible for acute adverse effects.24
Eslicarbazepine is a weak inducer of CYP 3A4, potentially decreasing plasma
concentrations of estrogen and other molecules metabolized by this enzyme,
and a weak inhibitor of CYP 2C19, potentially increasing the plasma concentration
of phenytoin and other drugs metabolized by this enzyme.
Eslicarbazepine acetate is effective against focal seizures. The recommended
starting dose is 400 mg once daily, to be increased to 800 mg once daily after
1 week. If needed, the dose can be increased again to 1200 mg/d after 1 week.
In a 2015 successful conversion to monotherapy study, a dose of 1600 mg/d
was used.25
Eslicarbazepine acetate has adverse effects similar to oxcarbazepine, although
less frequent. The most common dose-related adverse effects are dizziness,
somnolence, headache, diplopia, nausea, vomiting, fatigue, and ataxia.
Hyponatremia was less commonly reported than in oxcarbazepine trials. Sodium
levels of 125 mEq/L or lower were reported in up to 1.5% of individuals taking
1200 mg/d. Rash occurs in up to 3% of individuals at 1200 mg/d.
Place in Therapy
Eslicarbazepine acetate was first approved by the FDA as adjunctive treatment
for focal seizures. A monotherapy indication followed after successful
completion of a conversion to monotherapy trial.25 Like oxcarbazepine, it should
be avoided in idiopathic generalized epilepsy. Theoretical considerations suggest
eslicarbazepine acetate could be considered a first-line monotherapy for focal
seizures, with tolerability advantages over immediate-release oxcarbazepine.
However, financial considerations may be an obstacle.
VALPROIC ACID/DIVALPROEX SODIUM (VALPROATE)

Valproate has multiple mechanisms of action, including GABA potentiation,
blocking of T-type calcium channels (predictive of efficacy against absence
seizures), and blocking of sodium channels. It is available as oral preparations
(mainly in the form of divalproex sodium, a complex of valproate and sodium
valproate) and parenteral valproate sodium preparation. Oral bioavailability is
almost complete, although slightly less for the extended-release preparation. It is
highly protein bound at about 90%. The free fraction increases with increasing
total concentration and with coadministration of phenytoin, with which it
competes for protein binding.
Valproate is extensively metabolized by conjugation and oxidation. The
half-life in adults is 13 to 16 hours but shorter at about 9 hours with
enzyme-inducing drugs. It is a potent inhibitor, reducing the clearance of
phenobarbital, lamotrigine, rufinamide, and carbamazepine epoxide.
Valproate has a wide spectrum of efficacy against all focal and generalized
seizures, including generalized absence and myoclonic seizures. The divalproex
sodium formulation also has FDA indications for migraine prophylaxis and
bipolar disorder. It should be started at a low dose to improve tolerability. The
extended-release divalproex sodium preparation, which can be administered
once daily, is preferred. The recommended starting dose is 500 mg at bedtime for
the extended-release divalproex sodium preparation or 250 mg 2 times a day for
the delayed-release and immediate-release preparations. The dose can be
520 APRIL 2019

increased gradually as needed to achieve seizure control, up to 1000 mg/d to KEY POINTS
2000 mg/d. It should be avoided in women of childbearing potential because of
● Eslicarbazepine has a long
teratogenic risk. The recommended therapeutic range is 50 mg/L to 100 mg/L. A half-life in CSF, justifying
protein-free concentration should be checked at high levels and in other once-daily oral dosing.
circumstances in which the protein-free fraction is expected to rise.
The adverse effects of valproate include gastric irritation with nausea, ● Valproate has a broad
spectrum of efficacy against
vomiting, and anorexia. Other adverse effects include diarrhea, fatigue,
all focal and generalized
drowsiness, tremor, weight gain, hair loss, peripheral edema, and confusion. seizure types.
Tolerability is generally improved with the extended-release formulation.26–28
Dose-related thrombocytopenia may occur. Endocrine effects are most ● Valproate has the highest
recognized in women and include polycystic ovary syndrome, hyperandrogenism, rate of teratogenicity among
antiepileptic drugs and
hyperinsulinemia and insulin resistance.29,30 Reversible parkinsonism, gait should be avoided in women
disorder, dementia, and brain atrophy have been described with chronic use in of childbearing potential.
seniors. Encephalopathy and hyperammonemia may occur in polytherapy.
Idiosyncratic hepatotoxicity and pancreatitis are potentially life threatening
but rare. Risk factors are polytherapy and young age. Valproate is associated with
a dose-related teratogenicity rate higher than any other marketed AED, with risk
of major malformations higher than 30% at doses greater than 1100 mg/d.31 In
utero exposure is also associated with dose-dependent reduced verbal IQ and
autism.32,33
Place in Therapy
Valproate remains the most effective AED for idiopathic generalized epilepsy
with generalized tonic-clonic seizures and should remain a drug of first choice for
men with generalized epilepsy.34 Although equally effective as ethosuximide for
generalized absence seizures, it has more cognitive adverse effects.35 A large
cooperative US Department of Veterans Affairs study found it less well tolerated
and less effective than carbamazepine for complex partial seizures (focal
impaired awareness seizures), although equally effective for secondarily
generalized tonic-clonic seizures (focal to bilateral tonic-clonic seizures).36
ETHOSUXIMIDE
Ethosuximide blocks T-type calcium currents, which predicts efficacy against
absence seizures. It has an excellent oral bioavailability (greater than 90%).
Protein binding is very low. Ethosuximide is extensively metabolized in the liver.
It has a long half-life of 30 to 60 hours.
Ethosuximide is a narrow-spectrum AED, selective for generalized absence
seizures. The starting dose is 250 mg/d for patients between 3 and 6 years of age
and 250 mg 2 times a day for those older than 6 years of age. The dose can be
increased by 250 mg every week as needed for persistent seizures, not to exceed
500 mg 3 times a day. The recommended therapeutic range is 40 mg/L to
100 mg/L.
Adverse effects include nausea, abdominal discomfort, anorexia, vomiting,
diarrhea, drowsiness, insomnia, nervousness, dizziness, fatigue, ataxia, and
behavior changes. Most adverse effects are dose related and are helped by
administration of divided doses with meals. Headaches, psychosis, depression,
and hallucinations are not clearly dose related. Idiosyncratic adverse experiences
include rash, Stevens-Johnson syndrome, systemic lupus erythematosus, rare
aplastic anemia, thrombocytopenia, agranulocytosis, and rare autoimmune
thyroiditis.

ANTIEPILEPTIC DRUGS
Place in Therapy
Ethosuximide is the AED of choice for absence epilepsy with generalized absence
seizures as the only seizure type, a status supported by the large multicenter
double-blind randomized controlled trial comparing ethosuximide, valproic
acid, and lamotrigine.37
BENZODIAZEPINES
Benzodiazepines act mainly on the GABA-A receptor, increasing the frequency
of GABA-mediated chloride channel openings. Clobazam is the only 1,5-
benzodiazepine, referring to the position of nitrogen atoms in the heterocyclic
ring; other benzodiazepines are 1,4-benzodiazepines. Only clonazepam and
clobazam, used for chronic epilepsy management, are discussed here. In the
United States, they are available only as oral preparations.
Both have good oral bioavailability. Both are highly protein bound.
However, they differ in their metabolism.38 Clonazepam is converted to
inactive metabolites, while clobazam is metabolized in the liver to the active
N-desmethylclobazam. Both clonazepam and clobazam have long half-lives,
justifying once-daily dosing. Both clonazepam and clobazam are broad-spectrum
agents, although their FDA indication is limited to generalized seizure types.
Drowsiness is a common adverse effect that improves over time. It is less likely
with clobazam. With increasing doses, nystagmus, incoordination, unsteadiness,
and dysarthria may occur. Tolerance may develop to the therapeutic effect of
benzodiazepines, but this appears less likely with clobazam. Withdrawal seizures
may occur with abrupt discontinuation. All benzodiazepines are controlled
substances.
Place in Therapy
Both clonazepam and clobazam are typically used as adjunctive therapy and have
limited data to support monotherapy use. The clobazam FDA indication is for
adjunctive treatment of seizures associated with Lennox-Gastaut syndrome.
FELBAMATE
Felbamate was the first second-generation AED approved in the United States in
1993. It has multiple mechanisms of action, including N-methyl-D-aspartate
(NMDA) receptor antagonism, GABA enhancement, and sodium channel
blocking. It is available as an oral preparation.
Felbamate has excellent oral bioavailability; its protein binding is not
clinically significant. It is metabolized in the liver to inactive metabolites, with
a half-life of 20 to 23 hours. It is an inhibitor of CYP 2C19, CYP 1A2, and
β-oxidation, inhibiting the metabolism of phenobarbital, phenytoin, valproate,
carbamazepine epoxide, and warfarin, and it is a weak inducer of CYP 3A4,
decreasing carbamazepine levels and reducing oral contraceptive efficacy.
Felbamate is a broad-spectrum agent effective against focal seizures as well
as generalized seizures in the setting of Lennox-Gastaut syndrome. The
recommended starting dose is 600 mg 2 times a day, with subsequent titration
by 600 mg to 1200 mg per week up to 1200 mg 3 times a day.
The most common adverse effect of felbamate is gastrointestinal irritation
with anorexia, nausea, and vomiting, which can be helped by administration
with food. Felbamate may also cause insomnia, irritability, headache, and weight
loss. The most concerning toxicity is the potentially lethal aplastic anemia,
522 APRIL 2019

with an estimated risk of 1 in 5000 to 1 in 8000 patients, and hepatic failure, KEY POINTS
with an estimated risk of 1 in 26,000 to 1 in 54,000 patients. Both are unlikely
● Ethosuximide is the drug
after 1 year of treatment, and aplastic anemia has not been reported in patients of choice for pure absence
younger than 13 years of age. These two serious adverse effects have resulted in a seizures. While valproate is
boxed warning suggesting that felbamate should be used only for severe epilepsy equally effective, it is
where treatment benefits outweigh the risk. It is recommended to check a associated with more
cognitive adverse effects.
complete blood cell count and liver function test prior to starting felbamate
and to repeat the tests every 2 weeks in the first 6 months of treatment. The ● Tolerance may develop to
frequency of monitoring can be reduced considerably after 1 year of treatment. the therapeutic effect of
benzodiazepines; this
Place in Therapy appears less likely with
clobazam than clonazepam.
Although felbamate was approved for monotherapy, it is not indicated as a
first-line treatment because of its potential serious idiosyncratic toxicity. ● Felbamate-related
Adjunctive therapy or alternative monotherapy can be considered when other aplastic anemia and liver
appropriate and safer options have failed. failure are unlikely to start
after 1 year of treatment.
GABAPENTIN ● Gabapentin bioavailability

Gabapentin binds to the alpha-2-delta subunit of voltage-gated calcium channels, is low and decreases with
reducing the influx of calcium and associated neurotransmitter release under increasing doses.
hyperexcitable conditions. It is available as an oral preparation only.
Gabapentin bioavailability is low and variable between subjects and even in
the same subject. Because of its active saturable transport system from the gut,
its bioavailability decreases with increasing doses, from 60% after a single dose of
300 mg to 29% for 1600 mg 3 times a day.39 Protein binding is negligible. It is
eliminated unchanged in the urine. Its half-life is 5 to 7 hours. It has no known
interactions, other than potential antacid interference with its absorption.
Gabapentin is a narrow-spectrum agent against focal seizures. It may cause
exacerbation of myoclonus.40 It is also FDA approved for the treatment of
postherpetic neuralgia. An extended-release preparation (gabapentin enacarbil)
has been approved for the treatment of restless legs syndrome, and another
(gastroretentive dosage form) has been approved for the management of
postherpetic neuralgia.
The recommended starting dose of gabapentin is 300 mg/d to 400 mg/d, to
be increased by 300 mg to 400 mg every day up to 300 mg to 400 mg 3 times
a day. The dose can be increased as needed up to 4800 mg/d in 3 divided doses.
Adverse effects include drowsiness, dizziness, ataxia, tiredness, and weight
gain. It may cause myoclonus. It may cause cognitive slowing in the elderly and
emotional lability in children. Peripheral edema is more likely with increasing
age. Gabapentin was recently reclassified as a controlled substance in some states.
Place in Therapy
Gabapentin can be used as adjunctive treatment for focal seizures. It is often
chosen for its anecdotal benefit in the treatment of headache and other pain and
its benefit for sleep. Although approved in Europe for initial monotherapy, a
large randomized comparative trial found it less effective than lamotrigine.10
PREGABALIN
Pregabalin is structurally related to gabapentin and has a similar mechanism of
action. It is also available only as an oral preparation. Unlike gabapentin,
pregabalin has very good oral bioavailability, which is independent of dose. Like

ANTIEPILEPTIC DRUGS
gabapentin, it has no protein binding and is not metabolized in humans, and it

has no known interactions. It is excreted unchanged in the urine. Its half-life is
about 6 hours.
Pregabalin is a narrow-spectrum drug against focal seizures. The official FDA
epilepsy indication is adjunctive therapy for adult patients with partial onset
seizures. Like gabapentin, pregabalin has a narrow spectrum of efficacy against
focal seizures and may exacerbate generalized myoclonic and absence seizures. It
also has an FDA indication for neuropathic pain associated with diabetic
peripheral neuropathy, postherpetic neuralgia, and fibromyalgia. The starting
dose is 75 mg 1 time (at bedtime) or 2 times a day. The dose can then be increased
by 75 mg to 150 mg every week as needed, until seizure control, appearance of
adverse effects, or reaching a maximum dose of 300 mg 2 times a day.
The adverse effects of pregabalin include dizziness, somnolence, increased
appetite, weight gain, and peripheral edema. Myoclonus may occur with higher
doses in some individuals. Pregabalin is a controlled substance because of the
potential for abuse.
Place in Therapy
Pregabalin is indicated as adjunctive therapy for focal seizures. It was inferior to
lamotrigine as first-line therapy41 and should probably not be used as a first-line
treatment. However, a conversion-to-monotherapy study was successful.42
LAMOTRIGINE
Lamotrigine blocks sodium channels, like phenytoin and carbamazepine, but
must have other unrecognized actions to explain efficacy against absence
seizures. It is available as an oral preparation only.
Lamotrigine has an excellent oral bioavailability. Its protein binding is not
clinically significant. It is extensively metabolized in the liver, predominantly by
glucuronidation, and then eliminated in the urine. The half-life is about 24 hours
in monotherapy, at least twice as long when used with valproate, and about half
as long when used with an enzyme inducer. Estrogen and pregnancy increase
lamotrigine clearance.
Lamotrigine is a broad-spectrum AED, although its FDA indications are
limited to focal seizures, generalized tonic-clonic seizures, and Lennox-Gastaut
syndrome. It is less effective against generalized absence seizures than valproate
and ethosuximide. It may be effective against myoclonic seizures in some
patients but may exacerbate these seizures in others. Lamotrigine also has an
FDA indication for maintenance treatment in bipolar I disorder.
Lamotrigine requires a very slow titration to avoid the development of rash.
In monotherapy, it should be initiated with 25 mg/d for 2 weeks, followed by
50 mg/d for 2 weeks, then 100 mg/d. The dose can then be increased as needed
by 100 mg every 2 weeks. The titration rate is half as fast with adjunctive
valproic acid but can be twice as fast in the presence of an enzyme inducer
and absence of valproic acid. A serum concentration is helpful to guide further
titration if seizures are still not controlled at a dose of 600 mg/d. The suggested
therapeutic range is 2 mg/L to 20 mg/L.43 The extended-release preparation
allows once-daily dosing and reduces toxicity from peak levels. It may even
improve efficacy when used 2 times a day in patients who are drug resistant.44
Dose-related adverse effects include dizziness, blurred vision, diplopia,
unsteadiness, nausea and vomiting, headache, and tremor. A serum
524 APRIL 2019

concentration is indicated for symptoms that could be consistent with KEY POINTS
lamotrigine toxicity, particularly if the baseline concentration was greater than
● Like gabapentin,
10 mg/L.45 Rash is seen in about 3% of patients, with a higher incidence in pregabalin has a narrow
children, with coadministration of valproic acid, and with faster titration and spectrum of efficacy against
higher doses. The risk of rash is increased in patients with a prior rash when on focal seizures and may
carbamazepine or phenytoin.46 Stevens-Johnson syndrome, toxic epidermal exacerbate generalized
myoclonic and absence
necrolysis, hypersensitivity syndrome, and hemophagocytic lymphohistiocytosis
seizures.
are rare serious idiosyncratic adverse effects.
● Lamotrigine clearance is
Place in Therapy decreased by valproate and
Lamotrigine is an important first-line AED for focal seizures and generalized increased by estrogen and
pregnancy as well as by
tonic-clonic seizures. Several comparative trials have favored lamotrigine over other enzyme inducers.
AEDs for focal seizures in the balance of tolerability and efficacy.10,13 However, it
was inferior to valproic acid for idiopathic generalized epilepsy34 and inferior to
ethosuximide for generalized absence seizures.37 Lamotrigine is less sedating and
has fewer cognitive adverse effects than traditional AEDs. Its monotherapy use is
associated with one of the lowest rates of teratogenicity, favoring its use in women
of childbearing age. Lamotrigine may have pharmacodynamic interactions with
other classic sodium channel blockers, resulting in adverse effects at lower than
expected serum concentrations. However, its combination with valproate can be
synergistic, with greater efficacy than predicted.47,48
TOPIRAMATE
Topiramate has multiple mechanisms of action, including antagonism of
α-amino-3-hydroxy-5-methylisoxazole-4-proprionic acid (AMPA)/kainate
receptors, augmentation of GABA activity, and blocking of voltage-gated sodium
channels. It is also a weak carbonic anhydrase inhibitor, but this mechanism does
not contribute significantly to its efficacy. It is available as an oral preparation.
Topiramate has an excellent oral bioavailability. Its protein binding is not
clinically significant. It is partially metabolized in the liver, with about 70%
eliminated unchanged in the urine. Its half-life is approximately 21 hours. It is a
mild inducer of CYP 3A4, reducing the efficacy of the oral contraceptive at a dose
greater than 200 mg/d, and a mild inhibitor of CYP 2C19.
Topiramate is a broad-spectrum AED effective against focal and generalized
tonic-clonic seizures. A pilot trial suggested it is not effective for generalized
absence seizures.49 It is FDA approved for migraine prophylaxis and as a
weight-loss preparation in combination with phentermine. It is also frequently
used off-label for bipolar disorder. Topiramate has to be titrated gradually to
manage cognitive adverse effects. It is suggested to start with 25 mg/d and
increase the dose by 25 mg every week up to 100 mg/d. Further titration by 25 mg
to 50 mg every week can be considered, up to 400 mg/d in 2 divided doses.
Extended-release preparations with once-daily dosing may improve tolerability.
Topiramate is less well tolerated than lamotrigine, the main tolerability issue
being cognitive adverse effects, including cognitive slowing, decreased attention
and memory, impaired executive function, word-finding difficulty, and reduced
verbal fluency. Patients may not be aware of these cognitive difficulties.50,51
Other adverse effects include sedation, fatigue, dizziness, ataxia, and depression.
Kidney stones occur in about 1.5% of individuals. Decreased appetite and weight
loss may also occur. Paresthesia in the hands and feet can occur with initiation
and with dose increase but usually resolve. This is due to the carbonic anhydrase

ANTIEPILEPTIC DRUGS
inhibition activity of this drug. Oligohidrosis, hyperthermia, and metabolic

acidosis may occur, more commonly in children. Acute myopia and secondary
angle-closure glaucoma are reported rarely. Hyperammonemia may occur when
topiramate is used in conjunction with valproate. Topiramate is associated with
increased birth defects at a rate of approximately 4%, particularly oral clefts.52
Place in Therapy
Although topiramate is FDA approved for initial monotherapy for focal seizures
and generalized tonic-clonic seizures, it is not a drug of first choice because of its
cognitive adverse effects, unless its use is justified by comorbidity, such as
headache or obesity. It is effective as adjunctive therapy for focal and generalized
seizures and Lennox-Gastaut syndrome.
TIAGABINE
Tiagabine inhibits GABA reuptake at the synapse. It is available as an oral
preparation only.
Tiagabine has an excellent oral bioavailability. It is 96% protein bound, but this
is of limited importance because dosing decisions are not dependent on the level,
and its serum concentration is so low that it does not significantly compete for
protein binding. It is extensively metabolized in the liver. Its half-life is 7 to 9 hours
in monotherapy, shortened to 2 to 5 hours in the presence of an enzyme inducer.
Tiagabine has a narrow spectrum of efficacy against focal seizures only. It may
exacerbate generalized absence and myoclonic seizures. It is used off-label in the
management of spasticity in multiple sclerosis, in the treatment of addiction, and
to increase deep sleep proportion. It should be started at 4 mg at bedtime and
increased by 4 mg every week to an initial target dose of 8 mg 3 times a day. The
dose can be increased further by 4 mg every week up to 12 mg to 16 mg 3 times a
day. A higher dose may be used in the presence of an enzyme inducer.
The most common adverse effects are dizziness, asthenia, nervousness,
tremor, depression, and emotional lability, which are more common during
titration. Tiagabine may be associated with dose-related episodes of
nonconvulsive status epilepticus or encephalopathy, which may occur even in
the absence of epilepsy.53,54
Place in Therapy
Tiagabine should be reserved for use as adjunctive therapy for focal seizures.
LEVETIRACETAM
Levetiracetam’s main mechanism of action is binding to the synaptic vesicle
protein SV2A. This seems to result in nonspecific decrease in neurotransmitter
release in a state of neuronal hyperactivation.55 Levetiracetam is available in oral
and IV formulations.
Levetiracetam has an excellent oral bioavailability and very low protein
binding. It has no hepatic metabolism; 66% is excreted unchanged in the urine,
and the rest is hydrolyzed to inactive compounds. The half-life is 6 to 8 hours. It
has no known significant pharmacokinetic interactions.
Levetiracetam is a broad-spectrum drug, effective against focal seizures,
generalized tonic-clonic seizures, and generalized myoclonic seizures.
Levetiracetam is the only AED with Class I evidence for efficacy against
myoclonic seizures. It is best to start with 500 mg/d in 2 divided doses or once
526 APRIL 2019

at bedtime with the extended-release preparation. The dose can then be KEY POINTS
increased as needed and as tolerated up to 3000 mg/d to 4000 mg/d.
● Tiagabine may be
The most common adverse effects include somnolence, dizziness, and associated with
asthenia. Irritability and hostility may occur, more often in children. Depression dose-related episodes of
may also occur. nonconvulsive status
epilepticus or
encephalopathy, even
Place in Therapy
in subjects who do not
Although levetiracetam is not FDA approved for monotherapy in the have epilepsy.
United States, it is used widely as a first-line treatment for focal and generalized
tonic-clonic seizures and is approved for initial monotherapy in Europe. It is ● Levetiracetam is the only
also an excellent adjunctive treatment in view of its safety and absence of antiepileptic drug with Class
I evidence of efficacy
interactions. The IV preparation has been used as a second-line agent in the against generalized
treatment of status epilepticus.56–58 myoclonic seizures.
BRIVARACETAM ● Brivaracetam may have

fewer behavioral side
Brivaracetam is structurally related to levetiracetam and has a similar mechanism effects than levetiracetam.
of action through binding to SV2A but with approximately 20-fold higher
affinity and greater selectivity. It also has a higher brain permeability than
levetiracetam. It is available in oral and IV formulations.
Brivaracetam has an excellent bioavailability after oral administration. It is
weakly bound to plasma proteins. Its half-life is approximately 9 hours.
It is renally excreted after extensive metabolism, primarily by hydrolysis and
to a lesser extent hydroxylation mainly via CYP 2C19. Brivaracetam has more
interactions than levetiracetam. Its clearance is increased by enzyme inducers.
It may increase carbamazepine epoxide and may also increase phenytoin
concentration by up to 20%.
Although brivaracetam has a broad spectrum of efficacy in preclinical models,
human Class I trials have only been conducted in patients with focal seizures.
Pooled analyses demonstrated efficacy greater than placebo at 50 mg/d,
100 mg/d, and 200 mg/d administered in 2 divided doses as adjunctive therapy.59
The recommended starting dose is 50 mg 2 times a day, followed by adjustment
based on response and tolerability, either down to 25 mg 2 times per day or up
to 100 mg 2 times a day. The most commonly reported adverse experiences
occurring more often than placebo were somnolence, dizziness, and fatigue.
Irritability was reported only in 3.2% of patients receiving brivaracetam
compared with 1.1% of those receiving placebo.
Place in Therapy
Brivaracetam is FDA approved for the treatment of partial onset seizures in
patients 16 years of age and older. This indication includes monotherapy and
adjunctive use of the drug, although it has not specifically undergone
monotherapy trials. The avenue for this approval is via a new pathway for
monotherapy approval put forth by the FDA in a General Advice Letter in
September 2016, stating that for approved antiseizure drugs “it is acceptable
to extrapolate the efficacy and safety of drugs approved as adjunctive therapy
for the treatment of partial onset seizures (POS) to their use as monotherapy for
the treatment of POS.”60
Brivaracetam is not effective when added to levetiracetam.61 One small
open-label study suggested that behavioral adverse effects from levetiracetam
may improve after switching to brivaracetam.62

ANTIEPILEPTIC DRUGS
ZONISAMIDE
Zonisamide is structurally related to sulfonamides. It has multiple mechanisms of
action, including blocking T-type calcium channels (predictive of efficacy
against absence seizures), blocking sodium channels, and weak inhibition of
carbonic anhydrase activity. It is available only as an oral preparation.
Zonisamide has excellent oral bioavailability. Protein binding is not
clinically significant. It is metabolized in the liver to inactive metabolites. It
has a long half-life of about 60 hours. It is not a hepatic enzyme inducer
or inhibitor.
Zonisamide is considered a broad-spectrum AED, although Class I trials have
only been conducted in patients with focal seizures. The starting dose is 100 mg
at bedtime for 2 weeks, then 200 mg at bedtime. The dose can be increased by
100 mg every 2 weeks as needed, up to 600 mg/d once at bedtime or in 2 divided
doses. The suggested therapeutic range for plasma concentration is 10 mg/L
to 40 mg/L.
Adverse effects include sedation, ataxia, dizziness, nausea, fatigue,
agitation/irritability, and anorexia. Weight loss may occur. Cognitive slowing
and difficulty with concentration may be seen, particularly at higher doses, but
are less pronounced than with topiramate. Rarely, depression and psychosis may
occur. Serious rash, such as Stevens-Johnson syndrome and toxic epidermal
necrolysis, occurs rarely. Kidney stones occur in up to 4% of patients but may be
prevented with adequate hydration. Oligohidrosis, hyperthermia, and metabolic
acidosis occur rarely, more often in children.
Place in Therapy
Zonisamide is indicated as initial monotherapy for focal seizures in Europe. In
Japan, it is also indicated as monotherapy for generalized seizures. The official
FDA indication is for adjunctive therapy for focal seizures. Zonisamide is rarely
the first-choice agent for initial monotherapy because of its cognitive adverse
effects. However, its long half-life could be an advantage, reducing the impact of
a missed dose.
LACOSAMIDE
Lacosamide blocks sodium channels, enhancing slow inactivation, unlike most
classic sodium channel blockers, which enhance fast sodium channel
inactivation. It is available in oral as well as parenteral formulations.
Oral bioavailability is excellent. Protein binding is not clinically significant.
Lacosamide is converted in the liver to inactive metabolites, but approximately
40% is eliminated unchanged in the urine. The half-life is approximately
13 hours.
Lacosamide appears to be a narrow-spectrum AED against focal seizures.
However, preliminary data suggest that it does not exacerbate absence or
myoclonic seizures. The starting dose is 100 mg/d (once at bedtime or in 2
divided doses) for 1 week, then 100 mg 2 times a day. The dose can then be
titrated as needed by 100 mg every 1 to 2 weeks until seizures are controlled, side
effects appear, or a dose of 600 mg/d is reached.
The most common adverse effects include dizziness, headache, nausea,
vomiting, diplopia, fatigue, and sedation, all of which are more common at
higher doses. These adverse effects are also more likely when lacosamide is used
in conjunction with other sodium channel blockers.63 Lacosamide may produce a
528 APRIL 2019

dose-dependent prolongation in PR interval, which could be clinically significant KEY POINTS
in patients with known conduction problems, or if it is combined with other
● Zonisamide’s long
drugs that have a similar effect. half-life of about 60 hours
may be an advantage in
Place in Therapy reducing the impact of a
Lacosamide is indicated as monotherapy and as adjunctive therapy for focal missed dose.
seizures. The parenteral formulation is indicated as short-term replacement
● Lacosamide may produce
when oral administration is not feasible in patients taking oral lacosamide; a dose-dependent
several anecdotal reports also exist of efficacy in status epilepticus. When prolongation in PR interval,
lacosamide is used as adjunctive therapy, it may have greater efficacy and better which could be clinically
tolerability if it is combined with a non–sodium channel drug.63 significant in patients with
known conduction
problems, or if it is
VIGABATRIN combined with other drugs
Vigabatrin is an irreversible inhibitor of GABA transaminase, resulting in that have a similar effect.
accumulation of GABA. It is available as an oral formulation. Vigabatrin
● Long-term vigabatrin use
has excellent oral bioavailability and no protein binding. It is not significantly may be associated with
metabolized and is eliminated unchanged in the urine. The half-life is irreversible visual field
10.5 hours in young adults and 5 to 6 hours in infants. However, its duration constriction; hence, it
of action outlasts its presence in serum.64 Vigabatrin is a weak inducer of should only be continued if
it produces a remarkable
CYP 2C9.
improvement in seizure
Vigabatrin is a narrow-spectrum drug effective against focal seizures. It may control.
worsen absence and myoclonic seizures in idiopathic generalized epilepsy.40
However, it is effective against infantile spasms, particularly in the presence
of tuberous sclerosis. The starting adult dose is 500 mg 2 times a day, then
it is titrated by 500 mg per week up to 1.5 g 2 times a day. The dose can be
increased further, as needed, up to 3 g 2 times a day, but this increases
the risk of adverse effects with a low chance of additional therapeutic
benefit.
Common vigabatrin adverse effects include sedation, fatigue, dizziness,
and ataxia. Irritability, behavior changes, psychosis, and depression may also
be observed. Weight gain may occur. The most concerning adverse effect is
a progressive and permanent bilateral concentric visual field constriction,
which may occur in up to 30% to 40% of individuals.65 The risk increases with
increased daily dose and increased duration of therapy.66
Place in Therapy
Vigabatrin use is reserved for adjunctive therapy in subjects who have failed
several alternative treatments and monotherapy in infants with infantile spasms.
Because of the visual toxicity, periodic visual assessment is recommended at
baseline and every 3 months, and treatment should be continued only if
considerable benefit is observed in the first 3 months.
RUFINAMIDE
Rufinamide is a sodium channel blocker, although additional mechanisms of
action are likely. It is available only as an oral preparation. Oral bioavailability
is very good with food but is decreased in the absence of food. Protein binding
is not clinically significant. It is metabolized by enzymatic hydrolysis to an
inactive metabolite eliminated in the urine. The half-life is approximately 6
to 10 hours. It is a weak inhibitor of CYP 2E1 and a weak inducer of CYP 3A4
and uridine diphosphate glucuronyltransferase (UDP-GT). The addition of

ANTIEPILEPTIC DRUGS
valproate decreases rufinamide clearance and increases rufinamide levels by up

to 70%.
Rufinamide is a broad-spectrum AED, but its efficacy against focal seizures
was not sufficient for an FDA indication. The starting dose is 400 mg/d, after
which it is increased by 400 mg every other day until seizure control or until
a daily dose of 3200 mg is reached (in 2 divided doses).
The adverse effects of rufinamide include dizziness, fatigue, somnolence, and
headache. Vomiting may occur in children. Rufinamide may cause a shortening
of the QT interval.
Place in Therapy
Rufinamide is FDA indicated as adjunctive treatment for seizures associated with
Lennox-Gastaut syndrome.
EZOGABINE (RETIGABINE)
Ezogabine (known as retigabine outside the United States) was a promising
new AED with a novel mechanism of action as a potassium channel opener.
However, long-term use was associated with bluish pigmentation in the skin,
nails, and retina. Its use declined to the point that its maker withdrew it from
the market in 2017, which is why it will not be discussed further here.
PERAMPANEL
Perampanel is a selective noncompetitive AMPA glutamate receptor antagonist.
It is available as an oral preparation. It has excellent oral bioavailability and is
95% protein bound. It is extensively metabolized in the liver. It has a long
half-life of about 105 hours. At a dose of 12 mg (not 8 mg), it accelerates the
metabolism of levonorgestrel, a progesterone component of the oral
contraceptive pill.67 Perampanel is effective for focal seizures and generalized
tonic-clonic seizures.68
The adverse effects of perampanel include dizziness, somnolence, headache,
fatigue, ataxia, and blurred vision. Aggression and hostility may occur, with an
estimated incidence of about 20% at a dose of 12 mg/d, resulting in a boxed
warning.69 Behavioral changes were more common in patients with
intellectual disability.70
Place in Therapy
Perampanel is indicated for focal seizures (adjunctive and monotherapy) and as
adjunctive treatment for primary generalized tonic-clonic seizures. Although
there is no FDA indication for myoclonic seizures, several case reports and case
series suggest particular efficacy in progressive myoclonic epilepsies, which are
usually resistant to therapy.71–74
CANNABIDIOL
Cannabidiol was marketed in the United States in November 2018. It is a
cannabinoid but does not interact with the cannabinoid receptor CB1 and does
not share the psychoactive properties of tetrahydrocannabinol. Its exact
mechanisms of action are not known, but it may enhance GABA activity
through allosteric modulation of the GABA-A receptor and enhancement of
currents elicited by low GABA concentrations.75 Its bioavailability is
increased by administration with a high-fat meal. It is highly protein bound
530 APRIL 2019

(>94%). Cannabidiol is metabolized in the liver, primarily by CYP2C19 and KEY POINTS
CYP3A4 enzymes, and converted to an active then an inactive metabolite.
● Valproate reduces
Its clearance is increased by inducers and decreased by inhibitors of CYPC19 rufinamide clearance; as a
and CYP3A4. It interacts with several AEDs, most notably with clobazam, result, rufinamide has to be
increasing the concentration of its active metabolite N-desmethylclobazam.76,77 started at a lower dose and
Cannabidiol is available only as an oral solution. The recommended starting dose titrated more slowly in the
presence of valproate.
is 5 mg/kg/d in 2 divided doses for 1 week, then 10 mg/kg/d in 2 divided doses.
Its most common adverse effects are sedation, fatigue, decreased appetite, and ● Perampanel has a very
diarrhea. It may produce an increase in liver enzymes, particularly when used in long half-life, justifying
conjunction with valproate or with valproate and clobazam. Liver enzymes and once-daily dosing.
total bilirubin levels should be obtained before treatment and at 1, 3, and
● Antiepileptic drug
6 months after initiation of treatment. combinations with different
mechanisms of action may
Place in Therapy have a greater probability
Cannabidiol is FDA indicated for the treatment of seizures associated with of success.
Lennox-Gastaut syndrome or Dravet syndrome in patients 2 years of age and older
based on blinded controlled trials.78–80 Open-label trials also suggest efficacy for
other forms of epilepsy.81–83 Artisanal cannabidiol formulations are used without
prescription by many patients with epilepsy in the United States.
STIRIPENTOL
Stiripentol was FDA approved in 2018 for the treatment of seizures associated
with Dravet syndrome in patients also taking clobazam. Its mechanism of action
may involve both direct interaction with the GABA-A receptor and inhibition
of CYP enzyme activity resulting in increased concentration of clobazam and its
active metabolite. At the time of publication, it was not being marketed in the
United States.
USE OF ANTIEPILEPTIC DRUGS IN COMBINATION

If the first AED fails because of lack of tolerability, it should be replaced with
an alternative monotherapy. If the first AED fails because of lack of efficacy,
options of replacement monotherapy or adjunctive therapy seem to be equal.84
Substitution monotherapy is favored when the first AED was not well tolerated
or was totally ineffective. Substitution monotherapy would also be preferable in
elderly patients who already take other medications, in women of childbearing
potential contemplating pregnancy, in patients with compliance challenges, and
when financial restrictions exist. Add-on therapy would be preferred if the first
AED was well tolerated and partially effective or if the projected add-on agent
has not been tested in monotherapy. The add-on therapy should not have
negative pharmacokinetic interactions with the first AED or other concomitant
medications.85 For example, the use of an enzyme inducer with an AED whose
metabolism can be induced will reduce its efficacy. Enzyme inhibition is less of a
problem as long as dosing accommodations are made. Evidence exists that
combining two AEDs with different mechanisms of action is associated with
greater balance of tolerability and efficacy.86 In particular, combining two sodium
channel blockers tends to be associated with pharmacodynamic interactions
such that adverse effects may be seen even though serum concentrations are in
the therapeutic range. Several combinations seem to have synergistic efficacy in
animal models,87 but only one combination has been demonstrated to be
synergistic in humans, the combination of lamotrigine and valproate.48

ANTIEPILEPTIC DRUGS
CONCLUSION
In conclusion, many AEDs are available for the treatment of epilepsy, with
specific advantages and disadvantages. Some AEDs have additional efficacy in
the treatment of comorbidities such as migraine or bipolar disorder.
Considerations in AED choice include the AED’s efficacy profile as well as
patient-specific factors. AED combinations should avoid unfavorable
pharmacokinetic and pharmacodynamic interactions.
The most notable developments since the last version of this article are the
FDA approval of three new AEDs, new practice guidelines for the efficacy and
tolerability of the new AEDs,88,89 and new extrapolation policies of the FDA. The
pediatric extrapolation policy allows efficacy data against focal seizures in adults
to apply to children 4 years of age or older, although safety studies would still be
needed for pediatric approval. Based on this policy, lacosamide, eslicarbazepine,
brivaracetam, and perampanel were approved for the treatment of focal seizures
in children aged 4 years and older. The FDA also allowed extrapolation of
monotherapy use of a drug proven effective as adjunctive therapy. There has also
been increasing awareness of autoimmune pathophysiology underlying epilepsy
in many patients, often requiring immunotherapy for optimal management.
Improved understanding of the underlying pathophysiology of epilepsy in
individual patients will allow more specific AED therapy in the future.
REFERENCES
1 Tomson T, Battino D, Bonizzoni E, et al. Comparative 7 Brodie MJ, Overstall PW, Giorgi L. Multicentre,
risk of major congenital malformations with eight double-blind, randomised comparison between
different antiepileptic drugs: a prospective cohort lamotrigine and carbamazepine in elderly
study of the EURAP registry. Lancet Neurol 2018; patients with newly diagnosed epilepsy. The UK
17(6):530–538. doi:10.1016/S1474-4422(18)30107-8. Lamotrigine Elderly Study Group. Epilepsy Res
1999;37(1):81–87. doi:10.1016/S0920-1211(99)00039-X.
2 Hernandez-Diaz S, Smith CR, Shen A, et al.
Comparative safety of antiepileptic drugs during 8 Brodie MJ, Richens A, Yuen AW. Double-blind
pregnancy. Neurology 2012;78(21):1692–1699. comparison of lamotrigine and carbamazepine
doi:10.1212/WNL.0b013e3182574f39. in newly diagnosed epilepsy. UK Lamotrigine/
Carbamazepine Monotherapy Trial Group.
3 Vajda FJ, Eadie MJ. The clinical pharmacology of
Lancet 1995;345(8948):476–479. doi:10.1016/
traditional antiepileptic drugs. Epileptic Disord
S0140-6736(95)90581-2.
2014;16(4):395–408. doi:10.1684/epd.2014.0704.
9 Dam M, Ekberg R, Løyning Y, et al. A double-
4 Ding D, Zhang Q, Zhou D, et al. Cognitive and
blind study comparing oxcarbazepine and
mood effects of phenobarbital treatment in
carbamazepine in patients with newly
people with epilepsy in rural China: a prospective
diagnosed, previously untreated epilepsy.
study. J Neurol Neurosurg Psychiatry. 2012;83(12):
Epilepsy Res 1989;3(1):70–76. doi:10.1016/
1139–1144. doi:10.1136/jnnp-2012-303042.
0920-1211(89)90070-3.
5 Arif H, Buchsbaum R, Weintraub D, et al.
10 Marson AG, Al-Kharusi AM, Alwaidh M, et al.
Comparison and predictors of rash associated
The SANAD study of effectiveness of
with 15 antiepileptic drugs. Neurology 2007;
carbamazepine, gabapentin, lamotrigine,
68(20):1701–1709. doi:10.1212/01.wnl.0000261917.
oxcarbazepine, or topiramate for treatment of
83337.db.
partial epilepsy: an unblinded randomised
6 Mattson RH, Cramer JA, Collins JF, et al. controlled trial. Lancet 2007;369(9566):
Comparison of carbamazepine, phenobarbital, 1000–1015. doi:10.1016/S0140-6736(07)60460-7.
phenytoin, and primidone in partial and
11 Reinikainen KJ, Keränen T, Halonen T, et al.
secondarily generalized tonic-clonic seizures.
Comparison of oxcarbazepine and carbamazepine:
N Engl J Med 1985;313(3):145–151. doi:10.1056/
a double-blind study. Epilepsy Res 1987;1(5):
NEJM198507183130303.
284–289. doi:10.1016/0920-1211(87)90003-9.
532 APRIL 2019

12 Reunanen M, Dam M, Yuen AW. A randomised 23 Hebeisen S, Pires N, Loureiro AI, et al.
open multicentre comparative trial of lamotrigine Eslicarbazepine and the enhancement of slow
and carbamazepine as monotherapy in patients inactivation of voltage-gated sodium channels: a
with newly diagnosed or recurrent epilepsy. comparison with carbamazepine, oxcarbazepine
Epilepsy Res 1996;23(2):149–155. doi:10.1016/ and lacosamide. Neuropharmacology 2015;89:
0920-1211(95)00085-2. 122–135. doi:10.1016/j.neuropharm.2014.09.008.
13 Rowan AJ, Ramsay RE, Collins JF, et al. New onset 24 Nunes T, Rocha JF, Falcão A, et al. Steady-
geriatric epilepsy: a randomized study of state plasma and cerebrospinal fluid
gabapentin, lamotrigine, and carbamazepine. pharmacokinetics and tolerability of
Neurology 2005;64(11):1868–1873. doi:10.1212/ eslicarbazepine acetate and oxcarbazepine in
01.WNL.0000167384.68207.3E. healthy volunteers. Epilepsia 2013;54(1):108–116.
doi:10.1111/j.1528-1167.2012.03595.x.
14 Baulac M, Brodie MJ, Patten A, et al. Efficacy
and tolerability of zonisamide versus 25 Sperling MR, Harvey J, Grinnell T, et al. Efficacy
controlled-release carbamazepine for newly and safety of conversion to monotherapy
diagnosed partial epilepsy: a phase 3, with eslicarbazepine acetate in adults with
randomised, double-blind, non-inferiority trial. uncontrolled partial-onset seizures: a
Lancet Neurol 2012;11(7):579–588. doi:10.1016/ randomized historical-control phase III study
S1474-4422(12)70105-9. based in North America. Epilepsia 2015;56(4):
546–555. doi:10.1111/epi.12934.
15 Baulac M, Rosenow F, Toledo M, et al.
Efficacy, safety, and tolerability of lacosamide 26 Smith MC, Centorrino F, Welge JA, Collins MA.
monotherapy versus controlled-release Clinical comparison of extended-release
carbamazepine in patients with newly diagnosed divalproex versus delayed-release divalproex:
epilepsy: a phase 3, randomised, double-blind, pooled data analyses from nine trials. Epilepsy
non-inferiority trial. Lancet Neurol 2017;16(1): Behav 2004;5(5):746–751. doi:10.1016/j.yebeh.
43–54. doi:10.1016/S1474-4422(16)30292-7. 2004.07.007.
16 Brodie MJ, Perucca E, Ryvlin P, et al. Comparison 27 Pierre-Louis SJ, Brannegan RT, Evans AT. Seizure
of levetiracetam and controlled-release control and side-effect profile after switching
carbamazepine in newly diagnosed epilepsy. adult epileptic patients from standard to
Neurology 2007;68(6):402–408. doi:10.1212/01. extended-release divalproex sodium. Clin Neurol
wnl.0000252941.50833.4a. Neurosurg 2009;111(5):437–441.
17 Saetre E, Perucca E, Isojärvi J, et al. An 28 Thibault M, Blume WT, Saint-Hilaire JM, Zakhari R,
international multicenter randomized Sommerville KW. Divalproex extended-release
double-blind controlled trial of lamotrigine versus the original divalproex tablet: results of a
and sustained-release carbamazepine randomized, crossover study of well-controlled
in the treatment of newly diagnosed epilepsy epileptic patients with primary generalized
in the elderly. Epilepsia 2007;48(7):1292–1302. seizures. Epilepsy Research 2002;50(3):243–249.
doi:10.1111/j.1528-1167.2007.01128.x.
29 Belcastro V, D'Egidio C, Striano P, Verrotti A.
18 Berghuis B, van der Palen J, de Haan GJ, et al. Metabolic and endocrine effects of valproic acid
Carbamazepine- and oxcarbazepine-induced chronic treatment. Epilepsy Res 2013;107(1-2):1–8.
hyponatremia in people with epilepsy. Epilepsia
30 Verrotti A, Mencaroni E, Cofini M, et al. Valproic
2017;58(7):1227–1233. doi:10.1111/epi.13777.
acid metabolism and its consequences on sexual
19 Dong X, Leppik IE, White J, Rarick J. Hyponatremia functions. Curr Drug Metab 2016;17(6):573–581.
from oxcarbazepine and carbamazepine.
31 Vajda FJ, O'brien TJ, Hitchcock A, et al. Critical
Neurology 2005;65(12):1976–1978. doi:10.1212/01.
relationship between sodium valproate dose and
wnl.0000188819.45330.90.
human teratogenicity: results of the Australian
20 Azar NJ, Wright AT, Wang L, et al. Generalized register of anti-epileptic drugs in pregnancy.
tonic-clonic seizures after acute oxcarbazepine J Clin Neurosci 2004;11(8):854–858. doi:10.1016/
withdrawal. Neurology 2008;70(22 pt 2): j.jocn.2004.05.003.
2187–2188. doi:10.1212/01.wnl.0000313152.
32 Christensen J, Grønborg TK, Sørensen MJ, et al.
89906.5a.
Prenatal valproate exposure and risk of autism
21 Nolan SJ, Muller M, Tudur Smith C, Marson AG. spectrum disorders and childhood autism.
Oxcarbazepine versus phenytoin monotherapy JAMA 2013;309(16):1696–1703. doi:10.1001/
for epilepsy. Cochrane Database Syst Rev jama.2013.2270.
2013;(5):CD003615. doi:10.1002/14651858.
33 Meador KJ, Baker GA, Browning N, et al. Fetal
CD003615.pub3.
antiepileptic drug exposure and cognitive
22 Koch MW, Polman SK. Oxcarbazepine versus outcomes at age 6 years (NEAD study): a
carbamazepine monotherapy for partial onset prospective observational study. Lancet
seizures. Cochrane Database Syst Rev 2009;(4): Neurol 2013;12(3):244–252. doi:10.1016/S1474-
CD006453. 10.1002/14651858.CD006453.pub2. 4422(12)70323-X.

ANTIEPILEPTIC DRUGS
34 Marson AG, Al-Kharusi AM, Alwaidh M, et al. 46 Hirsch LJ, Arif H, Nahm EA, Buchsbaum R,
The SANAD study of effectiveness of valproate, Resor SR Jr, Bazil CW. Cross-sensitivity of skin
lamotrigine, or topiramate for generalised and rashes with antiepileptic drug use. Neurology
unclassifiable epilepsy: an unblinded 2008;71(19):1527–1534. doi:10.1212/01.
randomised controlled trial. Lancet 2007; wnl.0000334295.50403.4c.
369(9566):1016–1026. doi:10.1016/S0140-
47 Brigo F, Ausserer H, Tezzon F, Nardone R. When
6736(07)60461-9.
one plus one makes three: the quest for rational
35 Glauser TA, Cnaan A, Shinnar S, et al. antiepileptic polytherapy with supraadditive
Ethosuximide, valproic acid, and lamotrigine in anticonvulsant efficacy. Epilepsy Behav 2013;
childhood absence epilepsy: initial monotherapy 27(3):439–442. doi:10.1016/j.yebeh.2013.03.010.
outcomes at 12 months. Epilepsia 2013;54(1):
48 Poolos NP, Warner LN, Humphreys SZ, Williams S.
141–155. doi:10.1111/epi.12028.
Comparative efficacy of combination drug
36 Mattson RH, Cramer JA, Collins JF. A comparison therapy in refractory epilepsy. Neurology 2012;
of valproate with carbamazepine for the 78(1):62–68. doi:10.1212/WNL.0b013e31823ed0dd.
treatment of complex partial seizures and
49 Piña-Garza JE, Schwarzman L, Wiegand F, Hulihan
secondarily generalized tonic-clonic seizures
J. A pilot study of topiramate in childhood
in adults. The Department of Veterans Affairs
absence epilepsy. Acta Neurol Scand 2011;123(1):
Epilepsy Cooperative Study No. 264 Group.
54–59. doi:10.1111/j.1600-0404.2010.01347.x.
N Engl J Med 1992;327(11):765–771. doi:10.1056/
NEJM199209103271104. 50 Kockelmann E, Elger CE, Helmstaedter C.
Significant improvement in frontal lobe
associated neuropsychological functions after
withdrawal of topiramate in epilepsy patients.
childhood absence epilepsy. N Engl J Med 2010;
Epilepsy Res 2003;54(2-3):171–178.
362(9):790–799. doi:10.1056/NEJMoa0902014.
51 Lee S, Sziklas V, Andermann F, et al. The effects
38 Riss J, Cloyd J, Gates J, Collins S. Benzodiazepines
of adjunctive topiramate on cognitive function
in epilepsy: pharmacology and pharmacokinetics.
in patients with epilepsy. Epilepsia 2003;44(3):
Acta Neurol Scand 2008;118(2):69–86. doi:10.1111/
339–347.
j.1600-0404.2008.01004.x.
52 Hunt S, Russell A, Smithson WH, et al. Topiramate
39 Gidal BE, DeCerce J, Bockbrader HN, et al.
in pregnancy: preliminary experience from the
Gabapentin bioavailability: effect of dose and
UK Epilepsy and Pregnancy Register. Neurology
frequency of administration in adult patients
2008;71(4):272–276. doi:10.1212/01.wnl.
with epilepsy. Epilepsy Res 1998;31(2):91–99.
0000318293.28278.33.
doi:10.1016/S0920-1211(98)00020-5.
53 Azar NJ, Bangalore-Vittal N, Arain A, Abou-Khalil
40 Perucca E, Gram L, Avanzini G, Dulac O.
BW. Tiagabine-induced stupor in patients with
Antiepileptic drugs as a cause of worsening
psychogenic nonepileptic seizures:
seizures. Epilepsia 1998;39(1):5–17. doi:10.1111/
nonconvulsive status epilepticus or
j.1528-1157.1998.tb01268.x.
encephalopathy? Epilepsy Behav 2013;27(2):
41 Kwan P, Brodie MJ, Kälviäinen R, et al. Efficacy 330–332. doi:10.1016/j.yebeh.2013.02.016.
and safety of pregabalin versus lamotrigine in
54 Koepp MJ, Edwards M, Collins J, et al. Status
patients with newly diagnosed partial seizures: a
epilepticus and tiagabine therapy revisited.
phase 3, double-blind, randomised, parallel-
Epilepsia 2005;46(10):1625–1632. doi:10.1111/
group trial. Lancet Neurol 2011;10(10):881–890.
j.1528-1167.2005.00263.x.
doi:10.1016/S1474-4422(11)70154-5.
55 Fukuyama K, Tanahashi S, Nakagawa M, et al.
42 French J, Kwan P, Fakhoury T, et al. Pregabalin
Levetiracetam inhibits neurotransmitter release
monotherapy in patients with partial-onset
associated with CICR. Neurosci Lett 2012;518(2):
seizures: a historical-controlled trial.
69–74. doi:10.1016/j.neulet.2012.03.056.
Neurology 2014;82(7):590–597. doi:10.1212/
WNL.0000000000000119. 56 Gujjar AR, Nandhagopal R, Jacob PC, et al.
Intravenous levetiracetam vs phenytoin for
43 Hirsch LJ, Weintraub D, Du Y, et al. Correlating
status epilepticus and cluster seizures: a
lamotrigine serum concentrations with
prospective, randomized study. Seizure 2017;
tolerability in patients with epilepsy. Neurology
49:8–12. doi:10.1016/j.seizure.2017.05.001.
2004;63(6):1022–1026. doi:10.1212/01.WNL.
0000138424.33979.0C. 57 Beuchat I, Novy J, Rossetti AO. Newer
antiepileptic drugs for status epilepticus in
44 Ramey P, Osborn M, Abou-Khalil B. Conversion
adults: what's the evidence? CNS Drugs 2018;
from immediate-release to extended-release
32(3):259–267. doi:10.1007/s40263-018-0509-5.
lamotrigine improves seizure control. Epilepsy
Res 2014;108(9):1637–1641. doi:10.1016/ 58 Brigo F, Bragazzi N, Nardone R, Trinka E. Direct
j.eplepsyres.2014.08.004. and indirect comparison meta-analysis of
levetiracetam versus phenytoin or valproate for
45 Ramey P, Osborn MR, Lowen KM, et al.
convulsive status epilepticus. Epilepsy Behav
Unexplained spikes in lamotrigine serum
2016;64(pt A):110–115. doi:10.1016/j.yebeh.
concentration: nonlinear elimination? Acta Neurol
2016.09.030.
Scand 2017;135(2):240–246. doi:10.1111/ane.12588.
534 APRIL 2019

59 Ben-Menachem E, Mameniškienė R, Quarato PP, 72 Dirani M, Nasreddine W, Abdulla F, Beydoun A.
et al. Efficacy and safety of brivaracetam for Seizure control and improvement of neurological
partial-onset seizures in 3 pooled clinical studies. dysfunction in Lafora disease with perampanel.
Neurology 2016;87(3):314–323. doi:10.1212/ Epilepsy Behav Case Rep 2014;2:164–166.
WNL.0000000000002864. doi:10.1016/j.ebcr.2014.09.003.
60 Data on File (FDA General Advice Letter, dated 73 Goldsmith D, Minassian BA. Efficacy and
09/13/2016). tolerability of perampanel in ten patients with
Lafora disease. Epilepsy Behav 2016;62:132–135.
61 Klein P, Schiemann J, Sperling MR, et al. A
doi:10.1016/j.yebeh.2016.06.041.
randomized, double-blind, placebo-controlled,
multicenter, parallel-group study to evaluate the 74 Schorlemmer K, Bauer S, Belke M, et al. Sustained
efficacy and safety of adjunctive brivaracetam in seizure remission on perampanel in progressive
adult patients with uncontrolled partial-onset myoclonic epilepsy (Lafora disease). Epilepsy
seizures. Epilepsia 2015;56(12):1890–1898. Behav Case Rep 2013;1:118–121. doi:10.1016/
doi:10.1111/epi.13212. j.ebcr.2013.07.003.
62 Yates SL, Fakhoury T, Liang W, et al. An 75 Bakas T, van Nieuwenhuijzen PS, Devenish SO,
open-label, prospective, exploratory study McGregor IS, Arnold JC, Chebib M. The direct
of patients with epilepsy switching from actions of cannabidiol and 2-arachidonoyl
levetiracetam to brivaracetam. Epilepsy Behav glycerol at GABAA receptors. Pharmacol Res
2015;52(pt A):165–168. doi:10.1016/j.yebeh.2015. 2017;119:358–370. doi:10.1016/j.phrs.2017.02.022.
09.005.
76 Gaston TE, Bebin EM, Cutter GR, Liu Y, Szaflarski
63 Sake JK, Hebert D, Isojärvi J, et al. A pooled JP. Interactions between cannabidiol and
analysis of lacosamide clinical trial data grouped commonly used antiepileptic drugs. Epilepsia
by mechanism of action of concomitant 2017;58(9):1586–1592. doi:10.1111/epi.13852.
antiepileptic drugs. CNS Drugs 2010;24(12):
77 Geffrey AL, Pollack SF, Bruno PL, Thiele EA.
1055–1068. doi:10.2165/11587550-000000000-
Drug-drug interaction between clobazam and
00000.
cannabidiol in children with refractory epilepsy.
64 Ben-Menachem E. Medical management of Epilepsia 2015;56(8):1246–1251. doi:10.1111/epi.13060.
refractory epilepsy–practical treatment with
78 Devinsky O, Cross JH, Laux L, et al. Trial of
novel antiepileptic drugs. Epilepsia 2014;
cannabidiol for drug-resistant seizures in the
55(suppl 1):3–8. doi:10.1111/epi.12494.
Dravet syndrome. N Engl J Med 2017;376(21):
65 Kälviäinen R, Nousiainen I, Mäntyjärvi M, et al. 2011–2020. doi:10.1056/NEJMoa1611618.
Vigabatrin, a gabaergic antiepileptic drug, causes
79 Devinsky O, Patel AD, Cross JH, et al. Effect
concentric visual field defects. Neurology 1999;
of cannabidiol on drop seizures in the
53(5):922–926. doi:10.1212/WNL.53.5.922.
Lennox-Gastaut syndrome. N Engl J Med 2018;
66 Toggweiler S, Wieser HG. Concentric visual field 378(20):1888–1897. doi:10.1056/NEJMoa1714631.
restriction under vigabatrin therapy: extent
80 Thiele EA, Marsh ED, French JA, et al. Cannabidiol
depends on the duration of drug intake. Seizure
in patients with seizures associated with
2001;10(6):420–423. doi:10.1053/seiz.2000.0527.
Lennox-Gastaut syndrome (GWPCARE4): a
67 Patsalos PN. The clinical pharmacology profile of randomised, double-blind, placebo-controlled
the new antiepileptic drug perampanel: a novel phase 3 trial. Lancet 2018;391(10125):1085–1096.
noncompetitive AMPA receptor antagonist. doi:10.1016/S0140-6736(18)30136-3.
Epilepsia 2015;56(1):12–27. doi:10.1111/epi.12865.
81 Devinsky O, Marsh E, Friedman D, et al.
68 French JA, Krauss GL, Wechsler RT, et al. Cannabidiol in patients with treatment-resistant
Adjunctive perampanel for the treatment of epilepsy: an open-label interventional trial.
drug-resistant primary generalized tonic-clonic Lancet Neurol 2016;15(3):270–278. doi:10.1016/
(PGTC) seizures in patients with idiopathic S1474-4422(15)00379-8.
generalized epilepsy (IGE): a double-blind,
82 Devinsky O, Verducci C, Thiele EA, et al.
randomized, placebo-controlled phase III trial.
Open-label use of highly purified CBD (Epidiolex)
Neurology 2015;84(14 suppl):S31.007.
in patients with CDKL5 deficiency disorder and
69 Ettinger AB, LoPresti A, Yang H, et al. Psychiatric Aicardi, Dup15q, and Doose syndromes. Epilepsy
and behavioral adverse events in randomized Behav 2018;86:131–137. doi:10.1016/j.yebeh.2018.
clinical studies of the noncompetitive AMPA 05.013.
receptor antagonist perampanel. Epilepsia 2015;
83 Szaflarski JP, Bebin EM, Comi AM, et al.
56(8):1252–1263. doi:10.1111/epi.13054.
Long-term safety and treatment effects of
70 Andres E, Kerling F, Hamer H, et al. Behavioural cannabidiol in children and adults with
changes in patients with intellectual disability treatment-resistant epilepsies: expanded
treated with perampanel. Acta Neurol Scand access program results. Epilepsia 2018;59(8):
2017;136(6):645–653. doi:10.1111/ane.12781. 1540–1548. doi:10.1111/epi.14477.
71 Crespel A, Gelisse P, Tang NP, Genton P. 84 Semah F, Thomas P, Coulbaut S, Derambure P.
Perampanel in 12 patients with Unverricht- Early add-on treatment vs alternative monotherapy
Lundborg disease. Epilepsia 2017;58(4):543–547. in patients with partial epilepsy. Epileptic Disord
doi:10.1111/epi.13662. 2014;16(2):165–174. doi:10.1684/epd.2014.0650.

ANTIEPILEPTIC DRUGS
85 Zaccara G, Perucca E. Interactions between 88 Kanner AM, Ashman E, Gloss D, et al. Practice
antiepileptic drugs, and between antiepileptic guideline update summary: efficacy and
drugs and other drugs. Epileptic Disord 2014; tolerability of the new antiepileptic drugs II:
16(4):409–431. doi:10.1684/epd.2014.0714. treatment-resistant epilepsy: report of the
Guideline Development, Dissemination, and
86 Margolis JM, Chu BC, Wang ZJ, et al.
Implementation Subcommittee of the American
Effectiveness of antiepileptic drug combination
Academy of Neurology and the American
therapy for partial-onset seizures based on
mechanisms of action. JAMA Neurol 2014;71(8):
doi:10.1212/WNL.0000000000005756.
985–993. doi:10.1001/jamaneurol.2014.808.
89 Kanner AM, Ashman E, Gloss D, et al. Practice
87 Stafstrom CE. Mechanisms of action of
guideline update summary: efficacy and
antiepileptic drugs: the search for synergy. Curr
tolerability of the new antiepileptic drugs I:
Opin Neurol 2010;23(2):157–163. doi:10.1097/
treatment of new-onset epilepsy: report of the
WCO.0b013e32833735b5.
Guideline Development, Dissemination, and
Implementation Subcommittee of the American
Academy of Neurology and the American
doi:10.1212/WNL.0000000000005755.
536 APRIL 2019

Driving and Epilepsy: MEDICOLEGAL ISSUES

Ethical, Legal, and Health C O N T I N U UM A U D I O
ONLINE
Care Policy Challenges

By Joseph S. Kass, MD, JD, FAAN; Rachel V. Rose, JD, MBA
ABSTRACT
Although the principle of autonomy allows patients to refuse interventions
their physicians recommend, patients are not free to ignore legally
mandated restrictions on driving, and physicians are ethically justified in
constraining their patients’ driving rights in compliance with state law.
Furthermore, the standard of care for treatment of patients with epilepsy
includes counseling about lifestyle modifications that promote patient
safety and compliance with the law. Neurologists should not only counsel
their patients with epilepsy about legally mandated driving restrictions but
also document this counseling in the medical record. Failure to counsel
and to document may result in legal liability if patients experience seizures
while driving and injure either themselves or third parties. The
neurologist’s duty of care may be limited to the patient in some CITE AS:
jurisdictions but may be extended to injured third parties in others. CONTINUUM (MINNEAP MINN)
2019;25(2, EPILEPSY):537–542.
Furthermore, a patient’s own contributory negligence may limit or
completely foreclose recovery against the physician to varying degrees, Address correspondence to
depending on the state in which the injury occurred. Dr Joseph S. Kass, Baylor College
of Medicine, One Baylor Plaza
M-210, Houston, TX 77030,
kass@bcm.edu.
CASE
Note: This is a hypothetical case. Dr Kass serves as associate
A 27-year-old man diagnosed with juvenile myoclonic epilepsy at editor of medicolegal issues for
Continuum, as an associate
age 15 was seizure free on valproic acid until he began his work as a
editor for Continuum Audio, as a
management consultant, which required many late nights of work and neurology section editor of
frequent international travel. Because of his work schedule and frequent Ferri’s Clinical Advisor for
Elsevier, and as co-editor of
crossing of time zones, the man occasionally missed doses of valproic Neurology Secrets, Sixth Edition.
acid. He therefore experienced breakthrough generalized tonic-clonic Dr Kass has received personal
seizures. However, he commuted to his office by car and did not wish to compensation for CME lectures
from Pri-Med LLC. Ms Rose
stop driving because of the inconvenience of having to rely on taxis and serves on the editorial board of
ride-sharing services. At his last visit with his neurologist, the patient’s BC Advantage and receives book
valproic acid level was found to be subtherapeutic, and he admitted to royalties from the American Bar
Association.
occasionally missing doses of his medication and to having experienced
a breakthrough seizure 2 months earlier. The neurologist counseled the UNLABELED USE OF
patient about the importance of medication compliance and lifestyle USE DISCLOSURE:
modification but either forgot to discuss driving restrictions or discussed Dr Kass and Ms Rose report no
restrictions but failed to document the details of the counseling in the disclosures.
electronic health record. Two weeks after the appointment, the patient © 2019 American Academy
experienced a generalized tonic-clonic seizure while driving to work, striking of Neurology.

DRIVING AND EPILEPSY
an oncoming vehicle, seriously injuring both himself and the other driver, and
completely wrecking both cars.
DISCUSSION
I
n all but a few cities that have outstanding public transportation systems,
living an independent, socially and financially productive life in the
United States depends heavily on an adult’s ability to drive. Although most
adults take the right to drive for granted, people with epilepsy face
considerable driving restrictions, with each state establishing its own laws.
Most states use an inflexible interval of seizure freedom before legal driving
privileges can be reinstituted, ranging from 3 to 12 months, but several states do
allow individualization of restrictions based on features of a driver’s epilepsy
that mitigate the risk of seizures with driving. Whereas the majority of states do
not compel medical providers to report drivers who experience seizures to the
state department of motor vehicles, six states do impose such a restriction. In this
issue of Continuum, the article “Counseling and Management of the Risks of
Living With Epilepsy” by Katherine Noe, MD, PhD, FAAN,1 reviews in detail the
impact driving restrictions have on people with epilepsy, including data about
the impact of seizures on the risk of motor vehicle collisions and the impact of
shortening the duration of driving restrictions to 3 months of seizure freedom.1
Noe’s article concludes with data indicating that many, if not most, adults with
epilepsy do not recall being counseled about driving restrictions and then
explains that “[p]roviders may be challenged to balance advocating for the needs
of an individual patient with promoting the safety of the individual and the
public. . .”1 This call for improved counseling and the important insight about
providers experiencing conflict in their dual roles as advocates for the individual
patient and the public interest are two starting points for this article on the
ethical, legal, and health care policy challenges of driving and epilepsy.
This article first explores the ethical issues arising from the physician’s role in
managing driving restrictions in patients with epilepsy, and then, using the case
above, it explores the legal liability physicians may face for the actions of their
patients who cause motor vehicle collisions because they experienced a seizure
while driving. This article then discusses ways physicians can discharge their
ethical duty to their patients and simultaneously mitigate legal risk through
adequate patient counseling and documentation of that counseling in the
medical record.
Ethical Conflicts Arising From Driving Restrictions

“It is the responsibility of the state, not the physician, to determine who should or
should not drive an automobile.”2 However, state authorities often rely on
physicians’ medical assessments of patients in rendering their decisions, and
physicians have a legal and ethical obligation to comply with the law. When
recommending that a patient who recently experienced a seizure cease driving in
compliance with state law, neurologists may feel conflicted as they shift their
focus of concern from patient to public. Typically, physicians consider the
impact of their clinical recommendations on the patient they are treating and
leave explicit societal considerations out of the examination room. Furthermore,
in the United States, medical practice promotes patient autonomy through the
informed consent process and, therefore, does not restrict adults with
538 APRIL 2019

decision-making capacity from refusing medical interventions, even when a
medical provider believes that the informed refusal is potentially harmful to the
patient. Therefore, a shift from a patient-centric, autonomy-promoting approach
to a public-centric, communitarian approach changes the power balance and
risks threatening the therapeutic alliance, an uneasy situation for physician and
patient alike. Although both parties understand that uncontrolled seizures are
dangerous not only to others on the road but also to the patient and the patient’s
passengers, driving restrictions still impose serious, uncompensated
socioeconomic burdens on those prevented from driving. Driving restrictions,
however, do align with the principles of beneficence and nonmaleficence,
requiring providers to maximize patient benefit while minimizing patient harm.
Preventing serious bodily harm and economic damage fulfills both ethical
imperatives. However, patients and physicians alike understand that the costs of
driving restrictions represent a tremendous, very tangible harm to the patient,
weighed (from the patient’s perspective) against the merely theoretical risk of
experiencing a seizure while driving. Therefore, physicians need to muster the
virtues of courage, honesty, and even self-sacrifice (because of the risk of
angering and even losing a patient) to discuss restrictions on driving, even more
so in states that mandate physician reporting to state authorities.
Counseling and Documenting: Ethical Obligations and Legal Risk Mitigation

Because neurologic diseases render patients particularly vulnerable to harm,
lifestyle counseling and anticipatory guidance are critical components of the
neurologist’s therapeutic armamentarium. Therefore, the standard of care for the
treatment of a patient with epilepsy involves not merely prescribing medication
but also counseling the patient about lifestyle choices that may reduce the risk of
recurrent seizures (eg, obtaining adequate sleep) and of harm during a seizure
(eg, not swimming alone). Counseling about driving restrictions, including the
requirement that the patient self-report to the state authority regulating driving,
falls into the latter category because it reduces the risk of harm and promotes
compliance with the law. Thus, failing to counsel patients appropriately about
driving restrictions may open a clinician up to legal liability.
CASE CONTINUED
In the case presented, the victim sued both the patient and the neurologist
for negligence.
DISCUSSION CONTINUED
The victim’s suit against the patient is rather straightforward. “Under the
common law, a cause of action for negligence has three elements: (1) a legal duty;
(2) a breach of that duty; and (3) damages proximately resulting from the
breach.”3 The reasonable-person standard is generally used to judge whether the
defendant breached a legal duty. The patient has a duty to others sharing the
roads to drive safely and in compliance with the law. The patient in this case
breached that duty when he chose to operate his automobile under conditions
that were both illegal (given the recent seizure) and of particularly high risk of
breakthrough seizures (history of inconsistent medication adherence and poor
sleep hygiene). “But for” the patient’s negligence, the victim would not have

suffered physical, psychological, and financial injuries, making the patient’s

negligent action both the direct and the “proximate” cause of the victim’s injuries.
The less predictable part of the victim’s lawsuit relates to the allegation of
negligence against the neurologist. According to the victim, had the neurologist
warned the patient not to drive, the motor vehicle collision may not have
occurred. Therefore, the victim alleges that the neurologist owed a duty to third
parties who are not her patients to mitigate the risk of her patient’s unlawful
driving, and she acted negligently by not counseling the patient about the legal
restrictions on driving and the safety risks of driving under the extant conditions.
The threshold legal question in the victim’s lawsuit against the neurologist is
whether the neurologist owes the victim, a complete stranger, a duty of care. The
answer to this question is not uniform among the states. In no jurisdiction would
the neurologist be found to have acted negligently vis-à-vis the victim on the
basis of malpractice. Although an allegation of medical malpractice is a type of
negligence case, the victim’s lawsuit against the neurologist does not meet the
requirements for a malpractice case in any jurisdiction because of the lack of a
physician-patient relationship between the neurologist and the victim.
Some courts have decided that a physician does not owe “a duty to third
parties to warn an epileptic patient not to drive or to report the patient’s
condition to state authorities that govern the issuance of drivers’ licenses.”4
For example, in Praesel v Johnson,4 the Texas Supreme Court found that
“[b]alancing both the need for and the effectiveness of a warning to a patient
who already knows that he or she suffers from seizures against the burden of
liability to third parties … the benefit of warning an epileptic not to drive is
incremental [because a doctor’s warning may have little effect on the patient’s
behavior] but that the consequences of imposing a duty are great.”4 The Court
found that drivers carry the responsibility for safely operating their vehicles and
declined to impose on physicians a duty to third parties to warn a patient with
epilepsy not to drive.
Other courts have reached quite a different conclusion. Cases like the
California Supreme Court decision Tarasoff v Regents of University of California5
influenced a number of jurisdictions to impose duties of care beyond the
patient-physician dyad. Tarasoff imposed on mental health providers a duty to
protect from harm a patient’s intended victim. Cases such as Duvall v Goldin,6 in
which the Michigan Supreme Court, influenced by Tarasoff among other cases,
found that a physician’s failure to properly treat a patient with epilepsy with
antiepileptic drugs resulted in a breach of a duty of care to third-party drivers,
expanded the duty of care beyond a specific intended victim to all drivers at
large. Although recognizing that at common law, “no one has a duty to protect an
individual who is endangered by the conduct of another,” the Duvall Court
concluded that “[w]here the actor stands in a special relationship with either the
third-party victim or the person causing the injury, a duty of reasonable care
may arise.”6
Having found that the physician, by virtue of his special relationship with
the patient, did have a duty to third-party drivers to treat the patient’s seizures
appropriately, the Duvall Court next considered whether it was foreseeable that
the patient’s conduct would harm the victim. The Duvall Court cited Gooden v
Tips,7 a case in which a physician failed to warn a patient about the sedating
nature of a medication, resulting in the patient crashing and injuring another
party. The foreseeability of this consequence led the Gooden Court to conclude
540 APRIL 2019

that “under proper facts, a physician can owe a duty to use reasonable care to
protect the driving public where the physician’s negligence in diagnosis or
treatment of his patient contributes to plaintiff’s injuries.”7 The Duvall Court
concurred, finding that “it is foreseeable that a doctor’s failure to diagnose or
properly treat an epileptic condition may create a risk of harm to a third party.”6
Thus, depending on the jurisdiction in which the injury occurs, the victim’s
negligence case against the neurologist would be either dismissed or allowed
to proceed.
CASE CONTINUED
Finding himself the defendant in a negligence lawsuit, the patient sued the
neurologist for malpractice, alleging that the neurologist’s treatment did
not meet the standard of care because the neurologist failed to warn the
patient not to drive until he was seizure free for at least 3 months, the legal
requirement in the state in which the parties of the case reside.
DISCUSSION CONTINUED
To prevail in a malpractice lawsuit against the neurologist, the patient will have
to prove that a physician-patient relationship existed between patient and
neurologist and that the neurologist breached her duty to provide care at the
nationally accepted standard for neurologists, resulting in harm to the patient.
The outcome of the lawsuit will hinge on the jury’s decision about two issues:
(1) the neurologist failed in her duty to her patient by failing to warn the patient
not to drive, and (2) the neurologist’s breach of the standard of care harmed the
patient, resulting in damages. After hearing from expert witnesses in neurology
on both sides of the case, the jury may find that the lack of documentation
about driving restrictions supports the allegation that the neurologist did breach
the standard of care. The neurologist may claim that, regardless of her
documentation in this case, her standard practice is always to counsel about
driving restrictions. Yet the jury may find this argument unconvincing and
indicative of sloppy practice. However, even if it does conclude that the
physician breached the standard of care, the jury may also find that the patient
contributed to his own injury through his personal negligence. Depending on
how the state assigns damages, a finding that the patient’s behavior was also
negligent may have a profound impact on how much in damages the patient can
recover from the neurologist.
Every US jurisdiction uses one of four basic systems for allocating fault and
damages: (1) pure contributory negligence, (2) pure comparative fault,
(3) modified comparative fault, or (4) slight or gross negligence comparative
fault.8 Four states and the District of Columbia (the District of Columbia uses a
different standard for pedestrians and bicyclists) operate under a pure
contributory negligence regime, in which plaintiffs cannot recover if they are
even 1% at fault. In the 12 states with a pure comparative fault regime, fault is
apportioned proportionately, and the plaintiff’s damages are reduced by the
percentage of fault assigned to the plaintiff. In a modified comparative fault
jurisdiction, “each party is held responsible for damages in proportion to their
own percentage of fault, unless the plaintiff’s negligence reaches a certain
designated percentage.”8 In 10 states, plaintiffs cannot recover any damages if

they are 50% or more at fault, whereas in 23 states plaintiffs cannot recover any
damages if they are 51% or more at fault. Finally, only South Dakota uses the
slight or gross negligence comparative fault rule, under which a “plaintiff is
barred from any recovery for anything other than slight negligence.”8
Mitigation of Risk
How could the neurologist have avoided both lawsuits in the first place? The
best way for neurologists treating patients with epilepsy to mitigate risk is to
standardize both counseling about driving as well as documentation of this
counseling. In many electronic health records, providers can create standardized
templates to document their discussion with the patient. Standardized
documentation ensures that the neurologist has communicated all relevant
points and documents the details of the conversation relatively easily. In addition
to reviewing the state’s driving laws and reporting requirements, the neurologist
should remind the patient about the physical, financial, and legal (both civil
and criminal) risks of driving when state law forbids driving. Patients should also
be encouraged to follow the law by self-reporting and should be informed if the
physician will be reporting them to the state licensing authority. Even states
without mandatory physician reporting generally allow reporting but only at the
physician’s discretion. Physicians are ethically justified to report patients who
insist on driving illegally because they pose a heightened risk to themselves
and others.
Each state has its own process for self-reporting, and many states rely on the
patient’s own physician’s assessment to determine if a patient is safe to drive.
Other states use independent medical advisors. Some states offer statutory
immunity from lawsuits to physicians who render opinions about driver safety in
good faith, whereas others do not offer such protections. Physicians are
encouraged to search the Epilepsy Society’s State Driving Law Database9 for
more information about the relevant driving laws and reporting procedures in
every state and the District of Columbia.
REFERENCES
1 Noe K. Counseling and management of the 6 Duvall v Goldin, 139 Mich App 342 (1984) 362
risks of living with epilepsy. Continuum 2019; NW2d 275.
25(2, Epilepsy):477–491. doi:10.1212/
7 Gooden v Tips, 651 SW2d 364 (Tex App, 1983).
CON.0000000000000708.
8 Contributory negligence/comparative fault laws
2 Krumholz A. Carpe diem (seize the day). Agency
in all 50 states. Matthiesen, Wickert & Lehrer, S.C.
for Healthcare Research and Quality Patient
website. www.mwl-law.com/wp-content/
Safety Network. psnet.ahrq.gov/webmm/case/
uploads/2013/03/contributory-negligence-
85/Carpe-Diem-Seize-the-Day. Updated
comparative-fault-laws-in-all-50-states.pdf.
December 2004. Accessed January 31, 2019.
Accessed January 31, 2019.
3 Greater Houston Transp. Co v Phillips, 801 SW2s
9 Epilepsy Foundation. State driving laws
523 (1990).
database. www.epilepsy.com/driving-laws/
4 Praesel v Johnson 967 SW2d 391 (1998). 2008691. Accessed January 31, 2019.
5 Tarasoff v Regents of University of California, v17
Cal 3d 425; 131 Cal Rptr 14; 551 P2d 334 (1976).
542 APRIL 2019

Tips and Resources for PRACTICE ISSUES

Medication Reconciliation C O N T I N U UM A U D I O
ONLINE
By Anant M. Shenoy, MD, FAAN; Amy Bennett, JD; Alan Z. Segal, MD, FAAN
ABSTRACT
Medication errors occur despite best intentions and are often the result of
medication discrepancies. Medication reconciliation reduces the
likelihood of errors by addressing medication discrepancies that result
from multiple points of care, transitions in care, or patient report.
Providers and practices may feel overwhelmed by new record systems and
regulatory requirements, but multiple resources are available to assist
providers to perform medication reconciliation with their patients.
Providers and practices should implement medication reconciliation
strategies, such as adoption of a multidisciplinary approach, engagement
of patients to track medications, and identification of patients who are at
high risk for medication list discrepancies and errors. Medication
reconciliation will ultimately improve quality of care.
CASE
A 72-year-old woman with multiple medical problems including diabetes
mellitus, hypertension, dementia, and a 20-year history of epilepsy
presented to the emergency department with a breakthrough seizure, and
a neurologic consultation was called. Her seizures had previously been
well-controlled on a combination of topiramate and levetiracetam. She
had been discharged from the hospital 1 week previously after a long
hospitalization for an abdominal abscess requiring surgery. During the CITE AS:
admission, her topiramate had been held because she was not taking CONTINUUM (MINNEAP MINN)
2019;25(2, EPILEPSY):543–549.
anything by mouth.
During the neurologic consultation, the medication list from her recent
discharge was reviewed. Although it showed 15 different medications, it Dr Anant Shenoy, 300 Mount
did not list her topiramate, only her levetiracetam. Unfortunately, the Auburn St, Ste 316, Cambridge,
MA 02138; anant.shenoy@mah.
topiramate had not been restarted during the previous admission or at the harvard.edu.
time of discharge home.
Drs Shenoy and Segal and Ms
Bennett report no disclosures.
DISCUSSION
UNLABELED USE OF
A
medication error is any error that occurs in the medication-use PRODUCTS/INVESTIGATIONAL
process1 and can be either an act of commission or an act of USE DISCLOSURE:
Drs Shenoy and Segal and
omission. Medication errors include giving the wrong medication Ms Bennett report no
or dose, administering a drug to the wrong patient, and not disclosures.
prescribing a medication that was indicated.2 The Institute of
Medicine estimates that medication errors are associated with 1 of 131 outpatient © 2019 American Academy
and 1 of 854 inpatient deaths.3 of Neurology.

MEDICATION RECONCILIATION
One cause of medication error is medication list discrepancies, which are

differences in a patient’s documented medication regimen across different sites
of care.4,5 These discrepancies can lead patients to receive incorrect doses of
medicines or to not receive correct medications. Medication discrepancies are
common and can lead to adverse events, including prolonged hospital stays,
emergency department visits, and hospital readmissions.4,6 Patients are
especially vulnerable to medication list discrepancies during transitions between
sites of care, for example, when a patient is transferred from a hospital to a
skilled nursing facility. More than half of patients have at least one unintended
medication discrepancy at hospital admission, demonstrating that medication
errors are common and that better methods of ensuring an accurate medication
history at admission are needed.7 At discharge, medication regimens are
frequently confusing, with the inclusion of temporary hospital medications
(eg, stool softeners or gastric ulcer prophylaxis). Changes in patients’ medications
because of hospital formulary restrictions may further complicate medication
reconciliation, and crucial new medications necessary upon hospital discharge
are often missed. Better strategies for transitions of care are needed, such as
postdischarge follow-up telephone calls and patient education to empower
maintenance of accurate medication information that may reduce this risk.
Another pitfall of medication error is failing to identify potential drug-drug
interactions. Drug-drug interactions can lead to morbidity and mortality for a
patient and are often preventable.8 As medicine is becoming more and more
specialized, it is becoming harder to keep up with the growing list of medications
and potential interactions. The number of medications used within neurology has
increased. This complexity increases the risk of potential adverse interactions.
Electronic health records flag potential adverse medication interactions, but
messages about clinically important effects have the potential of being obscured
in the context of multiple less relevant messages being relayed to the clinician.
Caring for the pediatric populations adds another layer of complexity to
medication concerns. Most pediatric medications are weight based, many must
be mixed by pharmacists or others at the time of use, and some are available in
multiple formulations. Children are also less able to identify and report an error
or adverse event.9
One strategy to reduce medication list discrepancies and minimize drug-drug
interactions is medication reconciliation. The Institute for Healthcare
Improvement defines medication reconciliation as the “process of identifying the
most accurate list of all medications a patient is taking—including name, dosage,
frequency, and route—and using this list to provide correct medications for
patients anywhere within the health care system.”10 The goal of medication
reconciliation is for a patient to receive all the correct medications at their
intended doses in various care settings, thereby preventing medication errors.
Providers should reconcile at each care transition, reconcile admission orders
with home medication lists, reconcile medication orders when patients are
transferred to other levels of care, and reconcile discharge instructions and
prescriptions with the medications received during the patient’s stay. Having the
correct medication list also allows for computerized checking for drug-drug
interactions and allergies. Use of electronic health records is crucial to ensure
accuracy across different sites of care. This is particularly important with the
advent of electronic prescribing and regulations to track use of opioids and other
controlled substances.
544 APRIL 2019

Medication reconciliation has received much support from national and
international organizations. Since 2005, the Joint Commission has required US
hospitals to perform medication reconciliation at admission, during transfers,
and at discharge to help prevent “medication errors such as omissions,
duplications, dosing errors, or drug interactions.”11 In addition, the World Health
Organization has also encouraged medication reconciliation at transitions of care
as part of its High 5s global patient safety initiative.12 The Centers for Medicare &
Medicaid Services (CMS) previously required medication reconciliation for
Meaningful Use, and medication reconciliation activities continue to be
highlighted in the CMS Quality Payment Program (QPP) Merit-based Incentive
Payment System (MIPS).13,14 For 2019, eligible clinicians are able to use
medication reconciliation activities to meet a portion of MIPS’ Quality and
Improvement Activities requirements.13,14
Despite this international attention and that logic dictates that medication
reconciliation would be beneficial, evidence for the benefit of medication
reconciliation on patient outcomes has not been robust.1,4,15 Multiple studies have
shown that medication reconciliation helps decrease medication discrepancies.4
However, these studies have not consistently shown beneficial effect on adverse
patient outcomes and health care utilization.4,13 Some of this is because of the
logistic limitation of designing such studies.1,4 Future clinical trials encompassing
various clinical venues will be necessary to assess the true effect of medication
reconciliation on patient outcomes.
Medication reconciliation involves multiple challenges at various levels, but
strategies and resources exist to help neurologists overcome these challenges
(PRACTICE TABLE 1 and PRACTICE TABLE 2).16 The challenges start at the health
care provider level, where there can be a knowledge gap for health care
professionals who may not appreciate the dangers of medication list
discrepancies. For some providers, medication reconciliation is viewed as a
regulatory requirement. Providers must be convinced that medication
reconciliation is more than a regulatory obligation and see it as a benefit to
patients. This way of thinking encourages providers to perform medication
reconciliation with accuracy and to assign the appropriate level of priority to
medication reconciliation with their busy schedules and multiple time
constraints.20
Time constraints are an impediment to providers in all health care settings,
particularly smaller practices. In some cases, a team approach to medication
Strategies to Achieve Medication Reconciliation PRACTICE TABLE 1
◆ Discuss the technical process of performing, documenting, and reporting medication

reconciliation with the electronic health records vendor.
◆ Adopt a team-based approach in which individual staff members have clearly defined roles
in performing medication reconciliation.
◆ Engage patients to keep track of their own medications and bring them to all appointments.
Office staff can remind patients of this when they communicate appointment times.
Electronic resources, such as the patient portal, might help facilitate this.
◆ Identify patients at high risk based on health literacy, cognitive impairment, polypharmacy,
and multiple care sites.

reconciliation might alleviate some of this burden. Team members such as

medical assistants, nurses, and pharmacists can help verify a patient’s medication
history.4,16 Similar to providers having a colleague double-check identity and
procedure location before starting a procedure, providers could ask a team
member to confirm that a patient’s medications have been recorded accurately.
Multiple resources are available for providers, including those in solo and small
practices, to identify stops within the patient care process to confirm
medication accuracy.
The Agency for Healthcare Research and Quality has provided clinicians and
systems with a Medications at Transitions and Clinical Handoffs (MATCH)
Toolkit for Medication Reconciliation to help build a strong medication
reconciliation foundation and to pilot changes in practice.17 This toolkit guides
physicians to review and improve on their current practices of medication
reconciliation. It also encourages the standardization of various roles among team
members in medication reconciliation. For example, in some ambulatory
practices, a medical assistant might take the patient’s medication history, with
the provider performing the final medication reconciliation approval afterward.
The Institute for Healthcare Improvement has resources for providers
considering new medication reconciliation practices, including sample templates
to track provider medication reconciliation compliance, help avoid additional
costs, and find errors.9 In addition, the Pediatric Affinity Group’s “How-to
Guide: Prevent Adverse Drug Events Pediatric Supplement” is available on the
Institute for Healthcare Improvement website.9,21
Unreliable medication history is another direct obstacle to medication
reconciliation.16,20 Patients should be empowered to keep an accurate list of
medications to bring to medical encounters. Programs such as Partnership for
Patients through CMS have started to engage patients in becoming involved in
their health care, including transitions of care.22 Patients might also be
encouraged to track their medication list through electronic tools such as patient
portals.23 Practices may want to provide resources to engage patients and their
families in tracking their medications, such as the National Quality Forum
PRACTICE TABLE 2 Medication Reconciliation Resources
Medication
Overview of How to Reconciliation Flow Strategies to Sample Policies
Medication Engage Diagrams for Patient Engage Team and/or
Resource Reconciliation Patients Visits Members Documents
Medications at Yes Yes Yes Yes Yes

Transitions and Clinical
Handoffs Toolkit17
Institute for Healthcare Yes Yes Yes No Yes

Improvement9
National Quality Forum No Yes No No No

Patient Passport18
You: Your Own Best No Yes No No No

Medicine19
546 APRIL 2019

Patient Passport or the Minnesota Alliance for Patient Safety’s You: Your Own
Best Medicine.18,19
The MATCH toolkit also encourages providers to be especially cautious with
certain patients who are at high risk of medication list discrepancies based on
their health literacy, cognitive impairment, and multiple care sites.17 Patients
with polypharmacy should also be considered at risk for medication errors.1
Polypharmacy is compounded by confusion related to standing medications
versus as-needed medications and generic versus brand names. In addition,
many neurologic medications are now available in both immediate-release and
long-acting formulations, which adds to patient confusion. Clinical summaries
that detail medications, dosages, and schedules should be given at the end of
ambulatory and inpatient encounters to allow patients to have their most
up-to-date medication list on hand.
The US health care system can be disjointed, which makes medication
reconciliation even more challenging.24 Patients often receive medical care and
medications from multiple care settings and pharmacies. Furthermore, electronic
health records systems may silo patient information between certain hospitals
and practices. These entities will need to do a better job of collaborating in the
best interest of the patient.15,24 The health information exchange aims to better
connect various medical information systems and is an important resource for
medication reconciliation.25 The health information exchange allows for the
communication and transmission of electronic information between various
electronic health records systems.
In the case above, implementation of medication reconciliation best practices
could have prevented the patient’s breakthrough seizure. A medication
reconciliation of discharge medications with admission medications would have
led to the patient restarting her topiramate. The patient was at high risk for
medication list discrepancies given her cognitive problems, polypharmacy, and
transition of care sites. All neurologists should encourage their patients or the
patients’ caregivers to maintain an accurate list of their medications in case of
medical emergencies. Upon presentation for care, neurologists should review any
medication list provided by the patient; confirm accuracy and adherence through
secondary sources, such as family or pharmacy; review new medications
prescribed since time of arrival for accuracy and potential complications; and
educate patients and caregivers about any changes to medications made during
the course of treatment. For patients at high risk for medication list discrepancies
and errors, a follow-up telephone call upon discharge from inpatient care
services could help confirm medication accuracy with discharge care providers.
By ensuring existing medications and any new medications prescribed are
recorded accurately, a lower likelihood of error and greater chance of patient
adherence and improvement exist. Most often, best practice will require review
with multidisciplinary team members to confirm that accurate information is
represented in the electronic health record. Every team member is responsible
for medication reconciliation when caring for a patient.
CONCLUSION
Collaboration between all stakeholders will be crucial to ensure medication
reconciliation. Patients should be encouraged to maintain medication lists at

all times and keep them on their person. Together, physicians, hospitals,
and pharmacies can facilitate this process, ultimately improving the quality
of care.
REFERENCES
1 Wittich CM, Burkle CM, Lanier WL. Medication 12 World Health Organization. Action on patient
errors: an overview for clinicians. Mayo Clin Proc safety – high 5s. who.int/patientsafety/
2014;89(8)1116–1125. doi:10.1016/j.mayocp. implementation/solutions/high5s/ps_high5s_
2014.05.007. project_overview_fs_2010_en.pdf. Accessed
January 31, 2019.
2 Committee on Data Standards for Patient Safety,
Board on Health Care Services, Institute of 13 Centers for Medicare and Medicaid Services.
Medicine of the National Academies; Aspden P, Stage 2 eligible professional meaningful use core
Corrigan JM, Wolcott J, Erickson SM, eds. Patient measures: measure 14 of 17. cms.gov/
safety: achieving a new standard for care. Regulations-and-Guidance/Legislation/
Washington, DC: The National Academies Press, EHRIncentivePrograms/downloads/Stage2_
2004. EPCore_14_MedicationReconciliation.pdf.
3 Institute of Medicine Committee on Quality of Published October 2012. Accessed January 31,
Health Care in America; Kohn LT, Corrigan J, 2019.
Donaldson MS, eds. To err is human: building a 14 Centers for Medicare and Medicaid Services.
safer health system. Washington, DC: The Medicare Program; Revisions to Payment
National Academies Press, 2000. Policies Under the Physician Fee Schedule and
4 Mueller SK, Sponsler KC, Kripalani S, Schnipper Other Revisions to Part B for CY 2019; Medicare
JL. Hospital-based medication reconciliation Shared Savings Program Requirements; Quality
practices: a systematic review. Arch Intern Payment Program; Medicaid Promoting
Med 2012;172(14):1057–1069. doi:10.1001/ Interoperability Program; Quality Payment
archinternmed.2012.2246. Program-Extreme and Uncontrollable
Circumstance Policy for the 2019 MIPS Payment
5 Coleman EA, Smith JD, Raha D, Min SJ. Posthospital Year; Provisions From the Medicare Shared
medication discrepancies: prevalence and Savings Program-Accountable Care
contributing factors. Arch Intern Med 2005;165(16): Organizations-Pathways to Success; and
1842–1847. doi:10.1001/archinte.165.16.1842. Expanding the Use of Telehealth Services for the
6 Forster AJ, Murff HJ, Peterson JF, Gandhi TK, Bates Treatment of Opioid Use Disorder Under the
DW. Adverse drug events occurring following Substance Use-Disorder Prevention That
hospital discharge. J Gen Intern Med 2005;20(4): Promotes Opioid Recovery and Treatment
317–323. doi:10.1111/j.1525-1497.2005.30390.x. (SUPPORT) for Patients and Communities Act.
federalregister.gov/documents/2018/11/23/
7 Cornish PL, Knowles SR, Marchesano R, et al. 2018-24170/medicare-program-revisions-to-
Unintended medication discrepancies at the payment-policies-under-the-physician-fee-
time of hospital admission. Arch Intern Med 2005; schedule-and-other-revisions. Published
165(4):424–429. doi:10.1001/archinte.165.4.424. November 2018. Accessed January 31, 2019.
8 McDonnell PJ, Jacobs MR. Hospital admissions 15 Christensen M, Lundh A. Medication review in
resulting from preventable adverse drug reactions. hospitalised patients to reduce morbidity and
Ann Pharmacother 2002;36(9):1331–1336. mortality. Cochrane Database Syst Rev 2016;2:
doi:10.1345/aph.1A333. CD008986. doi:10.1002/14651858.CD008986.
9 Institute for Healthcare Improvement. Medication pub3.
reconciliation to prevent adverse drug events. 16 van Sluisveld N, Zegers M, Natsch S, Wollersheim
ihi.org/topics/ADEsMedicationReconciliation/ H. Medication reconciliation at hospital admission
Pages/default.aspx. Accessed January 31, 2019. and discharge: insufficient knowledge, unclear
10 Institute for Healthcare Improvement. task reallocation and lack of collaboration as major
Medication reconciliation review. ihi.org/ barriers to medication safety. BMC Health Serv
knowledge/Pages/Tools/Medication Res 2012;12:170. doi:10.1186/1472-6963-12-170.
ReconciliationReview.aspx. Accessed 17 Agency for Healthcare Research and Quality.
January 31, 2019. Medications at transitions and clinical handoffs
11 The Joint Commission. Sentinel Event Alert Issue (MATCH) toolkit for medication reconciliation.
35. jointcommission.org/assets/1/18/SEA_35.pdf. ahrq.gov/professionals/quality-patient-safety/
Published January 25, 2006. Accessed patient-safety-resources/resources/match/.
January 31, 2019. Published August 2012. Accessed January 31, 2019.
doi:10.1136/amiajnl-2013-001995.
548 APRIL 2019

18 National Quality Forum. Patient passport 22 Centers for Medicare and Medicaid Services.
encourages better engagement with providers. Welcome to the partnership for patients.
www.qualityforum.org/Patient_Passport.aspx. partnershipforpatients.cms.gov/. Accessed
Accessed January 31, 2019. January 31, 2019.
19 You: Your Own Best Medicine. 23 Heyworth L, Paquin AM, Clark J, et al. Engaging
ownbestmedicine.mn/. Accessed January 31, patients in medication reconciliation via a patient
2019. portal following hospital discharge. J Am Med
Inform Assoc 2014;21(e1):e157–e162.
20 Boockvar KS, Santos SL, Kushniruk A, et al.
Medication reconciliation: barriers and facilitators 24 Greenwald JL, Halasyamani L, Greene J, et al.
from the perspectives of resident physicians and Making inpatient medication reconciliation
pharmacists. J Hosp Med 2011;6(6):329–337. patient centered, clinically relevant and
doi:10.1002/jhm.891. implementable: a consensus statement on key
principles and necessary first steps. J Hosp Med
21 Institute for Healthcare Improvement. How-to
2010;5(8):477–485. doi:10.1002/jhm.849.
guide: prevent adverse drug events (medication
reconciliation) — pediatric supplement. ihi.org/ 25 High-tech approach to medication reconciliation
resources/Pages/Tools/HowtoGuidePrevent saves time, bolsters safety at hospital in northern
AdverseDrugEventsPediatricSupplement.aspx. Virginia. ED Manag 2011;23(10):117–119.
Accessed January 31, 2019.

SELF-ASSESSMENT AND CME TALLY SHEET
Instructions for Completing Continuum Postreading Self-Assessment

Postreading Self-Assessment and CME Test and CME Test Responses
For information on Continuum Audio CME, go to After completing this tally sheet,
continuum.audio-digest.org. please enter your answers online at
continpub.com/CME.
You must be a Continuum subscriber or have purchased a
back issue to earn self-assessment and CME (SA-CME)
credit for this issue. Institutional access does not include EPILEPSY
individual CME. To earn credits, complete the Postreading Vol. 25 No. 2 April 2019
Self-Assessment and CME Test in one of three ways:
1 Go to continpub.com/CME and complete the test online 1 A B C D E 21 A B C D E

(available to subscribers only)
2 A B C D E 22 A B C D E
2 Read through the questions in the issue and mark your
answers on the adjacent tally sheet before entering them
online at continpub.com/CME, or 3 A B C D E 23 A B C D E
3 Email ContinuumCME@aan.com for online access if you 4 A B C D E 24 A B C D E

are a nonsubscriber who has a single issue.
Participants have up to 3 years from the date of
publication to earn AMA PRA Category 1 Credit™ from this
issue. No credit will be awarded for this issue after
April 30, 2022.
US Participants
The ABPN has reviewed Continuum: Lifelong Learning in 8 A B C D E 28 A B C D E
Neurology and has approved this product as part of a
comprehensive lifelong learning and self-assessment 9 A B C D E 29 A B C D E
program, which is mandated by the ABMS as a necessary
component of maintenance of certification.
US participants may earn up to 20 AMA PRA Category 1
Credits in SA-CME by completing the Postreading Self-
Assessment and CME Test. AAN members can view and
download a report of SA-CME credits earned on the
AAN’s NeuroTracker™ on aan.com within 2 business days. 12 A B C D E 32 A B C D E
AAN nonmember subscribers may request a transcript
of credits earned by contacting AAN Member Services at 13 A B C D E 33 A B C D E
memberservices@aan.com or (800) 879-1960.
Canadian Participants
This program is an Accredited Self-Assessment Program
(Section 3) as defined by the Maintenance of Certification 15 A B C D E 35 A B C D E
Program of the Royal College of Physicians and Surgeons
of Canada and approved by the Office of Continuing 16 A B C D E 36 A B C D E
Medical Education and Professional Development,
University of Calgary, on April 1, 2017. Refer to the CME 17 A B C D E 37 A B C D E
tab on ContinuumJournal.com for dates of accreditation.
Canadian participants should visit MAINPORT 18 A B C D E 38 A B C D E
(www.mainport.org) to record learning and outcomes.
Canadian participants can claim a maximum of 20 hours 19 A B C D E 39 A B C D E
(credits are automatically calculated).
For all questions regarding Continuum CME, email 20 A B C D E 40 A B C D E

ContinuumCME@aan.com or call (612) 928-6161.

Postreading
SELF-ASSESSMENT
AND CME
Self-Assessment
and CME Test
By James W. M. Owens Jr, MD, PhD; Allyson R. Zazulia, MD
EPILEPSY
The Continuum Postreading Self-Assessment and CME Test is an integral
part of the issue that is intended to stimulate thought and help participants
assess general understanding of the material presented in this issue. The
Postreading Self-Assessment and CME Test is also approved by the
American Board of Psychiatry and Neurology (ABPN) to meet the Lifelong
Learning (CME), Self-Assessment (SA) (part 2) component for Maintenance
of Certification.
For each item, select the single best response. A tally sheet is provided
with this issue to allow the option of marking answers before entering
them online at continpub.com/CME. Nonsubscribers who have purchased
single back issues should email ContinuumCME@aan.com for instructions
to complete this test online.
US PARTICIPANTS: Upon the completion of the Postreading Self-Assessment

and CME Test and issue evaluation online at continpub.com/CME,
participants may earn up to 20 AMA PRA Category 1 CreditsTM toward
SA-CME. US participants have up to 3 years from the date of publication to
earn SA-CME credits. No SA-CME will be awarded for this issue after
April 30, 2022.
CANADIAN PARTICIPANTS: This

program is an Accredited Self-Assessment
Program (Section 3) as defined by the Maintenance of Certification
Program of the Royal College of Physicians and Surgeons of Canada and
approved by the Office of Continuing Medical Education and Professional
Development, University of Calgary, on April 1, 2017. Refer to the
CME tab on ContinuumJournal.com for dates of accreditation. Canadian
participants should visit MAINPORT (mainport.org) to record learning and
outcomes. Canadian participants can claim a maximum of 20 hours (credits
are automatically calculated).

POSTREADING TEST
ARTICLE 1: EPILEPSY OVERVIEW AND REVISED CLASSIFICATION OF

SEIZURES AND EPILEPSIES
1 An epilepsy diagnosis can be made if the patient has a single unprovoked

seizure and which of the following additional features?
A age younger than 2 years

B at least a 60% risk of another unprovoked seizure within 10 years
C history of febrile seizures in two or more first-degree relatives
D previous provoked seizure
E seizure duration longer than 20 minutes
2 According to the 2017 International League Against Epilepsy

classification system, which of the following terms should be used to
describe a seizure that originates within a neuronal network limited to
one hemisphere?
A discrete
B focal
C generalized
D hemispheric
E partial
3 A focal seizure characterized by irregular, arrhythmic jerking

movements is defined as which of the following seizure types?
A atonic
B clonic
C hyperkinetic
D myoclonic
E tonic
4 A 5-year-old girl with a history of normal birth and development is

brought to the emergency department by her father for evaluation after
a spell. According to her father, she reported a “tummy ache,” vomited,
and developed hypersalivation, pallor, and confusion; she then became
floppy, unresponsive, and incontinent of urine. She gradually regained
consciousness and began speaking after 45 minutes. Neurologic
examination is normal. EEG shows focal high-amplitude occipital
spikes. Which of the following is the most likely diagnosis?
A basilar migraine
B childhood epilepsy with centrotemporal spikes
C idiopathic childhood occipital epilepsy of Gastaut
D Lennox-Gastaut syndrome
E Panayiotopoulos syndrome
554 APRIL 2019

5 In which of the following etiologic categories does juvenile myoclonic
epilepsy belong?
A genetic
B immune
C metabolic
D structural
E unknown
ARTICLE 2: EVALUATION OF SEIZURE ETIOLOGY FROM ROUTINE TESTING

TO GENETIC EVALUATION
6 An infant with seizures developing at 1 week of age has an MRI showing

intracranial calcifications, polymicrogyria, and ventricular dilation.
Which of the following epilepsy etiologies is most likely?
A genetic
B immune
C infectious
D metabolic
E unknown
7 Screening for which of the following disorders is indicated in patients

with treatment-refractory status epilepticus and an antibody directed
against intraneuronal cytoplasmic antigens?
A genetic epilepsy
B inborn error of metabolism
C infection
D neoplasm
E structural cerebral malformation
8 A previously healthy 6-month-old girl presents with fever and left-sided

hemiconvulsions lasting 40 minutes without alteration of
consciousness. Over the next year, she experiences alternating
hemiconvulsive or generalized tonic-clonic seizures 1 or 2 times per
month despite increasing doses of antiepileptic medications. Her seizure
triggers include fever, illness, and vaccinations. Her neurologic
examination is normal. A pathogenic variant in the SCN1A gene is
detected. Which of the following is the most likely diagnosis?
A Dravet syndrome
B febrile seizures
C Lennox-Gastaut syndrome
D Panayiotopoulos syndrome
E severe infantile multifocal epilepsy

POSTREADING TEST
9 Which of the following is first-line therapy for epilepsy associated with

glucose transporter type 1 (GLUT1) deficiency?
A functional hemispherotomy
B ketogenic diet
C phenobarbital
D vagal nerve stimulation
E valproic acid
ARTICLE 3: OPTIMIZING MANAGEMENT OF MEDICALLY RESPONSIVE

EPILEPSY
10 According to the operational definition of seizure freedom put forth by

the International League Against Epilepsy, a patient starting a new
antiseizure medication may be considered seizure free after having gone
without seizures for what length of time?
A for 3 months after starting the new medication

B for 6 months after starting the new medication
C for 9 months after starting the new medication
D for a period equal to 2 times the preintervention interseizure
interval
E for a period equal to 3 times the preintervention interseizure
interval or 12 months, whichever is longer
11 Which of the following is the leading modifiable factor associated with

breakthrough seizures in patients with epilepsy?
A alcohol use
B antiseizure medication nonadherence
C medication interactions affecting antiseizure medication kinetics
D sleep deprivation
E switching antiseizure medications
12 Patients with which of the following epilepsy syndromes are most

susceptible to seizure precipitation by flashing or flickering lights?
A childhood epilepsy with centrotemporal spikes

B frontal lobe epilepsy
C idiopathic generalized epilepsy
D Rasmussen encephalitis
E temporal lobe epilepsy
556 APRIL 2019

ARTICLE 4: IDENTIFICATION AND TREATMENT OF DRUG-RESISTANT
EPILEPSY
13 Which of the following neurologic complications is most frequently

detected after resective epilepsy surgery of the mesial temporal lobe?
A intermittent nystagmus and vertigo

B mild unilateral limb weakness
C mild visual agnosia
D minor visual field deficits
E transient expressive aphasia
14 Which of the following clinical factors is associated with a relative

increased risk of postoperative cognitive decline after epilepsy
surgery?
A family history of dementia

B later age of epilepsy onset
C mesial temporal sclerosis on MRI
D personal history of depression
E preexisting language or memory deficits
15 A 34-year-old right-handed woman with drug-resistant epilepsy

undergoes a scalp EEG monitoring study. Three habitual electroclinical
seizures are captured with frontal bisynchronous onset followed
almost immediately by hyperkinetic motor activity associated with an
excess of muscle and movement artifact that obscures underlying
lateralizing or localizing ictal EEG manifestations. Interictal activity is
seen arising from the frontal lobes bilaterally, at times bisynchronously,
and at times independently, without a clear hemispheric predominance.
The patient’s structural MRI is normal, and a single-photon emission
computed tomography (SPECT) is unrevealing. Neuropsychological
testing reveals cognitive functioning within the average range and no
indication of lateralizing or localizing cognitive deficits. The patient is
highly motivated to pursue a resective surgical option. Which of the
following next steps would be most appropriate?
A intracranial study with use of subdural grid and strip electrodes

B obtain an ictal magnetoencephalogram
C proceed with right anterior temporal lobectomy
D repeat scalp monitoring study to capture more seizures
E stereo-EEG or depth electrodes

POSTREADING TEST
ARTICLE 5: ANTIEPILEPTIC DRUG TREATMENT OF EPILEPSY IN CHILDREN
16 In which of the following patients should initiation of antiepileptic

drug therapy be most strongly considered?
A a cognitively impaired 7-year-old with a single self-limited focal

seizure and a normal EEG and MRI
B a developmentally normal 3-year-old with a single brief focal daytime
seizure and rare multifocal interictal epileptiform transients
C a developmentally normal 5-year-old without seizures diagnosed
with tuberous sclerosis based on ash leaf spots and cardiac
rhabdomyomas on echocardiography
D a developmentally normal 9-year-old with a single generalized
tonic-clonic seizure and centrotemporal spikes on EEG
E a 6-month-old infant with recent developmental regression and
hypsarrhythmia on EEG but no clinical infantile spasms
17 For a child with infantile spasms due to tuberous sclerosis, which of the
following anticonvulsants would be the most appropriate initial choice?
A adrenocorticotropic hormone
B gabapentin
C oral prednisolone
D pyridoxine
E vigabatrin
18 A 6-year-old girl without significant past medical history is seen in

clinic for staring spells and worsening school performance. The parents
also report observing brief episodes at home in which she stops, stares,
fumbles with her hands, and then reinitiates whatever she was doing
after 10 to 20 seconds. One typical event occurs during the clinic visit,
but her neurologic examination is otherwise normal. Three typical
episodes are captured during a routine awake and asleep EEG: two
during hyperventilation and one that occurred spontaneously.
All three episodes were associated with an anterior dominant
generalized spike-and-slow-wave discharge, initially 3.5 Hz to 4 Hz
then slowing to 3 Hz by the end of the seizure. Given her history,
examination, and EEG findings, which of the following treatments is
best supported by evidence?
A clobazam
B ethosuximide
C ketogenic diet
D lamotrigine
E valproic acid
558 APRIL 2019

19 A 4-year-old boy is seen in clinic with refractory epilepsy and several
types of seizures including tonic, atonic, tonic-clonic, and atypical
absence. He is cognitively impaired but otherwise has a nonfocal
neurologic examination. His EEG reveals interictal runs of anterior
dominant 1.5-Hz to 2.5-Hz spike-and-slow-wave activity as well as
bursts of generalized paroxysmal fast activity. Of all his seizure types,
the drop seizures are the most concerning because they have been
associated with frequent injuries. Based on available evidence, which of
the following medications is most likely to decrease the frequency of
his drop seizures?
A clobazam
B lamotrigine
C rufinamide
D topiramate
E valproic acid
ARTICLE 6: TREATMENT OF WOMEN WITH EPILEPSY
20 A 19-year-old woman with temporal lobe epilepsy treated with

lamotrigine and topiramate reports clusters of seizures around the time
of her menstrual period each month. Which of the following treatments
taken perimenstrually is most likely to reduce seizure frequency?
A adrenocorticotropic hormone
B clobazam
C folate
D high-dose ethinyl estradiol
E testosterone propionate
21 Which of the following antiseizure medications is associated with the

greatest risk of sexual dysfunction and decreased libido in women?
A carbamazepine
B clobazam
C lamotrigine
D levetiracetam
E topiramate
22 Which of the following contraceptive methods is most appropriate to

recommend to a woman taking a hepatic enzyme–inducing antiseizure
medication?
A contraceptive vaginal ring

B copper intrauterine device
C levonorgestrel subdermal implant
D progestin-only contraceptive pill
E transdermal contraceptive patch

POSTREADING TEST
23 Which of the following effects is pregnancy most likely to have on

lamotrigine?
A hepatic enzyme induction

B hepatic enzyme suppression
C increased absorption
D reduced plasma volume of distribution
E reduced renal clearance
ARTICLE 7: ELECTROENCEPHALOGRAPHY IN EPILEPSY EVALUATION
24 Which of the following scalp EEG interictal epileptiform discharge

locations is the most specific for mesial temporal lobe epilepsy?
A centrotemporal
B frontotemporal
C medial frontal
D midtemporal
E posterior temporal
25 A 13-year-old girl presents for evaluation of a generalized convulsive

episode with stiffening and shaking lasting for 1 to 2 minutes.
An EEG is performed, which demonstrates interictal 4 Hz to 6 Hz
anterior dominant generalized spike-and-wave and polyspike
and wave bursts. Bursts of similar character are provoked by photic
stimulation at several frequencies of flickering light. No clinical events
or electroclinical seizures are captured. Which of the following
electroclinical syndromes is most likely in this patient?
A childhood absence epilepsy

B juvenile myoclonic epilepsy
C Lennox-Gastaut syndrome
D Panayiotopoulos syndrome
E West syndrome
26 A 31-year-old woman presents for evaluation of seizures characterized

by a difficult-to-describe visceral sensation followed by alteration of
consciousness with orofacial automatisms and with amnesia for the
majority of the event. An EEG is performed, during which one of her
typical seizures is captured. Which of the following electrographic
patterns is most likely to be seen during the altered-awareness phase of
the seizure?
A 5 Hz to 9 Hz frontotemporal rhythmic activity

B irregular 2 Hz to 5 Hz lateral temporal rhythm
C medial frontal monomorphic 2 Hz to 3 Hz activity
D no clear ictal change during the event
E rhythmic sinusoidal 1 Hz to 4 Hz activity over the anterior temporal leads
560 APRIL 2019

27 A 34-year-old woman without a history of epilepsy is brought to the
emergency department for evaluation of altered mental status. An
EEG performed in the emergency department reveals frequent
generalized, bifrontal-maximum, 1.5 Hz to 2 Hz runs of sharp
transients without clear evolution and without any clinical
accompaniment. Which of the following tests or interventions is the
appropriate next step?
A administer an IV antiseizure drug

B increase the number of EEG scalp electrodes
C monitor with EEG for 24 hours without intervention
D place depth electrodes
E use trending and computational analysis
ARTICLE 8: EPILEPSY EMERGENCIES: STATUS EPILEPTICUS, ACUTE

REPETITIVE SEIZURES, AND AUTOIMMUNE ENCEPHALITIS
28 IV administration of which of the following antiseizure drugs is

considered first-line treatment for convulsive status epilepticus?
A fosphenytoin
B levetiracetam
C lorazepam
D phenytoin
E valproate
29 A 38-year-old woman with a remote history of traumatic brain

injury and no neurologic deficits is admitted to the hospital 1 day
after onset of nonfluent aphasia. MRI shows left frontal
encephalomalacia. EEG shows intermittent left hemisphere slowing
and 0.5 Hz to 1 Hz left frontally maximal lateralized periodic
discharges. In the first 4 hours of video-EEG monitoring, she has
three episodes characterized by global aphasia without altered
consciousness coinciding with left frontotemporal seizure activity.
Between electrographic seizures, she is limited to one-word or
two-word answers with difficulty initiating speech. Which of the
following best describes this patient’s condition?
A acute repetitive seizures

B convulsive status epilepticus
C new-onset refractory status epilepticus
D nonconvulsive status epilepticus
E nonepileptic episodic events

POSTREADING TEST
30 A previously healthy 33-year-old woman with a 3-week history of

emotional lability, hallucinations, and facial dyskinesias is admitted to
the hospital with seizures. She has deterioration in mental status and
develops status epilepticus refractory to four antiseizure drugs. She is
intubated and maintained in burst suppression with midazolam and
pentobarbital. Attempts to wean anesthesia are unsuccessful, and there
is no improvement with IV steroids or IV immunoglobulin (IVIg). Brain
MRI shows bilateral mesial temporal lobe hyperintensities on T2-weighted
images. Lumbar puncture shows elevated CSF protein. CT chest, abdomen,
and pelvis and positron emission tomography (PET) are negative for
malignancy. Which of the following is the next best step in management?
A brexanolone
B hypothermia
C ketamine
D rituximab
E transcranial magnetic stimulation
ARTICLE 9: COUNSELING AND MANAGEMENT OF THE RISKS OF LIVING

WITH EPILEPSY
31 Which of the following antidepressants is most associated with an

increased risk of seizures?
A bupropion
B citalopram
C escitalopram
D mirtazapine
E venlafaxine
32 Which of the following antiseizure drugs is most likely to worsen

depression or increase suicidality in a patient with a preexisting mood
disorder?
A gabapentin
B lamotrigine
C levetiracetam
D oxcarbazepine
E valproate
33 Mortality among individuals with epilepsy is most commonly due to

which of the following causes?
A drowning
B injuries sustained during falls
C motor vehicle collisions
D sudden unexpected death in epilepsy (SUDEP)
E underlying medical disorders
562 APRIL 2019

34 Which of the following is the strongest risk factor for sudden
unexpected death in epilepsy (SUDEP)?
A alcohol and other substance abuse

B antiseizure medication polytherapy
C concomitant psychiatric illness
D presence of structural brain lesion
E uncontrolled generalized tonic-clonic seizures
ARTICLE 10: NONEPILEPTIC EPISODIC EVENTS
35 A 25-year-old woman is seen in clinic for stereotyped episodes, which

begin with diaphoresis and a sense of the “world closing in” followed
by loss of consciousness, brief convulsive movements, and rapid return
to consciousness (clear thinking). Which of the following diagnoses is
most likely?
A epileptic seizure
B migraine
C psychogenic nonepileptic seizure
D syncope
E transient ischemic attack
36 A 27-year-old left-handed woman is being evaluated for episodes

concerning for epileptic seizures. A full episode, serendipitously
captured on video, is reviewed. Without any clear prodromal change,
the patient suddenly cries out, closes her eyes and her mouth, and then
turns her head to the right with extension of her right arm and flexion
of her left arm at the elbow. This lasts for 20 seconds and is followed by
nearly symmetric jerks of her arms. During the event, she appears to
not be breathing. She does not remember any of the event. Which of
the following features of this episode is most suggestive of psychogenic
nonepileptic seizures?
A amnesia for the event

B brief duration of event
C eye and mouth closure
D initial cry at onset
E sudden onset

POSTREADING TEST
ARTICLE 11: UPDATE ON ANTIEPILEPTIC DRUGS 2019
37 Which of the following antiepileptic drugs is a potent hepatic P450

system inducer?
A levetiracetam
B oxcarbazepine
C phenobarbital
D pregabalin
E valproic acid
38 A 45-year-old man presents to the emergency department with

new-onset ataxia, confusion, headache, and nausea. He has a history of
epilepsy and is taking an antiseizure drug, although he cannot recall
the name. He was started on erythromycin a few days before the onset
of his new symptoms. Which of the following antiepileptic drugs is he
most likely taking?
A carbamazepine
B lacosamide
C levetiracetam
D oxcarbazepine
E valproic acid
39 Which of the following antiseizure drugs can be effectively

administered 1 time per day?
A gabapentin
B lacosamide
C lamotrigine
D perampanel
E rufinamide
40 Which of the following antiepileptic drugs is most likely to increase

the PR interval on ECG?
A ethosuximide
B felbamate
C lacosamide
D phenytoin
E rufinamide
564 APRIL 2019

Postreading
SELF-ASSESSMENT
AND CME
Self-Assessment and
CME Test—Preferred
Responses
By James W. M. Owens Jr, MD, PhD; Allyson R. Zazulia, MD
EPILEPSY
Following are the preferred responses to the questions in the Postreading
Self-Assessment and CME Test in this Continuum issue. The preferred
response is followed by an explanation and a reference with which you
may seek more specific information. You are encouraged to review the
responses and explanations carefully to evaluate your general
understanding of the article topic. The comments and references included
with each question are intended to encourage independent study.
US PARTICIPANTS: Upon completion of the Postreading Self-Assessment and

CME Test and issue evaluation online at continpub.com/CME, participants
may earn up to 20 AMA PRA Category 1 Credits™ toward SA-CME. US
participants have up to 3 years from the date of publication to earn
SA-CME credits. No SA-CME will be awarded for this issue after
April 30, 2022.
CANADIAN PARTICIPANTS: This

program is an Accredited Self-Assessment
Program (Section 3) as defined by the Maintenance of Certification
Program of the Royal College of Physicians and Surgeons of Canada and
approved by the Office of Continuing Medical Education and Professional
Development, University of Calgary, on April 1, 2017. Refer to the CME
tab on ContinuumJournal.com for dates of accreditation. Canadian
participants should visit MAINPORT (mainport.org) to record learning and
outcomes. Canadian participants can claim a maximum of 20 hours
(credits are automatically calculated).
ARTICLE 1: EPILEPSY OVERVIEW AND REVISED CLASSIFICATION OF

SEIZURES AND EPILEPSIES
1 The preferred response is B (at least a 60% risk of another unprovoked seizure
within 10 years). Epilepsy may be diagnosed when (1) an individual has at least two
unprovoked or reflex seizures more than 24 hours apart; (2) when an individual
has a single unprovoked or reflex seizure and a probability of having another

POSTREADING TEST—PREFERRED RESPONSES
seizure that is similar to the general occurrence risk after two unprovoked
seizures (≥60%) over the next 10 years (at least 60% reflects the 60% lower limit of
the confidence interval for someone with two unprovoked seizures having
another seizure within 10 years); or (3) when an individual has an epilepsy
syndrome. For more information, refer to page 308 of the Continuum article
“Epilepsy Overview and Revised Classification of Seizures and Epilepsies.”
2 The preferred response is B (focal). In the 2017 International League Against

Epilepsy classification system, the term “focal” replaces “partial,” which
had been in use since 1981, to describe seizures originating from a discretely
localized or more widely distributed network limited to one hemisphere.
For more information, refer to pages 309–310 of the Continuum article “Epilepsy
Overview and Revised Classification of Seizures and Epilepsies.”
3 The preferred response is D (myoclonic). Each of the answer options is a motor

onset manifestation of focal seizures. Both myoclonic and clonic seizures
present with jerking, but myoclonic jerking is irregular and not rhythmic, whereas
clonic jerking is regular and rhythmic. Focal atonic seizures involve loss of tone in
a body part. Tonic seizures involve increased tone of an extremity or the neck.
Hyperkinetic seizures involve excessive muscle movement such as thrashing or
pedaling. For more information, refer to pages 310–311 of the Continuum article
“Epilepsy Overview and Revised Classification of Seizures and Epilepsies.”
4 The preferred response is E (Panayiotopoulos syndrome). This child had

a prolonged autonomic seizure with EEG findings demonstrating focal occipital
spikes, which are characteristic of Panayiotopoulos syndrome. Idiopathic
childhood occipital epilepsy of Gastaut also produces occipital discharges on
EEG, but seizures consist of visual hallucinations or blindness that last fewer
than 30 minutes. Lennox-Gastaut syndrome is characterized by multiple seizure
types (usually atonic, atypical absence, and tonic) as well as cognitive,
behavioral, and psychiatric symptoms. Childhood epilepsy with centrotemporal
spikes (rolandic epilepsy) is characterized by unilateral facial sensory-motor
symptoms and centrotemporal spikes on EEG. Occipital and rolandic epilepsy
are commonly associated with migraine, and migraine may produce autonomic
symptoms, but migraine would not account for the EEG findings in this case. For
more information, refer to page 317 of the Continuum article “Epilepsy Overview
and Revised Classification of Seizures and Epilepsies.”
5 The preferred response is A (genetic). The new classification system includes six
etiologic categories: structural, genetic, infectious, metabolic, immune, and
unknown. Genetic etiology is defined by having a known mutation, clinical
presentation with supportive data and family history, or a syndrome with
evidence from research studies to suggest a genetic etiology. The genetic
etiology of juvenile myoclonic epilepsy is inferred from research studies. For
566 APRIL 2019

more information, refer to pages 318–319 of the Continuum article “Epilepsy
Overview and Revised Classification of Seizures and Epilepsies.”
ARTICLE 2: EVALUATION OF SEIZURE ETIOLOGY FROM ROUTINE TESTING

TO GENETIC EVALUATION
6 The preferred response is C (infectious). The likely etiology in this infant

presenting with seizures in the first month of life who has malformations
of cortical development and intracranial calcifications is congenital
cytomegalovirus infection. For more information, refer to pages 335–336 of the
Continuum article “Evaluation of Seizure Etiology From Routine Testing to
Genetic Evaluation.”
7 The preferred response is D (neoplasm). Patients with autoimmune-related

seizures often present with a high frequency of seizures or treatment-refractory
status epilepticus. Antibodies may be directed against neuronal cell surface
antigens (eg, N-methyl-D-aspartate [NMDA] receptor, voltage-gated potassium
channel, contactin-associated proteinlike 2) or intraneuronal nuclear or
cytoplasmic antigens (eg, antineuronal nuclear antibody type 1, collapsin
response mediator protein 5, glutamic acid decarboxylase 65). Both groups are
associated with neoplasm, with a risk greater than 90% in those with nuclear or
cytoplasmic antibodies; thus, screening for cancer is indicated. The testing
strategy depends on the specific autoantibody or clinical syndrome and may
include whole-body fludeoxyglucose positron emission tomography (FDG-PET);
CT of chest, abdomen, and pelvis; breast imaging with a mammogram or MRI; or
testicular ultrasonography. For more information, refer to pages 336–337 and
Table 2-4 of the Continuum article “Evaluation of Seizure Etiology From Routine
Testing to Genetic Evaluation.”
8 The preferred response is A (Dravet syndrome). Dravet syndrome is a severe

epileptic encephalopathy that typically presents with early, prolonged febrile
seizures and is characterized by intractable generalized tonic-clonic, tonic, and
hemiconvulsive seizures, with myoclonic seizures and developmental regression
emerging later in the course. Eighty percent of patients with Dravet syndrome
are found to have a pathogenic variant in the SCN1A gene. For more information,
refer to page 338 of the Continuum article “Evaluation of Seizure Etiology From
Routine Testing to Genetic Evaluation.”
9 The preferred response is B (ketogenic diet). Glucose transporter type 1 (GLUT1)

deficiency syndromes associated with the SLC2A1 gene mutation present in the
first months of life with seizure clusters and status epilepticus during fasting.
A diagnostic clue is a low fasting CSF glucose level. The ketogenic diet is highly
effective in controlling seizures and improving outcome. For more information,

refer to page 339 of the Continuum article “Evaluation of Seizure Etiology From
Routine Testing to Genetic Evaluation.”
ARTICLE 3: OPTIMIZING MANAGEMENT OF MEDICALLY RESPONSIVE

EPILEPSY
10 The preferred response is E (for a period equal to 3 times the preintervention

interseizure interval or 12 months, whichever is longer). The International
League Against Epilepsy defines seizure free as seizure freedom of 12 months
after starting the new medication or for a period equal to 3 times the longest
previous interseizure interval, whichever is longer. This definition is referred to
as “the rule of three,” although subsequent studies suggest that a waiting time
up to 6 times this interval should be required for patients with a low
preintervention probability of seizure freedom. For more information, refer to
page 344 of the Continuum article “Optimizing Management of Medically
Responsive Epilepsy.”
11 The preferred response is B (antiseizure medication nonadherence). While all

the answer options represent modifiable factors associated with breakthrough
seizures in patients with epilepsy, the leading factor is nonadherence to
antiseizure medication. In surveys of patients with breakthrough seizures,
antiseizure medication nonadherence is associated with 27% to 56% of
breakthrough seizures. For more information, refer to page 346 of the
Continuum article “Optimizing Management of Medically Responsive Epilepsy.”
12 The preferred response is C (idiopathic generalized epilepsy). Patients with

idiopathic generalized epilepsies such as absence epilepsy or juvenile myoclonic
epilepsy have 3 to 4 times the prevalence of response to photic stimulation
during EEG than those with symptomatic focal epilepsies. For more information,
refer to page 355 of the Continuum article “Optimizing Management of
Medically Responsive Epilepsy.”
ARTICLE 4: IDENTIFICATION AND TREATMENT OF DRUG-RESISTANT

EPILEPSY
13 The preferred response is D (minor visual field deficits). The most common
neurologic complication following resective epilepsy surgery is minor visual field
deficits. For more information, refer to page 365 of the Continuum article
“Identification and Treatment of Drug-Resistant Epilepsy.”
568 APRIL 2019

14 The preferred response is B (later age of epilepsy onset). The risk of
postoperative cognitive decline is greater with later age of epilepsy onset.
Mesial temporal sclerosis and preexisting language or memory deficits portend
a lower likelihood of postoperative cognitive decline. For more information,
refer to pages 365–366 of the Continuum article “Identification and Treatment
of Drug-Resistant Epilepsy.”
15 The preferred response is E (stereo-EEG or depth electrodes). In this patient

for whom lateralization of seizure onset has not yet been determined,
stereo-EEG is the preferred method for further evaluation. Subdural grid and
strip electrodes are preferable when a focus has been lateralized but requires
further localization. Since three habitual seizures were already captured during
the patient’s scalp monitoring study, it is unlikely that repeating this study will
contribute additional helpful information. An ictal magnetoencephalography
would be very difficult to arrange and also technically challenging and
potentially hazardous for a patient with hyperkinetic seizures. Finally, since
the patient’s seizure behavior and interictal EEG manifestations are highly
suggestive of extratemporal epilepsy localization, anterior temporal lobectomy
without further lateralization or localization would not be indicated in this case.
For more information, refer to pages 369–370 and Figure 4-3 of the Continuum
article “Identification and Treatment of Drug-Resistant Epilepsy.”
ARTICLE 5: ANTIEPILEPTIC DRUG TREATMENT OF EPILEPSY IN CHILDREN
16 The preferred response is E (a 6-month-old infant with recent developmental

regression and hypsarrhythmia on EEG but no clinical spasms). A patient with
developmental regression and hypsarrhythmia on EEG is at very high risk of
infantile spasms, and beginning treatment should be strongly considered.
Developmental regression in this child with hypsarrhythmia on EEG could be
secondary to epileptic encephalopathy. The developmentally normal child with
rare multifocal interictal activity and the cognitively impaired child with a normal
EEG and MRI are at relatively low risk of recurrence. The patient with what
appears to be childhood epilepsy with centrotemporal spikes (formerly called
benign epilepsy with centrotemporal spikes or benign rolandic epilepsy) is very
likely to have few seizures, and treatment is not indicated at this point. Although
most patients with tuberous sclerosis do have epilepsy, a patient who is
diagnosed outside of infancy without epilepsy and with normal development
should not be treated with anticonvulsants prophylactically. For more
information, refer to page 383 of the Continuum article “Antiepileptic Drug
Treatment of Epilepsy in Children.”
17 The preferred response is E (vigabatrin). For patients with infantile

spasms due to tuberous sclerosis, the best evidence-supported treatment at
present is vigabatrin. Intramuscularly administered adrenocorticotropic
hormone is the preferred choice in infantile spasms due to other etiologies. For

more information, refer to page 391 of the Continuum article “Antiepileptic Drug
Treatment of Epilepsy in Children.”
18 The preferred response is B (ethosuximide). This patient has typical childhood

absence epilepsy for which the best treatment is ethosuximide. The Childhood
Absence Epilepsy study compared treatment with ethosuximide, lamotrigine,
and valproic acid. It found that ethosuximide and valproic acid were more
efficacious than lamotrigine and that ethosuximide has fewer behavioral side
effects than valproic acid. Clobazam and the ketogenic diet could be
considered in the treatment of refractory absence. For more information, refer
to pages 393–394, Table 5-3, and Table 5-6 of the Continuum article
“Antiepileptic Drug Treatment of Epilepsy in Children” and the following
reference.
Glauser TA, Cnaan A, Shinnar S, et al. Ethosuximide, valproic acid, and lamotrigine in childhood
absence epilepsy. N Engl J Med 2010;362(9):790–799. doi:10.1056/NEJMoa0902014.
19 The preferred response is A (clobazam). Of all the medications listed, clobazam

has shown the most efficacy against drop seizures compared with placebo. The
other medications listed all have evidence for efficacy in Lennox-Gastaut
syndrome. For more information, refer to Table 5-3 and Table 5-7 of the
Continuum article “Antiepileptic Drug Treatment of Epilepsy in Children.”
ARTICLE 6: TREATMENT OF WOMEN WITH EPILEPSY
20 The preferred response is B (clobazam). This patient has catamenial epilepsy,

defined as a cyclic seizure exacerbation at certain points in the menstrual cycle
attributed to fluctuations in sex hormones. No treatments for catamenial
epilepsy have been approved by the US Food and Drug Administration (FDA),
but interventions with some supporting data include benzodiazepines, such as
clobazam and clonazepam, acetazolamide, progesterone, and a temporary
increase in maintenance seizure medications. Estrogen’s effects on seizures are
complex and depend on the dose, route of administration, and chronicity of
administration, but high-dose estrogen has proconvulsant effects. For more
information, refer to pages 411–412 and Table 6-2 of the Continuum article
“Treatment of Women With Epilepsy.”
21 The preferred response is B (carbamazepine). Sexual dysfunction and

decreased libido are important problems in women with epilepsy that may
reflect effects of both seizures and antiseizure medications. Sexual side
effects are most commonly reported in women taking hepatic enzyme–
inducing antiseizure medications, such as carbamazepine, phenytoin, and
phenobarbital. For more information, refer to pages 413–415 of the Continuum
article “Treatment of Women With Epilepsy.”
570 APRIL 2019

22 The preferred response is B (copper intrauterine device). The use of hepatic
enzyme–inducing antiseizure medications, such as phenytoin, carbamazepine,
oxcarbazepine, eslicarbazepine acetate, phenobarbital, and primidone, can
result in contraceptive failure by lowering the level of the progestin component
of hormonal contraceptive agents. Contraceptive failure can also occur in users
of the less potent hepatic enzyme–inducers perampanel, topiramate, and
felbamate at high doses. Intrauterine devices (IUDs), including the nonhormonal
copper IUD and levonorgestrel-releasing IUD, are not affected by
enzyme-inducing antiseizure medications, making them a reasonable choice
of contraception for women with epilepsy. For more information, refer to
pages 415–416 of the Continuum article “Treatment of Women With Epilepsy.”
23 The preferred response is A (hepatic enzyme induction). The most common

cause of seizure recurrence during pregnancy for women with epilepsy is
likely a reduced plasma concentration of antiseizure medications related to
changes in metabolism, which may be particularly problematic for lamotrigine.
Declining plasma lamotrigine concentrations leading to increased seizure
frequency may occur in 40% of women taking lamotrigine during pregnancy. This
is a result of a sex hormone–mediated induction of metabolism through hepatic
glucuronidation and a twofold to threefold higher rate of drug clearance. Other
contributors to reduced plasma antiseizure medication concentrations during
pregnancy may include decreased drug absorption and increased plasma
volume of distribution. For more information, refer to page 417 of the
Continuum article “Treatment of Women With Epilepsy.”
ARTICLE 7: ELECTROENCEPHALOGRAPHY IN EPILEPSY EVALUATION
24 The preferred response is B (frontotemporal). Frontotemporal interictal

epileptiform discharges are the most specific for mesial temporal lobe
epilepsy. Centrotemporal interictal epileptiform discharges may be seen in
childhood epilepsy with centrotemporal spikes (formerly referred to as benign
rolandic epilepsy). Neocortical temporal lobe epilepsy more often produces
midtemporal and posterior temporal interictal epileptiform discharges. For
more information, refer to pages 433–434 of the Continuum article
“Electroencephalography in Epilepsy Evaluation.”
25 The preferred response is B (juvenile myoclonic epilepsy). Faster frequency

(4 Hz to 6 Hz) generalized bursts with spikes and polyspikes together with
provocation by photic stimulation strongly suggests a diagnosis of juvenile
myoclonic epilepsy. Childhood absence epilepsy typically has slower burst
frequencies of 3 Hz, and polyspikes are less common. Lennox-Gastaut
syndrome may be associated with slow-spike slow-wave bursts, which are less
than 3 Hz in frequency. Panayiotopoulos syndrome is an age-limited occipital
epilepsy syndrome that is not associated with generalized interictal
epileptiform discharges. West syndrome is characterized by hypsarhythmia

with multifocal interictal epileptiform discharges. For more information, refer to

pages 436–438 and Figure 7-8 of the Continuum article “Electroencephalography
in Epilepsy Evaluation.”
26 The preferred response is A (5 Hz to 9 Hz frontotemporal rhythmic activity).

This patient’s epilepsy is clinically most consistent with hippocampal-onset
seizures, and the organized ictal electrographic change during the seizure
would be 5 Hz to 9 Hz frontotemporal rhythmic activity. Rhythmic sinusoidal
activity of 1 Hz to 4 Hz over the anterior temporal leads is temporal intermittent
rhythmic delta activity, an interictal finding in medial temporal lobe epilepsy
(at least on the basis of scalp recordings) but may also an early ictal pattern
that soon evolves into organized theta-range activity. While an irregular lateral
temporal rhythm in the delta-theta range may be an ictal pattern, it is also
seen interictally or in association with seizures originating outside the mesial
temporal lobe. Although some focal seizures may occur without EEG changes,
answer D is unlikely in view of the alteration of awareness, which signifies that
enough brain regions have been involved by the seizure to manifest as an ictal
EEG discharge. For more information, refer to pages 439–441 and Figure 7-10
of the Continuum article “Electroencephalography in Epilepsy Evaluation.”
27 The preferred response is A (administer an IV antiseizure drug). In this context,

given the relative uncertainty regarding whether the EEG pattern is ictal or
interictal, administration of an IV antiseizure drug is indicated because the
patient’s mental status is altered, even if spatiotemporal evolution is not
present (the presence of which would suggest nonconvulsive status
epilepticus). While a continuous EEG for 24 hours would be warranted in this
patient, simply monitoring without intervening will not expeditiously address
whether the findings are ictal or interictal. For more information, refer to
Table 7-1 of the Continuum article “Electroencephalography in Epilepsy
Evaluation.”
ARTICLE 8: EPILEPSY EMERGENCIES: STATUS EPILEPTICUS, ACUTE

REPETITIVE SEIZURES, AND AUTOIMMUNE ENCEPHALITIS
28 The preferred response is C (lorazepam). Lorazepam has been well

established as first-line therapy for convulsive status epilepticus with
superiority over phenytoin demonstrated in the US Department of Veterans
Affairs Cooperative Study. All other answer options represent second-line
treatments. For more information, refer to pages 456–458 and Figure 8-2 of the
Continuum article “Epilepsy Emergencies: Status Epilepticus, Acute Repetitive
Seizures, and Autoimmune Encephalitis” and the following reference.
Treiman DM, Meyers PD, Walton NY, et al. A comparison of four treatments for generalized
convulsive status epilepticus. Veterans Affairs Status Epilepticus Cooperative Study Group. N Engl J
Med 1998;339(12):792–798.
572 APRIL 2019

29 The preferred response is D (nonconvulsive status epilepticus). This is a case
of aphasic status, a form of nonconvulsive status epilepticus. It is considered
nonconvulsive status epilepticus rather than acute repetitive seizures because
nonfluent aphasia persisted and the patient did not return to baseline between
electrographic seizures. A designation of refractory status epilepticus requires
persistent seizures after adequate doses of antiseizure medications, whereas
this patient has not received any antiseizure medications. For more
information, refer to pages 462–464 and Figure 8-5 of the Continuum article
“Epilepsy Emergencies: Status Epilepticus, Acute Repetitive Seizures, and
Autoimmune Encephalitis.”
30 The preferred response is D (rituximab). This patient with new-onset refractory

status epilepticus has a high likelihood of autoimmune encephalitis based on
the subacute onset of neuropsychiatric changes, movement disorder, and
rapidly progressive mental status changes; brain MRI showing signal changes
consistent with limbic encephalitis; and CSF findings consistent with
inflammation. Because she did not respond to first-line immunotherapy with
steroids and IV immunoglobulin (IVIg), second-line therapy with rituximab or
cyclophosphamide is indicated. Brexanolone, an allosteric modulator of
γ-aminobutyric acid A (GABA-A) receptors, was no more effective than
placebo in weaning patients off a medically induced coma in a recent trial. The
other answer choices are treatment options for super-refractory status
epilepticus but are not specific to the likely etiology in this case. For more
information, refer to pages 470 and 472 of the Continuum article “Epilepsy
Emergencies: Status Epilepticus, Acute Repetitive Seizures, and Autoimmune
Encephalitis.”
ARTICLE 9: COUNSELING AND MANAGEMENT OF THE RISKS OF LIVING

WITH EPILEPSY
31 The preferred response is A (bupropion). A fear that antidepressant

medications exacerbate seizures may contribute to the undertreatment of
mood disorders in people with epilepsy, but the preponderance of data
suggest that most antidepressants may be safely used in this population.
Bupropion has long been known to lower seizure threshold and was taken off
the market shortly after its release because of a high incidence of seizures;
it was reapproved when it was found that most cases of seizures occurred
when the drug was taken in excess of 450 mg/d. For more information, refer
to pages 479–480 of the Continuum article “Counseling and Management of
the Risks of Living With Epilepsy.”
32 The preferred response is C (levetiracetam). Levetiracetam has been

demonstrated to increase risk of suicidal thoughts and behavior and is best

avoided in patients with epilepsy who have a preexisting mood disorder. Other
high-risk medications for increasing suicidal behavior include topiramate,
tiagabine, and vigabatrin. Lamotrigine, valproate, and carbamazepine have
mood-stabilizing properties and can be good choices for treatment of mild
depression in people with epilepsy. For more information, including a
discussion about the US Food and Drug Administration (FDA) safety alert for
increased risk of suicidal behavior and suicidal ideation with antiepileptic drugs
as a class, refer to page 481 of the Continuum article “Counseling and
Management of the Risks of Living With Epilepsy.”
33 The preferred response is E (underlying medical disorders). Mortality is

increased among individuals with epilepsy, with odds ratios of 1.6 to 3.0 for
adults and 6.4 to 7.5 for those with childhood-onset epilepsy. The excess
mortality is primarily attributable to the medical disorders underlying epilepsy
rather than to seizure-related injury or sudden unexpected death in epilepsy
(SUDEP). For more information, refer to pages 486–487 of the Continuum article
“Counseling and Management of the Risks of Living With Epilepsy.”
34 The preferred response is E (uncontrolled generalized tonic-clonic seizures).

Sudden unexpected death in epilepsy (SUDEP) is the most common cause of
mortality directly associated with a seizure, and SUDEP is most likely to occur
during sleep in young people with epilepsy. The strongest risk factor for SUDEP
is uncontrolled generalized tonic-clonic seizures, with a 15-fold increased risk
among individuals having three or more seizures per year. For more information,
refer to pages 486–487 of the Continuum article “Counseling and Management
of the Risks of Living With Epilepsy.”
ARTICLE 10: NONEPILEPTIC EPISODIC EVENTS
35 The preferred response is D (syncope). The prodromal diaphoresis and a sense

of the world “closing in” are typical for syncope, as is a brief convulsive phase
and rapid return of awareness/consciousness. Epileptic seizures usually
involve postictal amnesia, and preictal diaphoresis or a sense of visual
constriction would be uncommon. Neither transient ischemic attacks nor
migraine equivalents are typically associated with loss of consciousness.
Psychogenic nonepileptic seizures will generally be longer and would not be
expected to have prodromal diaphoresis. For more information, refer to
pages 494–495 of the Continuum article “Nonepileptic Episodic Events.”
36 The preferred response is C (eye and mouth closure). Closure of the eyes and
mouth during an event are more typical of psychogenic nonepileptic seizures
than epileptic seizures. Amnesia for the event, an initial cry at the onset of
the episode (as opposed to emotional crying during the event), and sudden
onset are all typical of epileptic seizures. Epileptic seizures tend to be briefer
574 APRIL 2019

than psychogenic nonepileptic seizures. For more information, refer to
Table 10-1 of the Continuum article “Nonepileptic Episodic Events.”
ARTICLE 11: UPDATE ON ANTIEPILEPTIC DRUGS 2019
37 The preferred response is C (phenobarbital). Phenobarbital is a potent hepatic

P450 system inducer. Levetiracetam and pregabalin are not hepatically
metabolized and do not affect the P450 system. Oxcarbazepine has only weak
P450 effects, whereas valproic acid is a potent P450 system inhibitor. For
more information, refer to pages 509–510 of the Continuum article “Update on
Antiepileptic Drugs 2019.”
38 The preferred response is A (carbamazepine). Erythromycin is a cytochrome

P450 3A4 (CYP3A4) inhibitor, which can increase levels of carbamazepine,
causing toxicity. Despite a similar structure to carbamazepine, CYP3A4
inhibitors do not affect oxcarbazepine levels. Lacosamide, levetiracetam, and
valproic acid levels are also not affected by CYP3A4 inhibition. For more
information, refer to pages 517–518 of the Continuum article “Update on
Antiepileptic Drugs 2019.”
39 The preferred response is D (perampanel). Perampanel has a plasma half-life

of about 105 hours, and it can, therefore, be administered 1 time per day.
Gabapentin and rufinamide have short half-lives, whereas lacosamide and
lamotrigine have intermediate half-lives. For more information, refer to
page 530 of the Continuum article “Update on Antiepileptic Drugs 2019.”
40 The preferred response is C (lacosamide). Lacosamide can prolong the PR

interval on the ECG. Rufinamide can shorten the QT interval. The other
anticonvulsants listed, at therapeutic doses, are not known to affect either the
PR or the QT interval in a clinically significant manner. For more information,
refer to page 528 of the Continuum article “Update on Antiepileptic
Drugs 2019.”

ERRATUM
In the December 2018 issue of Continuum

(Neurocritical Care, Vol. 24, No. 6), the following
errors occurred:
In TABLE 6-9 of “Coma and Brain Death” by

Alejandro A. Rabinstein, MD, FAAN (Continuum:
Lifelong Learning in Neurology 2018;24:1728), two
entries were listed incorrectly. “Interelectrode
distance ≤10 cm” should be listed as “Interelectrode
distance ≥10 cm” and “High-frequency filter <30 Hz
and low-frequency filter >1 Hz” should be listed as
“High-frequency filter ≥30 Hz and low-frequency
filter ≤1 Hz.”
The author refers readers to the following reference:
Stecker MM, Sabau D, Sullivan LR, et al. American Clinical
Neurophysiology Society guideline 6: minimum technical standards for
EEG recording in suspected cerebral death. Neurodiagn J 2016;56(4):
276–284. doi:10.1080/21646821.2016.1245575.
See the corrected table row below.

Rabinstein AA. Coma and brain death. Continuum (Minneap Minn)
2018;24(6, Neurocritical Care):1708–1731. doi:10.1212/CON.
0000000000000666.
The editors regret these errors.
TABLE 6-9 Ancillary Tests for the Determination of Brain Death
Diagnostic Finding Compatible

Test Testing Conditions With Brain Death Possible Pitfalls
EEG Minimum of eight electrodes: Complete absence of cerebral electric Electric artifacts (common in
activity, including lack of reactivity to intense, the intensive care unit); mostly
Interelectrode distance painful, visual, and auditory stimulation evaluates the cortex
≥10 cm
Interelectrode impedance
between 100 and 10,000 Ω
Sensitivity ≥2 μV
High-frequency filter ≥30 Hz and

low-frequency filter ≤1 Hz
Duration ≥30 minutes
576 APRIL 2019

LEARNING OBJECTIVES AND CORE COMPETENCIES
Learning Objectives ◆ Discuss the classification, evaluation strategies,

and treatment approaches for patients with
Upon completion of this Continuum: Lifelong
nonepileptic episodic events
Learning in Neurology Epilepsy issue,
participants will be able to: ◆ Discuss the spectrum of efficacy, clinical
pharmacology, and modes of use for individual
◆ Review the updated International League Against antiepileptic drugs
Epilepsy seizure and epilepsy classification system,
which now includes epilepsy syndrome and an ◆ Recognize the legal risks and responsibilities in
emphasis on etiology, and understand how the counseling patients with epilepsy about driving
classification system provides a framework for limitations
clinicians, researchers, and patients and their
families ◆ Discuss the importance of medication reconciliation
and identify strategies to implement
◆ Evaluate a first seizure or established epilepsy to medication reconciliation
determine the seizure type, epilepsy syndrome,
and etiology
◆ Define medically responsive epilepsy and discuss Core Competencies

strategies to maintain sustained seizure freedom in This Continuum: Lifelong Learning in Neurology
patients with medically responsive epilepsy Epilepsy issue covers the following
◆ Define drug-resistant epilepsy, review the current
core competencies:
diagnostic and therapeutic tools for surgical
treatment of epilepsy, and discuss the current ◆ Patient Care
evidence in favor of treating patients with drug-resistant
epilepsy with surgical treatment over continued ◆ Medical Knowledge
medical treatment
◆ Practice-Based Learning and Improvement
◆ Discuss the various factors that influence the
initiation, selection, and discontinuation of ◆ Interpersonal and Communication Skills
antiepileptic drugs in children with epilepsy
◆ Professionalism
◆ Discuss the latest data regarding management
of epilepsy in women ◆ Systems-Based Practice
◆ Recognize common interictal and ictal EEG findings

in epilepsy
◆ Describe the diagnosis and treatment for the epilepsy

emergencies of status epilepticus, acute repetitive
seizures, and autoimmune encephalitis
◆ Discuss the aspects of epilepsy beyond seizure

control that can adversely impact safety and quality
of life
296 APRIL 2019

LIST OF ABBREVIATIONS
Epilepsy LGI1 Leucine-rich, glioma inactivated 1

LPD Lateralized periodic discharge
MATCH Medications at Transitions and Clinical Handoffs
AAN American Academy of Neurology MEG Magnetoencephalography
ACHR Acetylcholine receptor MIPS Merit-based Incentive Payment System
ACTH Adrenocorticotropic hormone MRI Magnetic resonance imaging
ADNFLE Autosomal dominant nocturnal frontal lobe epilepsies
mTOR Mammalian target of rapamycin
AED Antiepileptic drug
NAAPR North American AED Pregnancy Registry
AMPA α-Amino-3-hydroxy-5-methylisoxazole-4-proprionic
-Amino-3-hydroxy-5-methylisoxazole-4-proprionic acid
NEAD Neurodevelopmental Effects of Antiepileptic Drugs
APE Antibody Prevalence in Epilepsy
NMDA -methyl-D-aspartate
N-methyl-
BIRD Brief ictal rhythmic discharge
NORSE New-onset refractory status epilepticus
CBT Cognitive-behavioral therapy
CMS Centers for Medicare & Medicaid Services PEACE Pediatric
P ediatric Epilepsy Academic Consortium
for Extrapolation
CSF Cerebrospinal fluid
CSWS Continuous spikes in slow-wave sleep PEARLS Program to Encourage Active, Rewarding Lives
CT Computed tomography PET Positron emission tomography

CYP Cytochrome P450 PLED Periodic lateralized epileptiform discharge
DEXA Dual energy x-ray absorptiometry PMSE Polyhydramnios,
Polyhydramnios, megalencephaly,
ECG Electrocardiogram and symptomatic
symptomatic epilepsy syndrome
EEG Electroencephalogram PNES Psychogenic nonepileptic seizures
ESES Electrical status epilepticus in slow-wave sleep PR interval Electrocardiographic interval from the
EURAP European Registry of Antiepileptic Drugs beginning of the P wave to the middle of the
and Pregnancy QRS complex
FDA US Food and Drug Administration QPP Quality Payment Program

FIRES Febrile infection–related epilepsy syndrome QT interval Electrocardiographic interval from the
FLAIR Fluid-attenuated inversion recovery beginning of the QRS complex to the end of the
T wave
fMRI Functional magnetic resonance imaging
GABA γ-Aminobutyric
-Aminobutyric acid RITE Response to Immunotherapy in Epilepsy
GAD Glutamic acid decarboxylase RNS Responsive neurostimulation
GEFS+ Generalized epilepsies with febrile seizures plus SANAD Standard and New Antiepileptic Drugs
GLUT1 Glucose transporter type 1 SPECT Single-photon emission computed tomography
GPD Generalized periodic discharge SSRI Selective serotonin reuptake inhibitor
HIV Human immunodeficiency virus SUDEP Sudden unexpected death in epilepsy
ILAE International League Against Epilepsy TIRDA Temporal intermittent rhythmic delta activity
IM Intramuscular UDP-GT Uridine diphosphate glucuronyltransferase
IQ Intelligence quotient UPLIFT Using
Using Practice and Learning to Increase
IUD Intrauterine device Favorable Thoughts
IV Intravenous VNS Vagal nerve stimulation
© 2019 American Academy of Neurology.

Epilepsy
Article 1: Epilepsy Overview and Revised
Classification of Seizures and Epilepsies
Alison M. Pack, MD, MPH. Continuum (Minneap Minn). April 2019; 25 (2 Epilepsy):306–321.
ABSTRACT
PURPOSE OF REVIEW:
The classification of seizures, epilepsies, and epilepsy syndromes creates a framework for
clinicians, researchers, and patients and their families. This classification has evolved over the
years, and in 2017 the International League Against Epilepsy (ILAE) published an operational
classification of seizures and epilepsies. Understanding this classification is important in the
diagnosis, treatment, and understanding of seizures and epilepsies, including epilepsy incidence.
RECENT FINDINGS:
The 2017 ILAE classification system builds on newly formulated definitions of seizures and
epilepsy. Seizure classification begins by determining whether the initial manifestations of the
seizure are focal or generalized. If the onset of the seizure is missed or unclear, the seizure is of
unknown onset. Focal seizures are classified according to the individual’s level of awareness,
the most prominent motor or nonmotor features of the seizure, and whether the focal seizure
evolves to a bilateral tonic-clonic seizure. Similarly, generalized seizures are classified
according to motor or nonmotor manifestations. Motor seizures are either tonic-clonic or other
motor seizures. Nonmotor generalized seizures primarily refer to absence seizures. Similar to
seizure classification, the epilepsies can be classified as focal or generalized. In addition, the
new classification system recognizes two new categories: combined generalized and focal
epilepsy and unknown epilepsy. The concept of an epilepsy syndrome has been introduced
under the new classification system and refers to a cluster of features incorporating seizure
types, EEG, imaging, and other features including genetics. The new classification system
emphasizes the etiology of seizures and epilepsies.
SUMMARY:
The recent ILAE seizure and epilepsy classification system aims to create a framework to better
classify seizures and the epilepsies. Universal adoption and implementation of this system will
enable patients, their families, clinicians, and researchers to better define and treat the
epilepsies. Incidence studies have not generally classified seizures and the epilepsies, and use
of this classification system, which emphasizes etiology, will lead to a better understanding of
epilepsy incidence.

KEY POINTS
• In 2014 the International League Against Epilepsy redefined epilepsy as a disease and not a disorder to
emphasize the importance and impact of epilepsy. Epilepsy occurs when an individual has an epileptic
seizure and his or her “brain demonstrates a pathologic and enduring tendency to have recurrent
seizures.”
• Prior and current classification systems aim to group seizures according to clinical presentation and brain
region onset and epilepsies according to seizure type, age of onset, probability of remission, EEG findings,
radiologic findings, and genetics.
• The 2017 International League Against Epilepsy seizure classification addresses limitations of the 1981 seizure
classification, which include the following: (1) some seizure types can have either focal or generalized onset,
(2) lack of knowledge about seizure onset makes a seizure unclassifiable and difficult to place within the
1981 system, (3) retrospective seizure descriptions often do not include a level of consciousness, (4) terms
used in the 1981 seizure classification such as complex partial or simple partial are difficult to understand, and
(5) some seizure types are not included in the 1981 classification.
• Focal seizures originate within a neuronal network limited to one hemisphere that may be discretely localized
or more widely distributed, whereas generalized seizures originate at some point within the brain and rapidly
engage bilateral distributed networks.
• Focal seizures are classified according to the patient’s level of awareness and the first most prominent motor
or nonmotor features of the seizure. These early prominent features are important to consider when
localizing the seizure onset or epileptogenic zone. The final feature used in classification is whether the focal
seizure evolves to a bilateral tonic-clonic seizure.
• Generalized seizures are classified according to motor or nonmotor manifestations. Broadly, motor seizures
are either tonic-clonic or other motor seizures. Nonmotor generalized seizures primarily refer to absence
seizures.
• Seizures of unknown onset can be classified by motor (tonic-clonic, epileptic spasms) or nonmotor (behavior
arrest) presentation. If information is inadequate or if the seizure cannot be categorized, then the seizure
is considered unclassified.
• Similar to seizure classification, the epilepsies are classified as generalized or focal. The new classification
system additionally recognizes two new categories: combined generalized and focal epilepsy and unknown
epilepsy.
• Patients with generalized epilepsy have one or more of the generalized seizure types, and their EEGs typically
display generalized spike-wave activity. For individuals who have generalized seizure types and a normal
EEG, other data are needed to determine whether the epilepsy is generalized.
• Patients with one or more focal seizure types have focal epilepsy. These epilepsies can be either unifocal or
multifocal.
• Designation of combined generalized and focal epilepsy is for patients with both focal and generalized
seizures. EEG may reveal both focal and generalized electrographic findings. Examples of combined
generalized and focal epilepsy include Dravet syndrome and Lennox-Gastaut syndrome.
• When the patient has epilepsy as defined by the International League Against Epilepsy, but it remains
undetermined whether the patient has focal or generalized epilepsy, the classification of unknown epilepsy
type is used.
• The epilepsy syndrome is a new addition to the current classification system and is defined as “a cluster of
features incorporating seizure types, EEG, and imaging features that tend to occur together.”
• Idiopathic generalized epilepsies include childhood absence epilepsy, juvenile absence epilepsy, juvenile
myoclonic epilepsy, and generalized tonic-clonic seizures alone.
• Reflex epilepsy syndromes are epilepsies in which seizures are provoked by a specific stimulus.
• Well-described focal epilepsy syndromes include childhood epilepsy with centrotemporal spikes and
Panayiotopoulos syndrome.
• The etiology of seizures and epilepsies is emphasized in the 2017 International League Against Epilepsy
classification system.

• A structural etiology is determined when a structural abnormality is seen on neuroimaging and when the signs
and symptoms of seizures, in combination with EEG data, suggest this abnormality is the probable cause of
the seizures.
• Genetic etiologies are determined if there is a known or presumed genetic mutation in which seizures are a
core symptom of the disorder.
• Infectious etiologies are the most common worldwide etiology of epilepsy.
• Epilepsies with a metabolic etiology occur secondary to a known or presumed metabolic disorder in which
seizures are a core symptom of the disorder.
• Immune etiologies are increasingly recognized as potential causes of epilepsy.
Article 2: Evaluation of Seizure Etiology

From Routine Testing to Genetic
Evaluation
Stephan U. Schuele, MD, MPH, FAAN. Continuum (Minneap Minn). April 2019; 25 (2
Epilepsy):322–342.
ABSTRACT
PURPOSE OF REVIEW:
Recognizing the cause of a first seizure and identifying the etiology of epilepsy are essential for
management. A systematic approach to patients who present with a first seizure helps distinguish
between an acute symptomatic seizure, a provoked or unprovoked seizure, and potential
mimickers. Routine testing with EEG and MRI may reveal a predisposition for further seizures and
help to establish the underlying epilepsy syndrome. An acquired etiology can be identified in
30% of patients with established epilepsy. The remaining 70% of patients have a presumably
genetic etiology. Particularly in patients with specific epilepsy syndromes or suspicion for an
autosomal dominant inheritance, genetic testing and counseling should be considered.
RECENT FINDINGS:
Neuroimaging, autoimmune antibodies, and genetic testing have revolutionized our ability to
investigate the etiology of many epilepsies. The new epilepsy classification distinguishes
structural, metabolic, genetic, infectious, and immune-mediated etiologies, which often help
determine prognosis and treatment.
SUMMARY:
There is growing acceptance and demystification of the term epilepsy as the most common cause
for recurrent seizures. The new classification of epilepsy does not stop with the recognition of
particular epilepsy syndromes but aims to determine the underlying etiology. This can lead to
earlier recognition of surgical candidates, a better understanding of many of the genetic
epilepsies, and medical treatments aimed at the underlying mechanism causing the disease.
KEY POINTS
• Syncope is the most common mimicker of a new-onset seizure. A situational trigger, an aura of “going to
faint,” and a quick recovery often support a diagnosis of vasovagal syncope.
• Laboratory testing after a first seizure should include a complete blood cell count; blood chemistry,
including calcium, magnesium, and phosphate; thyroid-stimulating hormone; and urine toxicology. A 12-lead
ECG should also be considered.

• Acute brain insults can have a risk of leading to late seizures in up to 20% of patients and do not warrant
prolonged seizure prophylaxis, even if the injury was associated with early seizures or status. Exceptions are
possibly a penetrating head trauma and herpes encephalitis, which are associated with a 50% risk of
• Typical MRI lesions associated with surgically amenable epilepsy, such as glioneuronal tumors, low-grade
gliomas, hippocampal sclerosis, small cavernous malformations, and subtle malformations of cortical
development (eg, cortical dysplasia or isolated periventricular nodular heterotopia), are not seen on brain CT
scans and often missed on standard MRIs.
• An epilepsy-protocol brain MRI differs from a typical MRI in that it includes thin 1- to 3-mm slices without
interslice gap and coronal fluid–attenuated inversion recovery sequences, which offer additional sensitivity
over standard-protocol MRIs.
• A brain CT and an EEG should be considered even in the presence of an alternative etiology for a provoked
seizure.
• According to national guidelines, patients who continue to have seizures after 1 year on medical therapy
should be referred to an epilepsy center.
• Genetic testing should be considered in all patients with an autosomal dominant familial epilepsy, epileptic
encephalopathies, and progressive myoclonic epilepsy and patients with clinical findings suggestive of a
genetic etiology.
• Whole exome sequencing and whole genome sequencing are usually performed within the setting of a
genetic clinic and research program with access to genetic counseling and the ability to interpret variants of
unknown significance.
• It is worthwhile to be familiar with glucose transporter type 1 (GLUT1) deficiency syndromes associated with
SLC2A1 gene mutation given the excellent response to dietary treatment and the recent discovery of milder,
later-onset variants.
• Genetic counseling for a specific epilepsy syndrome can be straightforward in patients with a 100% penetrant
syndrome but complicated in situations of incomplete penetrance and de novo mutations (eg, Dravet
syndrome) and should be performed with the help of a genetic counselor.
• The high prevalence of comorbidities associated with epilepsy is best addressed within a comprehensive
care setting including a social worker, psychologist, psychiatrist, neuropsychologist, dietician, and genetic
counselor.
Article 3: Optimizing Management of

Medically Responsive Epilepsy
Derek Bauer, MD; Mark Quigg, MD, MSc, FANA, FAES. Continuum (Minneap Minn). April
2019; 25 (2 Epilepsy):343–361.
ABSTRACT
PURPOSE OF REVIEW:
This article reviews the management of patients with medically responsive epilepsy, including
discussion of factors that may lead to transient breakthrough seizures and patient and physician
strategies to maintain freedom from seizures.
RECENT FINDINGS:
Imperfect adherence, unanticipated changes in ongoing medical therapy, inadvertent use of
proconvulsants or concurrent medications that alter epilepsy medication kinetics, and a variety
of seizure precipitants such as stress or sleep deprivation may alter long-term seizure control.

SUMMARY:
The majority of patients with epilepsy are medically responsive. Many potential factors may lead
to breakthrough seizures in these patients. Identification of these factors, patient education, and
use of self-management techniques including mindfulness therapy and cognitive-behavioral
therapy may play a role in protecting patients with epilepsy against breakthrough seizures.
KEY POINTS
• Approximately two-thirds of patients with epilepsy will become seizure free with pharmacotherapy.
• Comparatively little evidence exists for evaluating risk factors for breakthrough seizures. The available data
suggest that nonadherence is the most important factor leading to breakthrough seizures.
• Antiseizure medication nonadherence is common and should be screened for when a patient’s seizures are
uncontrolled.
• Increasing complexity of the antiseizure medication regimen, including increased dosing frequency or
increased numbers of medications, is associated with an increased risk of medication nonadherence.
• Many demographic factors may convey an increased risk of nonadherence, including socioeconomic status
and race, although other factors such as gender and age are not clearly associated with increased risk.
• Current data suggest that adherence may be improved by increasing patient education and providing
feedback to address concerns that may be obstacles to adherence.
• Medications and supplements may cause seizures, either through interactions with antiseizure medication or
through proconvulsant properties.
• Switching antiseizure medications may result in seizure recurrence, even in the setting of prolonged seizure
freedom.
• Current data from large population studies suggest that generic antiepileptic drugs are bioequivalent to name
brand medication.
• Stress and sleep deprivation are the most common seizure precipitants in patients with epilepsy.
• Insomnia is common in patients with epilepsy and is correlated with poor seizure control.
• Mood disorders are common in patients with epilepsy. Screening should be performed to address both
inherent mood concerns but also to potentially improve seizure control.
• Self-management techniques may improve psychiatric comorbidities associated with epilepsy, and some
evidence suggests this may translate into improved seizure frequency.
• The first-line treatment for insomnia is cognitive-behavioral therapy; sedative-hypnotic use should be
avoided as much as possible in patients with epilepsy because of risks of polypharmacy and the fluctuation
of seizure thresholds, such as in the withdrawal from and habituation to benzodiazepines.
Article 4: Identification and Treatment of

Drug-Resistant Epilepsy
Ji Yeoun Yoo, MD; Fedor Panov, MD. Continuum (Minneap Minn). April 2019; 25 (2
ABSTRACT
PURPOSE OF REVIEW:
Drug-resistant epilepsy is a potentially life-threatening condition affecting one-third of people
living with epilepsy. Despite existing evidence of improved outcomes in patients who received
surgical treatment compared to continued medical treatment, epilepsy surgery remains
underused in patients with drug-resistant epilepsy. This article discusses the gap between

evidence and practice and common misconceptions about epilepsy surgery and reviews the
current diagnostic and therapeutic surgical options.
RECENT FINDINGS:
Three randomized controlled trials comparing the medical versus surgical treatment for patients
with drug-resistant epilepsy have shown the superiority of surgery in controlling seizures and
improving patients’ quality of life. In addition to resective surgery, neuromodulation through
devices such as responsive neurostimulation and vagal nerve stimulation have also shown
efficacy in seizure control that increases over time. Diagnostic and therapeutic surgical tools are
tailored to the needs of each patient.
SUMMARY:
Appropriate patients with drug-resistant epilepsy benefit more from epilepsy surgery than from
continuing medical treatment. These patients should be referred to comprehensive epilepsy
centers where a thorough presurgical workup and surgical options can be provided. The gap
between evidence and practice can be bridged by education, community outreach, and
providers’ earnest efforts to improve the quality of life of patients with epilepsy.
KEY POINTS
• Drug-resistant epilepsy is defined as the “failure of adequate trials of two tolerated, appropriately chosen
and used antiepileptic drug schedules (whether as monotherapies or in combination) to achieve sustained
seizure freedom.”
• Drug-resistant epilepsy affects about one-third of people living with epilepsy.
• Currently, three randomized controlled trials have shown superiority of surgery compared to continued
medical treatment in patients with drug-resistant epilepsy, not only for seizure control but also for quality
of life.
• A formal practice guideline by the American Academy of Neurology, American Epilepsy Society, and
American Association of Neurological Surgeons recommends that the patients for whom appropriate trials of
first-line antiepileptic drugs have failed should be considered for referral to an epilepsy surgery center.
• A gap between evidence and practice exists in regard to treatment for patients with drug-resistant epilepsy.
Although evidence clearly dictates referral of patients with drug-resistant epilepsy to a comprehensive
epilepsy center, such referrals are still not completed due to a variety of reasons.
• Epilepsy surgery is generally safe; most complications are minor and transient.
• Cognition and memory deficits are not contraindications to epilepsy surgery.
• The risk of cognitive and memory decline is less when an abnormality is seen in imaging such as mesial
temporal sclerosis, or with earlier age of seizure onset, or with preexisting memory and language deficits.
• After failure (having recurrent seizures despite adequate trials) of two antiepileptic drugs, the chance of a
third antiepileptic drug effectively controlling a patient’s seizures is very low.
• “Normal” brain MRIs are not always normal. A careful review by dedicated neuroradiologists, with
attention to the patient’s seizure symptomatology, and correlation with EEG findings, can help identity subtle
findings.
• A focal lesion is not always the culprit for the patient’s seizures, and careful review through multiple
modalities is still necessary.
• Involvement of eloquent cortex is not a contraindication to epilepsy surgery. When the resection of the
epileptogenic zone is not possible, a device can be an effective treatment option.
• A treatment-viable seizure onset zone can still be identified even if multiple or diffuse lesions are seen on
MRI or if multifocal spikes are seen on EEG.
• What seems to be generalized epilepsy or bilateral spikes on EEG may still have a unilateral seizure onset
zone, and in cases of truly generalized epilepsy, a device can be an effective option.
• A referral to an epilepsy center is not only for possible epilepsy surgery but is also to identify a
“pseudoresistance” and to adjust treatments accordingly.

• Invasive diagnostic tools are used for delineating the intracranial seizure onset zone and for functional brain
mapping. The diagnostic tools include grids, strips, depth electrodes, or stereo-EEG, and these tools can
be used as dictated by the needs of the specific patient.
• Different surgical options are available, including resection, laser ablation, and neuromodulatory devices,
with therapeutic surgery tailored to the specific patient.
• The efficacy of vagal nerve stimulation improves over time, with 60% of patients experiencing a significant
response (>50% seizure reduction), but the goal of seizure freedom is unlikely to be attained with this therapy
(8% seizure freedom after 2 years).
Article 5: Antiepileptic Drug Treatment of

Epilepsy in Children
Ahsan N. V. Moosa, MD. Continuum (Minneap Minn). April 2019; 25 (2 Epilepsy):381–407.
ABSTRACT
PURPOSE OF REVIEW:
The treatment of epilepsy in children is highly individualized at each and every major step in the
management. This review examines various factors that modify the treatment from the point
of initiation of therapy to the decision to stop an antiepileptic drug (AED).
RECENT FINDINGS:
AED therapy leads to seizure freedom in about 70% of all children with epilepsy. AED initiation
could be delayed until a second seizure in most children and may be avoided altogether in many
children with self-limited childhood focal epilepsies. Three key factors influence the choice
of AED: seizure type(s), efficacy of the drug for the seizure type, and the side effect profile of
the drug(s). For epileptic spasms, steroids and vigabatrin are the most effective treatment
options. For absence seizures, ethosuximide and valproic acid are superior to lamotrigine. For
focal seizures, many newer AEDs have favorable side effect profiles with efficacy comparable to
older-generation drugs. For generalized epilepsies, valproic acid remains the most effective
drug for a broad range of seizure types. Genetic and metabolic etiologies may guide unique
treatment choices in some children. After 2 years or more of seizure freedom, if the recurrence
risk after AED withdrawal is acceptable, slow weaning of AEDs should be done over the span of
6 weeks or longer. After discontinuation, about 70% of patients remain seizure free, and of those
with recurrence, the majority achieve seizure control with restarting an AED. When treatment
with two or more AEDs fails, other treatment opportunities for drug-resistant epilepsy, including
epilepsy surgery, vagal nerve stimulation, and dietary therapies should be considered.
SUMMARY:
Carefully selected medical therapy guided by seizure type and AED characteristics is effective in
more than two-thirds of children with epilepsy.
KEY POINTS
• A clinical diagnosis of epilepsy is met when a child has two or more unprovoked seizures or if the predicted
risk of recurrence of seizure is 60% or more after the first seizure.
• In selected children with self-limited childhood focal epilepsies (with centrotemporal or occipital spikes),
long-term daily antiepileptic drug treatment may not be needed because children invariably outgrow their
tendency to have seizures.

• The single most important determinant of treatment choice is the type of seizure that is being targeted.
• Drugs with low risk of drowsiness and cognitive dulling are preferred.
• A higher risk of Stevens-Johnson syndrome in patients of Asian ancestry with the HLA-B*1502 allele and a risk
of liver failure with valproic acid in patients with a POLG1 mutation are two noteworthy situations in which
pharmacogenomics factors influence antiepileptic drug selection.
• The results of a network meta-analysis showed that, for focal seizures, lamotrigine and levetiracetam were
significantly better than carbamazepine, which was better than phenytoin and phenobarbital.
• The results of a network meta-analysis showed that, for generalized seizures, valproic acid was superior to
phenobarbital, topiramate, and carbamazepine.
• The American Academy of Neurology quality committee recommends initiation of therapy within 7 days of
the onset of epileptic spasms to improve developmental outcome.
• Steroids (oral prednisone or IM adrenocorticotropic hormone) and vigabatrin are the mainstays of therapy for
epileptic spasms.
• Reversible hyperintense signal abnormalities in white matter, basal ganglia, thalamus, and brainstem have
been reported in as many as 30% of infants treated with vigabatrin.
• Both valproate and ethosuximide produce comparable rates of seizure control for absence seizures, but
ethosuximide has a favorable side effect profile, particularly in the behavioral domain.
• The focus of management in Lennox-Gastaut syndrome should be on the overall quality of life and not seizure
count per se.
• Valproic acid is frequently used as a first-line medication for children with Lennox-Gastaut syndrome
because of its broad spectrum of action against various seizure types.
• Two commonly used treatment options for electrical status epilepticus in slow-wave sleep include
high-dose nightly benzodiazepines and steroids.
• Examples of antiepileptic drugs known to worsen epilepsy in certain diseases include phenytoin,
carbamazepine, and lamotrigine in Dravet syndrome due to SCNA1 mutation.
• A pooled analysis of five trials in children provided evidence to support a minimum 2-year seizure-free period
before considering the need for continued treatment.
• The majority of children with epilepsy belong in the “uncertain” group for whom the long-term chances for
seizure remission without antiepileptic drugs are not very predictable.
• Pooled analyses of studies in children who had achieved sustained seizure freedom show that 70% remained
seizure free after discontinuation of antiepileptic drugs, and 30% had one or more recurrences.
• Risk factors for increased recurrence include epileptiform abnormalities on EEG, epilepsy onset before
2 years of age and after 10 years of age, intellectual disability (IQ <70), history of status epilepticus, and higher
seizure burden before and during treatment.
• A comparison of slow weaning over 9 months versus relatively faster weaning over 6 weeks showed no
significant difference in seizure recurrence risk.
Article 6: Treatment of Women With

Epilepsy
Mona Sazgar, MD, FAES. Continuum (Minneap Minn). April 2019; 25 (2 Epilepsy):408–430.
ABSTRACT
PURPOSE OF REVIEW:
This article provides the latest information to guide practitioners in counseling and treating
women with epilepsy.

RECENT FINDINGS:
There is an increasing body of literature on the multidirectional effects of sex hormones on
seizure frequency and severity and of seizures altering areas of the brain involved in
neuroendocrine function. Ongoing pregnancy outcome data from pregnancy registries and
meta-analysis of observational studies have provided key information on the safety of using
antiseizure medications during pregnancy and the risk to the fetus.
SUMMARY:
In treating and counseling women with epilepsy from puberty to menopause, it is important to
understand the complex interactions of sex hormones, seizures, and antiseizure medications on
reproductive health and pregnancy outcomes.
KEY POINTS
• Nearly 1.5 million women of childbearing age in the United States live with epilepsy.
• Catamenial epilepsy is reported in at least one-third of women with epilepsy.
• Women with catamenial epilepsy have a cyclic exacerbation of their seizures with the fluctuation of their hormones.
• As a rule, estrogens are proconvulsant and progesterone is anticonvulsant.
• Menstrual disorders are estimated to occur in one-third of women with epilepsy as compared with 12% to 14%
of women in the general population.
• The oral contraceptive failure rate is 1% in healthy women but 3% to 6% in the population of women with epilepsy.
• Neurologists should be aware of the complex and at times bidirectional interactions between antiepileptic
drugs and hormonal contraception.
• Intrauterine devices are a safe and effective method of contraception in women with epilepsy.
• In most women with epilepsy, pregnancy has no effect or a protective effect on their seizure frequency.
• Having a seizure disorder that was active in the year before pregnancy appears to be the best predictor of
seizure recurrence during pregnancy.
• There is evidence for better seizure control during pregnancy in women with catamenial epilepsy compared
with women with epilepsy in general.
• The most common cause of seizure recurrence in pregnancy is likely a reduced plasma concentration of
antiepileptic drugs and changes in antiepileptic drug metabolism.
• American Academy of Neurology guidelines recommend checking antiepileptic drug levels at baseline
before conception and monthly thereafter.
• In counseling women with epilepsy who plan to become pregnant, the practitioners should discuss the need
for staying on antiseizure medication, simplifying the medication regimen, attempting monotherapy, and
selecting medications with a more favorable side effect profile.
• Exposure to valproate in utero may adversely affect child IQ and contribute to autistic traits.
• In the Neurodevelopmental Effects of Antiepileptic Drugs study, breast-fed children exposed to
antiepileptic drugs in utero and through breast milk exhibited higher IQs and enhanced verbal abilities
compared with children exposed to antiepileptic drugs in utero who were not breast-fed.
• At perimenopause, seizures may worsen because of a temporary increased estrogen-to-progesterone ratio.
After menopause is established, women with catamenial epilepsy may experience improvement in seizures.
• Antiepileptic drugs used to treat epilepsy can adversely affect bone health and calcium metabolism.
Article 7: Electroencephalography in
Epilepsy Evaluation
Hai Chen, MD, PhD; Mohamad Z. Koubeissi, MD, FAAN. Continuum (Minneap Minn).
April 2019; 25 (2 Epilepsy):431–453.

ABSTRACT
PURPOSE OF REVIEW:
Epilepsy is a heterogeneous disorder that is often associated with abnormal
electroencephalogram (EEG) findings. This article provides an overview of common EEG findings
in epileptic disorders. The physiologic basis of EEG and intracranial EEG studies is also discussed.
RECENT FINDINGS:
EEG is widely used in clinical practice. Because of the paroxysmal nature of seizure disorders,
interictal epileptiform discharges, such as spikes and sharp waves, are often used to support the
diagnosis of epilepsy when a habitual seizure is not captured by EEG. Interictal and ictal EEG
findings also underlie the classification of seizures and epilepsy. Continuous critical care EEG
monitoring has become an invaluable study in the diagnosis and treatment of subclinical seizures
and nonconvulsive status epilepticus. Intracranial EEG with subdural or intraparenchymal
electrodes is warranted when localization of the seizure focus and mapping of eloquent brain
areas are required to plan epilepsy surgery.
SUMMARY:
The EEG is a key tool in the diagnosis of epilepsy. Interictal and ictal EEG findings are crucial for
the confirmation and classification of seizure disorders. Intracranial EEG monitoring is also
indispensable for planning surgery for some patients.
KEY POINTS
• The EEG signal is generated by the summation of extracellular postsynaptic potentials.
• Paroxysmal depolarizing shifts are considered the intracellular correlate of the interictal epileptiform
discharge on EEG recordings.
• The finding of interictal epileptiform discharges may aid in the diagnosis of epilepsy.
• Ictal discharges seen on EEG in patients with epilepsy vary in morphology, frequency, and distribution and
often show pattern evolutions.
• Seizures are classified as focal-onset or generalized-onset seizures.
• Periodic discharges can be considered as ictal or interictal phenomena in various clinical settings.
• Continuous EEG monitoring is important in the diagnosis and treatment of nonconvulsive status epilepticus.
• Intracranial electrodes allow for earlier seizure detection and spatial resolution than do scalp recordings.
• Intracranial monitoring is an invaluable tool in surgical planning.
Article 8: Epilepsy Emergencies: Status

Epilepticus, Acute Repetitive Seizures,
and Autoimmune Encephalitis
Stephen VanHaerents, MD; Elizabeth E. Gerard, MD. Continuum (Minneap Minn).
April 2019; 25 (2 Epilepsy):454–476.
ABSTRACT
PURPOSE OF REVIEW:
This article reviews epilepsy emergencies, including status epilepticus, acute repetitive
seizures, autoimmune encephalitis, and the current perspective on their diagnosis
and treatment.

RECENT FINDINGS:
Recent guidelines on the treatment of status epilepticus from the Neurocritical Care Society in
2012 and the American Epilepsy Society in 2016 highlight areas of consensus in the treatment of
status epilepticus as well as areas of uncertainty. The TRENdS (Treatment of Recurrent
Electrographic Nonconvulsive Seizures) trial is the first prospective randomized clinical trial to
evaluate the efficacy of IV antiseizure medications in controlling nonconvulsive seizures on
continuous EEG. It demonstrated that IV lacosamide is noninferior to fosphenytoin in this setting.
Autoimmune encephalitis is an increasingly recognized cause of new-onset seizures or status
epilepticus. Recently described scoring systems, the Antibody Prevalence in Epilepsy score and
the Response to Immunotherapy in Epilepsy score, can help in the assessment of autoimmune
encephalitis.
SUMMARY:
Status epilepticus, acute repetitive seizures, and autoimmune encephalitis are neurologic
emergencies. For all these conditions, rapid and appropriate treatment may influence patient
prognosis and mitigate neuronal injury. For convulsive status epilepticus, there is reasonable
consensus on the initial steps that need to be taken. There is less agreement about the
management of acute repetitive seizures and nonconvulsive status epilepticus. An increasingly
recognized etiology of status epilepticus is autoimmune encephalitis, which may not be as rare
as previously thought.
KEY POINTS
• Currently, the most commonly accepted definition for convulsive status epilepticus is either 5 minutes or
more of continuous seizure activity or two or more discrete seizures between which there is incomplete
recovery of consciousness.
• Refractory status epilepticus refers to either clinical or electrographic seizures that persist after adequate
doses of an initial benzodiazepine and an acceptable second-line antiseizure drug.
• Super-refractory status epilepticus is defined as status epilepticus that continues or recurs 24 hours or more
after the initiation of anesthetic therapy.
• Benzodiazepine therapy has been well established as the first-line treatment for convulsive status
epilepticus.
• Both guidelines (from the Neurocritical Care Society and American Epilepsy Society) state that phenytoin,
fosphenytoin, valproate sodium, levetiracetam, and phenobarbital are reasonable choices for second-line
antiseizure drugs with little evidence to guide the choice between them.
• One proposed clinical definition of acute repetitive seizures is three or more seizures within 24 hours for
patients whose habitual seizure frequency is fewer than three seizures per day.
• In a patient who is critically ill, his or her mental status may be affected by the underlying cause of seizures,
so it can be difficult to determine if the patient would return to baseline between seizures.
• In a recent randomized controlled trial, lacosamide was noninferior to fosphenytoin in the treatment of
nonconvulsive seizures.
• No clinical consensus exists for how aggressively to treat nonconvulsive status epilepticus, but in clinical
practice, the determination is often based on the patient’s mental status and clinical course.
• New-onset refractory status epilepticus (NORSE) is a recently described clinical presentation that can affect
all ages and occurs in patients without active epilepsy or other preexisting neurologic disorders and has no
clear acute or active structural, toxic, or metabolic cause.
• Febrile infection–related epilepsy syndrome (FIRES) is thought to be a subcategory of new-onset refractory
status epilepticus that requires a preceding febrile infection, with fever starting between 2 weeks and
24 hours before the onset of refractory status epilepticus.
• In one study, an Antibody Prevalence in Epilepsy score of 4 or greater had a sensitivity of 82.6% and a
specificity of 82.0% for detecting a positive antibody.

• In one study, a Response to Immunotherapy in Epilepsy score of 7 or greater predicted favorable seizure
outcome and had a sensitivity of 87.5% and specificity of 83.8%.
• When choosing a treatment strategy for any antibody-associated central nervous system disorder, it is key to
choose objective markers to monitor treatment response.
Article 9: Counseling and Management of

the Risks of Living With Epilepsy
Katherine Noe, MD, PhD, FAAN. Continuum (Minneap Minn). April 2019; 25 (2 Epilepsy):
477–491.
ABSTRACT
PURPOSE OF REVIEW:
For patients living with epilepsy, quality of life is determined not only by seizure control but by
mood, antiepileptic drug adverse effects, relationships, and access to education, employment,
and transportation. This article reviews some of the most commonly encountered concerns
associated with epilepsy, including mood disorders, driving, injuries, mortality, bone health,
genetic burden, and impact on relationships.
RECENT FINDINGS:
People with epilepsy are at increased risk for anxiety, depression, and suicide. Depression is
underrecognized in patients with epilepsy, but effective validated screening tools are available
for use. Mortality rates for people with epilepsy are 2 times higher than those of the general
population, but much of this is attributable to underlying conditions rather than seizures. Sudden
unexpected death in epilepsy (SUDEP) occurs in an estimated 1:1000 adults with epilepsy per
year, and the risk can be reduced by improved observation and seizure control. An increased risk
of injury, including fractures, is also present in patients with epilepsy. Reduced bone health
leading to increased fracture risk is an important negative consequence of long-term use of
antiepileptic medication. Seizures while driving can also cause accidents and injury. Despite the
importance of driving for people with epilepsy, physicians are underperforming in providing
counsel about driving.
SUMMARY:
Optimal care of the patient with epilepsy includes addressing risks to emotional health, physical
health including fractures and SUDEP, social health, and an independent lifestyle. Identification
of and treatments to reduce these risks can do more to improve quality of life than a narrow
clinical focus on seizure control alone.
KEY POINTS
• Depression risk is increased 2 to 3 times in people with epilepsy. People with epilepsy are twice as likely to
report suicidal thoughts and up to 3 times more likely to die of suicide compared to the general population.
• Depression in patients with epilepsy is underrecognized and undertreated.
• The often-cited fear that a selective serotonin reuptake inhibitor will worsen seizure control is not supported
by evidence and should not be a barrier to treatment of depression in the patient with epilepsy.
• People living with epilepsy rated their ability to drive as the most important factor determining quality of life.
• Physicians are underperforming in counseling patients with epilepsy about driving.
• Fracture risk is 2 to 6 times higher in people with epilepsy than in the general population, and fracture risk
increases with duration of antiepileptic drug therapy.

• Evidence-based or expert consensus recommendations on prevention, screening, and treatment of bone
disease and fractures in people with epilepsy are lacking. Medical providers must rely on common sense or
strategies used in the general population.
• Mortality in patients with epilepsy is primarily attributable to underlying medical disorders, rather than to
seizure-related injury or sudden unexpected death in epilepsy.
• Uncontrolled generalized tonic-clonic seizures are an important risk factor for sudden unexpected death
in epilepsy.
• The majority of people with epilepsy want to be informed by their physician about sudden unexpected death
in epilepsy.
• People with epilepsy tend to overestimate the risk of passing on epilepsy to a child, which may falsely
influence decisions about having children.
Article 10: Nonepileptic Episodic Events

Jennifer L. Hopp, MD, FAAN. Continuum (Minneap Minn). April 2019; 25 (2 Epilepsy):
492–507.
ABSTRACT
PURPOSE OF REVIEW:
This review addresses the scope, evaluation, treatments, and outcomes of patients with
nonepileptic episodic events with a focus on psychogenic nonepileptic seizures. Differentiation
of the types of events, including a review of terminology, is included, as well as a brief review
of special patient populations with these disorders.
RECENT FINDINGS:
There are continued efforts to develop tools to improve the diagnosis of these disorders. A
thorough evaluation with trained personnel and physicians knowledgeable in the assessment
and treatment of these disorders is important. Although inpatient video-EEG monitoring in an
epilepsy monitoring unit remains the gold standard for diagnosis, the assessment of clinical and
historical factors is critical and can be useful in expediting the process and improving diagnostic
certainty. International efforts have recently assisted in providing guidelines for the evaluation
of the psychogenic disorders and may help target educational and other resources to
underserved areas.
SUMMARY:
The prompt and accurate diagnosis of nonepileptic episodic events and psychogenic
nonepileptic seizures is possible with current technology, and the appropriate and targeted use
of evidence-based treatments may help improve patient quality of life and avoid unnecessary
disability in patients with these disorders.
KEY POINTS
• The term nonepileptic episodic events is broad and includes disorders of both physiologic and psychogenic
origin. Nonepileptic events of a psychogenic origin are often referred to as psychogenic nonepileptic
seizures.
• Syncope may be mistaken for an epileptic seizure when followed by tonic-clonic movements (convulsive
syncope), and a thorough evaluation should be performed to exclude physiologic causes of loss of
consciousness.
• No single sign can differentiate between types of physiologic or psychogenic nonepileptic episodic events or
between nonepileptic episodic events and epilepsy.

• Patients with psychogenic nonepileptic seizures may account for 20% of those seen in an outpatient setting
and up to one-third of patients in an epilepsy monitoring unit.
• Psychiatric disorders, such as depression and anxiety, are common in patients with nonepileptic episodic
events and epileptic seizures, and their presence should not be used to discriminate between the two
disorders when making a diagnosis.
• Common clinical signs in patients with psychogenic nonepileptic seizures include waxing and waning
movements or a fluctuating course, a long duration of events, eye closure, ictal crying, gradual onset,
asynchronous movements, pelvic thrusting, recall during the period of apparent unresponsiveness,
and hyperventilation.
• Although a psychogenic nonepileptic seizure was traditionally considered a manifestation of a conversion,
somatization, or dissociative disorder, it is now considered to have a multifactorial etiology that also
comprises biological and social factors.
• The diagnosis of nonepileptic episodic events, including psychogenic nonepileptic seizures, should be a
stepwise process that includes clinical and historical assessment and video and EEG monitoring capturing
typical events for the patient.
• Inpatient video-EEG monitoring will demonstrate wakefulness during events for patients with psychogenic
nonepileptic seizures compared with some events in patients with physiologic nonepileptic episodic events
or epileptic seizures.
• Prolactin levels may help distinguish epileptic seizures from psychogenic nonepileptic seizures by
demonstrating a twofold rise from baseline in the 10 to 20 minutes after a seizure but should be interpreted
with caution.
• The diagnosis of psychogenic nonepileptic seizures should be presented to the patient in an honest, positive
manner with an outline of plans for further evaluation and treatment.
• Barriers to the treatment of psychogenic nonepileptic seizures include a lack of consistent follow-up with
neurologists, shortage of trained treatment providers, and stigma related to psychological and psychiatric care.
• Cognitive-behavioral therapy is an evidence-based treatment shown to reduce seizure frequency and
improve the quality of life in patients with psychogenic nonepileptic seizures.
• Although up to 70% of patients with psychogenic nonepileptic seizures may continue to have seizures after
diagnosis, higher education, shorter time to diagnosis, and lower somatoform and dissociative scores may
predict better outcomes.
• Up to 10% of patients with psychogenic nonepileptic seizures may also have epileptic seizures, and careful
attention is needed during evaluation and treatment to identify and manage both disorders.
• Physiologic causes of nonepileptic episodic events in children that may mimic epilepsy include reflux, sleep
disorders, and breath-holding spells.
Article 11: Update on Antiepileptic

Drugs 2019
Bassel W. Abou-Khalil, MD, FAAN. Continuum (Minneap Minn). April 2019; 25 (2
ABSTRACT
PURPOSE OF REVIEW:
This article is an update from the article on antiepileptic drug (AED) therapy published in the last
Continuum issue on epilepsy and is intended to cover the vast majority of agents currently
available to the neurologist in the management of patients with epilepsy. Treatment of epilepsy
starts with AED monotherapy. Knowledge of the spectrum of efficacy, clinical pharmacology,

and modes of use for individual AEDs is essential for optimal treatment for epilepsy. This article
addresses AEDs individually, focusing on key pharmacokinetic characteristics, indications, and
modes of use.
RECENT FINDINGS:
Since the previous version of this article was published, three new AEDs, brivaracetam,
cannabidiol, and stiripentol, have been approved by the US Food and Drug Administration (FDA),
and ezogabine was removed from the market because of decreased use as a result of bluish
skin pigmentation and concern over potential retinal toxicity.
Older AEDs are effective but have tolerability and pharmacokinetic disadvantages. Several
newer AEDs have undergone comparative trials demonstrating efficacy equal to and tolerability
at least equal to or better than older AEDs as first-line therapy. The list includes lamotrigine,
oxcarbazepine, levetiracetam, topiramate, zonisamide, and lacosamide. Pregabalin was found
to be less effective than lamotrigine. Lacosamide, pregabalin, and eslicarbazepine have
undergone successful trials of conversion to monotherapy. Other newer AEDs with a variety of
mechanisms of action are suitable for adjunctive therapy. Most recently, the FDA adopted a
policy that a drug’s efficacy as adjunctive therapy in adults can be extrapolated to efficacy in
monotherapy. In addition, efficacy in adults can be extrapolated for efficacy in children 4 years
of age and older. Both extrapolations require data demonstrating that an AED has equivalent
pharmacokinetics between its original approved use and its extrapolated use. In addition, the
safety of the drug in pediatric patients has to be demonstrated in clinical studies that can be
open label. Rational AED combinations should avoid AEDs with unfavorable pharmacokinetic
interactions or pharmacodynamic interactions related to mechanism of action.
SUMMARY:
Knowledge of AED pharmacokinetics, efficacy, and tolerability profiles facilitates the choice of
appropriate AED therapy for patients with epilepsy.
KEY POINTS
• Phenobarbital, primidone, phenytoin, and carbamazepine are potent inducers of liver enzymes, reducing the
efficacy of drugs metabolized by the cytochrome P450 enzyme system.
• Long-term phenobarbital use is associated with decreased bone density, Dupuytren contractures, plantar
fibromatosis, and frozen shoulder.
• In addition to sedation and other adverse effects of phenobarbital, primidone use is associated with an acute
toxic reaction unrelated to phenobarbital, with potentially debilitating drowsiness, dizziness, ataxia, nausea,
and vomiting.
• Phenytoin has saturable nonlinear kinetics. Beyond a certain serum concentration, usually within the
accepted therapeutic range, phenytoin concentration increases disproportionately with an increase in
the dose. Small increments are necessary when increasing the dose at a serum concentration in the
therapeutic range.
• The traditional sodium channel blockers phenytoin, carbamazepine, and oxcarbazepine may exacerbate
generalized absence and myoclonic seizures and should be avoided in idiopathic generalized epilepsy. Other
antiepileptic drugs that have similar potential are gabapentin, pregabalin, tiagabine, and vigabatrin.
• Unlike phenytoin, the phenytoin prodrug fosphenytoin may be administered intramuscularly, with reliable
absorption, in the absence of IV access.
• Carbamazepine induces its own metabolism, so it has to be titrated gradually to the target dose.
• The HLA-B1502 allele is predictive of a carbamazepine-induced severe rash in individuals of Asian descent.
• Oxcarbazepine is more likely to cause hyponatremia than carbamazepine. Older individuals taking a diuretic
are at particularly high risk.
• Eslicarbazepine has a long half-life in CSF, justifying once-daily oral dosing.
• Valproate has a broad spectrum of efficacy against all focal and generalized seizure types.

• Valproate has the highest rate of teratogenicity among antiepileptic drugs and should be avoided in women
of childbearing potential.
• Ethosuximide is the drug of choice for pure absence seizures. While valproate is equally effective, it is
associated with more cognitive adverse effects.
• Tolerance may develop to the therapeutic effect of benzodiazepines; this appears less likely with clobazam
than clonazepam.
• Felbamate-related aplastic anemia and liver failure are unlikely to start after 1 year of treatment.
• Gabapentin bioavailability is low and decreases with increasing doses.
• Like gabapentin, pregabalin has a narrow spectrum of efficacy against focal seizures and may exacerbate
generalized myoclonic and absence seizures.
• Lamotrigine clearance is decreased by valproate and increased by estrogen and pregnancy as well as by
enzyme inducers.
• Tiagabine may be associated with dose-related episodes of nonconvulsive status epilepticus or
encephalopathy, even in subjects who do not have epilepsy.
• Levetiracetam is the only antiepileptic drug with Class I evidence of efficacy against generalized myoclonic
seizures.
• Brivaracetam may have fewer behavioral side effects than levetiracetam.
• Zonisamide’s long half-life of about 60 hours may be an advantage in reducing the impact of a missed dose.
• Lacosamide may produce a dose-dependent prolongation in PR interval, which could be clinically significant
in patients with known conduction problems, or if it is combined with other drugs that have a similar effect.
• Long-term vigabatrin use may be associated with irreversible visual field constriction; hence, it should only
be continued if it produces a remarkable improvement in seizure control.
• Valproate reduces rufinamide clearance; as a result, rufinamide has to be started at a lower dose and titrated
more slowly in the presence of valproate.
• Perampanel has a very long half-life, justifying once-daily dosing.
• Antiepileptic drug combinations with different mechanisms of action may have a greater probability of success.

Issue Overview
Epilepsy, Volume 25, Issue 2, April 2019
Continuum: Lifelong Learning in Neurology® is designed to help practicing neurologists stay
abreast of advances in the field while simultaneously developing lifelong self-directed learning
skills.
Learning Objectives
Upon completion of this Continuum: Lifelong Learning in Neurology Epilepsy issue, participants will be
able to:
 Review the updated International League Against Epilepsy seizure and epilepsy
classification system, which now includes epilepsy syndrome and an emphasis on
etiology, and understand how the classification system provides a framework for
clinicians, researchers, and patients and their families
 Evaluate a first seizure or established epilepsy to determine the seizure type, epilepsy
syndrome, and etiology
 Define medically responsive epilepsy and discuss strategies to maintain sustained seizure
freedom in patients with medically responsive epilepsy
 Define drug-resistant epilepsy, review the current diagnostic and therapeutic tools for
surgical treatment of epilepsy, and discuss the current evidence in favor of treating
patients with drug-resistant epilepsy with surgical treatment over continued medical
treatment
 Discuss the various factors that influence the initiation, selection, and discontinuation of
antiepileptic drugs in children with epilepsy
 Discuss the latest data regarding management of epilepsy in women
 Recognize common interictal and ictal EEG findings in epilepsy

 Describe the diagnosis and treatment for the epilepsy emergencies of status epilepticus,
acute repetitive seizures, and autoimmune encephalitis
 Discuss the aspects of epilepsy beyond seizure control that can adversely impact safety
and quality of life
 Discuss the classification, evaluation strategies, and treatment approaches for patients
with nonepileptic episodic events
 Discuss the spectrum of efficacy, clinical pharmacology, and modes of use for individual
antiepileptic drugs
 Recognize the legal risks and responsibilities in counseling patients with epilepsy about
driving limitations
 Discuss the importance of medication reconciliation and identify strategies to implement
medication reconciliation
Core Competencies
This Continuum: Lifelong Learning in Neurology Epilepsy issue covers the following core
competencies:
 Patient Care
 Medical Knowledge
 Practice-Based Learning and Improvement
 Interpersonal and Communication Skills
 Professionalism
 Systems-Based Practice

Contributors
Cynthia L. Harden, MD, Guest Editor

Head of Clinical Epilepsy Development, Xenon Pharmaceuticals Inc, Burnaby, British
Columbia, Canada; Attending Neurologist, 14th Street Medical Arts Center, New York, New
York
Relationship Disclosure: Dr Harden is a full-time employee of Xenon Pharmaceuticals Inc and holds stock options
of unknown value in Xenon Pharmaceuticals Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Harden reports no disclosure.
Erik K. St. Louis, MD, MS, FAAN, Guest Editor

Co-director, Mayo Center for Sleep Medicine; Associate Professor of Neurology, Consultant in
Medicine and Neurology, Mayo Clinic College of Medicine and Science, Rochester, Minnesota
Relationship Disclosure: Dr St. Louis has served as a consultant for Axovant Sciences, Inc and has received
research/grant support from Axovant Sciences, Inc; Mayo Clinic Center for Clinical and Translational Science; the
Michael J. Fox Foundation; the National Institutes of Health/National Institute on Aging; National Heart, Lung, and
Blood Institute; and Sunovian Pharmaceuticals Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr St. Louis reports no disclosure.
Bassel W. Abou-Khalil, MD, FAAN

Professor of Neurology, Vanderbilt University Medical Center, Nashville, Tennessee
Relationship Disclosure: Dr Abou-Khalil has served on the editorial board of Clinical Neurophysiology and has
received research/grant support from Biogen, the National Institute of Neurological Disorders and Stroke, SK-
Pharma, Sunovion Pharmaceuticals Inc, and UCB SA.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Abou-Khalil discusses the unlabeled/investigational

use of primidone for the treatment of essential tremor, valproate for the treatment of generalized myoclonic and
generalized tonic-clonic seizures, gabapentin for the treatment of headache and sleep disorders, lamotrigine as a
first-line treatment for epilepsy, zonisamide as initial monotherapy for epilepsy, and cannabidiol and clobazam for
the treatment of focal seizures.
Derek Bauer, MD
Assistant Professor, Department of Neurology, University of Virginia, Charlottesville, Virginia
Relationship Disclosure: Dr Bauer is a site subinvestigator for GWEP-1521, a double-blind, randomized, placebo-
controlled study to investigate the efficacy and safety of cannabidiol as add-on therapy in patients with tuberous
sclerosis complex who experience inadequately controlled seizures.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Bauer reports no disclosure.
Amy Bennett, JD
Manager, Quality Improvement, American Academy of Neurology, Minneapolis, Minnesota
Relationship Disclosure: Ms Bennett reports no disclosure.
Unlabeled Use of Products/Investigational Use Disclosure: Ms Bennett reports no disclosure.
Hai Chen, MD, PhD

Assistant Professor of Neurology, George Washington University, Washington, DC
Relationship Disclosure: Dr Chen reports no disclosure.

Unlabeled Use of Products/Investigational Use Disclosure: Dr Chen reports no disclosure.
Elizabeth E. Gerard, MD
Associate Professor of Neurology, Feinberg School of Medicine, Northwestern University,
Chicago, Illinois
Relationship Disclosure: Dr Gerard has received personal compensation as a lecturer for the Society for Maternal-
Fetal Medicine, Society of OB/GYN Hospitalists, and UCB China, and has received research/grant support from the
Eleanor Wood-Prince Grant from the Woman’s Board of Northwestern Memorial Hospital, the National Institute of
Neurological Disorders and Stroke/National Institutes of Health, SAGE Therapeutics, and Sunovion Pharmaceutical
Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Gerard discusses the unlabeled/investigational use of
immunosuppressant medications for the treatment of autoimmune encephalitis (cyclophosphamide, IV
immunoglobulin, IV methylprednisolone/corticosteroids, plasma exchange, and rituximab) and the pharmacologic
and nonpharmacologic therapies for the treatment of seizures and status epilepticus (convulsive and nonconvulsive),
which include diazepam, fosphenytoin/phenytoin, levetiracetam, lorazepam, midazolam, phenobarbital, and
valproate sodium/valproic acid. Dr Gerard discusses the unlabeled/investigational use of several agents for the
treatment of refractory and super-refractory status epilepticus, which include corticosteroids/methylprednisolone,
electroconvulsive therapy, hypothermia, isoflurane, IV immunoglobulin, ketamine, ketogenic diet, midazolam,
pentobarbital, propofol, thiopental, transcranial magnetic stimulation, and vagal nerve stimulation.
Jennifer L. Hopp, MD, FAAN

Associate Professor of Neurology, University of Maryland School of Medicine; Director,
Epilepsy Division, University of Maryland Medical Center, Baltimore, Maryland
Relationship Disclosure: Dr Hopp has received grant/research support as a site principal investigator of the
Established Status Epilepticus Treatment Trial from the National Institute of Neurological Disorders and Stroke and
from SAGE Therapeutics. Dr Hopp has received personal compensation as a speaker for J. Kiffin Penry Epilepsy
MiniFellow Network’s Epilepsy MiniFellowship and Residents Epilepsy Program. Dr Hopp receives publishing
royalties from UpToDate, Inc and has given expert medical testimony for Venable LLP.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Hopp discusses the unlabeled/investigational use of
sertraline for the treatment of psychogenic nonepileptic seizures.
Joseph S. Kass, MD, JD, FAAN

Associate Dean, Office of Student Affairs; Professor of Neurology, Psychiatry, and Medical
Ethics; Director, Alzheimer’s Disease and Memory Disorders Center, Baylor College of
Medicine; Chief of Neurology, Ben Taub General Hospital, Houston, Texas
Relationship Disclosure: Dr Kass serves as associate editor of medicolegal issues for Continuum, as an associate
editor for Continuum Audio, as a neurology section editor of Ferri’s Clinical Advisor for Elsevier, and as co-editor
of Neurology Secrets, Sixth Edition. Dr Kass has received personal compensation for CME lectures from Pri-Med
LLC.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Kass reports no disclosure.
Mohamad Z. Koubeissi, MD, FAAN

Professor, Department of Neurology; Director, Epilepsy Center, George Washington University,
Washington, DC
Relationship Disclosure: Dr Koubeissi serves on the editorial boards of Epilepsy Currents and Functional Neurology
and as the surgery and device editor of Epilepsy.com. Dr Koubeissi has received personal compensation for serving
on the speakers’ bureaus of Sunovion Pharmaceuticals Inc and UCB SA, and has received publishing royalties from
Springer for his book, Epilepsy Board Review.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Koubeissi reports no disclosure.

Ahsan N. V. Moosa, MD (Ahsan Moosa Naduvil Valappil, MD)
Staff Physician, Pediatric Epilepsy, Epilepsy Center, Cleveland Clinic; Assistant Professor of
Medicine, Lerner College of Medicine, Case Western Reserve University, Cleveland, Ohio
Relationship Disclosure: Dr Moosa reports no disclosure.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Moosa reports no disclosure.

Associate Professor of Neurology, Mayo Clinic College of Medicine, Phoenix, Arizona
Relationship Disclosure: Dr Noe has received research/grant support from NeuroPace, Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Noe reports no disclosure.
Alison M. Pack, MD, MPH

Associate Professor of Neurology, Columbia University Irving Medical Center, New York, New
York
Relationship Disclosure: Dr Pack receives research/grant support from the National Institutes of Health/National
Institute of Neurological Disorders and Stroke as co-investigator of a multicenter study and receives royalties from
UpToDate, Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Pack reports no disclosure.
Fedor Panov, MD
Assistant Professor, Department of Neurosurgery, Icahn School of Medicine at Mount Sinai;
Associate Director, Adult Epilepsy Program, Mount Sinai, New York, New York
Relationship Disclosure: Dr Panov has received personal compensation as a consultant for NeuroPace, Inc and
Zimmer Biomet.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Panov reports no disclosure.
Mark Quigg, MD, MSc, FANA, FAES

Professor, Department of Neurology, University of Virginia, Charlottesville, Virginia
Relationship Disclosure: Dr Quigg has received research/grant support as principal investigator of studies from the
National Institutes of Health/National Institute of Neurological Disorders and Stroke, the University of Virginia
Brain Institute, and ZETO Inc. Dr Quigg has received publishing royalties from Elsevier and has given expert
medical testimony.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Quigg reports no disclosure.
Rachel V. Rose, JD, MBA

Attorney, Rachel V. Rose Attorney at Law PLLC; Affiliated Faculty, Baylor College of
Medicine, Houston, Texas
Relationship Disclosure: Ms Rose serves on the editorial board of BC Advantage and receives book royalties from
the American Bar Association.
Unlabeled Use of Products/Investigational Use Disclosure: Ms Rose reports no disclosure.

Clinical Professor, Department of Neurology, University of California, Irvine, Orange,
California
Relationship Disclosure: Dr Sazgar has received personal compensation for serving on the scientific advisory board
for UCB SA and for serving on the speaker’s bureau of Eisai Co, Ltd; Sunovion Pharmaceuticals Inc; and UCB SA.
Dr Sazgar has received grants or research support from Biogen, Sunovion Pharmaceuticals Inc, and UCB SA as the
principal investigator for clinical trials and has received publishing royalties from Springer for her book
Controversies in Caring for Women with Epilepsy.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Sazgar discusses the unlabeled/investigational use of
acetazolamide, clobazam, depot-medroxyprogesterone acetate, and progesterone products for the treatment of
women with catamenial epilepsy.
Stephan U. Schuele, MD, MPH, FAAN

Professor of Neurology and Physical Medicine and Rehabilitation, Feinberg School of Medicine,
Northwestern University, Chicago, Illinois
Relationship Disclosure: Dr Schuele has received personal compensation for serving on the speaker’s bureau of
Eisai Co, Ltd and Sunovion Pharmaceuticals Inc. Dr Schuele serves on the board of directors of the American
Clinical Neurophysiology Society and the Epilepsy Foundation Greater Chicago, and on the editorial board of
Journal of Clinical Neurophysiology. Dr Schuele has received research/grant support from the Danny Did
Foundation and the National Institutes of Health and has been asked to provide expert medical testimony in cases
involving sudden unexpected death in epilepsy.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Schuele reports no disclosure.
Alan Z. Segal, MD, FAAN

Associate Professor of Clinical Neurology, Weill Cornell Medical College, New York, New
York
Relationship Disclosure: Dr Segal reports no disclosure.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Segal reports no disclosure.
Anant M. Shenoy, MD, FAAN

Chief of Neurology, Mount Auburn Hospital, Cambridge, Massachusetts
Relationship Disclosure: Dr Shenoy reports no disclosure.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Shenoy reports no disclosure.
Stephen VanHaerents, MD
Assistant Professor of Neurology, Feinberg School of Medicine, Northwestern University,
Chicago, Illinois
Relationship Disclosure: Dr VanHaerents has received research/grant support from the Citizens United for Research
in Epilepsy, National Institute of Mental Health, National Institute on Aging, National Institutes of Health, and
SAGE Therapeutics. Dr VanHaerents has received travel honoraria from NeuroPace and SAGE Therapeutics.
Unlabeled Use of Products/Investigational Use Disclosure: Dr VanHaerents discusses the unlabeled/investigational

use of immunosuppressant medications for the treatment of autoimmune encephalitis (cyclophosphamide, IV
immunoglobulin, IV methylprednisolone/corticosteroids, plasma exchange, and rituximab) and the pharmacologic
and nonpharmacologic therapies for the treatment of seizures and status epilepticus (convulsive and nonconvulsive),
which include diazepam, fosphenytoin/phenytoin, levetiracetam, lorazepam, midazolam, phenobarbital, and
valproate sodium/valproic acid. Dr VanHaerents discusses the unlabeled/investigational use of several agents for the
treatment of refractory and super-refractory status epilepticus, which include corticosteroids/methylprednisolone,
electroconvulsive therapy, hypothermia, isoflurane, IV immunoglobulin, ketamine, ketogenic diet, midazolam,
pentobarbital, propofol, thiopental, transcranial magnetic stimulation, and vagal nerve stimulation.

Ji Yeoun Yoo, MD
Assistant Professor, Icahn School of Medicine at Mount Sinai, New York, New York
Relationship Disclosure: Dr Yoo has received publishing royalties from Elsevier for the book Rowan’s Primer of
EEG and has received personal compensation as a lecturer for the Korean Epilepsy Society and for serving on the
advisory board of Zimmer Biomet. Dr Yoo has received research/grant support from the NeuroNEXT Program of
the National Institutes of Health/National Institute of Neurological Disorders and Stroke.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Yoo reports no disclosure.
Self-Assessment and CME Test Writers

James W. M. Owens Jr, MD, PhD
Associate Professor of Neurology, Adjunct Associate Professor of Pediatrics, University of
Washington School of Medicine, Seattle, Washington
Relationship Disclosure: Dr Owens serves as CME co-editor for Neurology and receives publishing royalties from
UpToDate, Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Owens reports no disclosure.
Allyson R. Zazulia, MD
Professor of Neurology and Radiology, Associate Dean for Continuing Medical Education,
Washington University, St. Louis, Missouri
Relationship Disclosure: Dr Zazulia reports no disclosure.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Zazulia reports no disclosure.

Methods of Participation and Instructions for Use
Continuum: Lifelong Learning in Neurology® is designed to help practicing neurologists stay
abreast of advances in the field while simultaneously developing lifelong self-directed learning
skills. In Continuum, the process of absorbing, integrating, and applying the material presented is
as important as, if not more important than, the material itself.
The goals of Continuum include disseminating up-to-date information to the practicing
neurologist in a lively, interactive format; fostering self-assessment and lifelong study skills;
encouraging critical thinking; and, in the final analysis, strengthening and improving patient
care.
Each Continuum issue is prepared by distinguished authors who are acknowledged leaders in
their respective fields. Six issues are published annually and are composed of review articles,
case-based discussions on ethical and practice issues related to the issue topic, coding
information, and comprehensive continuing medical education (CME) and self-assessment
offerings. For detailed instructions regarding Continuum CME and self-assessment activities,
visit continpub.com/CME.
The review articles emphasize clinical issues emerging in the field in recent years. Case reports
and vignettes are used liberally, as are tables and illustrations. Audio interviews with the authors
of Continuum articles are published alongside each article, and video material relating to the
issue topic accompanies issues when applicable.
The text can be reviewed and digested most effectively by establishing a regular schedule of
study in the office or at home, either alone or in an interactive group. If subscribers use such
regular and perhaps new study habits, Continuum’s goal of establishing lifelong learning patterns
can be met.

Continum Epilepsy.2019 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Continum Epilepsy.2019 PDF

Uploaded by

Copyright:

Available Formats

APRIL 2019

VOL. 25 NO. 2 Epilepsy

304 Editor’s Preface

306 Epilepsy Overview and Revised Classiﬁcation of Seizures

322 Evaluation of Seizure Etiology From Routine Testing to Genetic

343 Optimizing Management of Medically Responsive Epilepsy

362 Identiﬁcation and Treatment of Drug-Resistant Epilepsy

381 Antiepileptic Drug Treatment of Epilepsy in Children

408 Treatment of Women With Epilepsy

431 Electroencephalography in Epilepsy Evaluation

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

508 Update on Antiepileptic Drugs 2019

543 Tips and Resources for Medication Reconciliation

SELF-ASSESSMENT AND CME

296 Learning Objectives and Core Competencies

551 Instructions for Completing Postreading Self-Assessment and CME

553 Postreading Self-Assessment and CME Test

565 Postreading Self-Assessment and CME Test—Preferred Responses

List of Abbreviations (Back Cover)

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Cynthia L. Harden, MD, Bassel W. Abou-Khalil, MD,

Unlabeled Use of Products/Investigational

298 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Mohamad Z. Koubeissi, MD, Alison M. Pack, MD, MPH

Unlabeled Use of Products/Investigational

Unlabeled Use of Products/Investigational

300 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Unlabeled Use of Products/Investigational Relationship Disclosure: Dr Segal reports no

Anant M. Shenoy, MD, FAAN

Unlabeled Use of Products/Investigational

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Stephen VanHaerents, MD Ji Yeoun Yoo, MD

302 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Self-Assessment and CME Test Writers

James W. M. Owens Jr, MD, PhD Allyson R. Zazulia, MD

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Cloudy With a 60% Chance of

304 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

RECENT FINDINGS: The 2017 ILAE classification system builds on newly

306 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

HISTORICAL OVERVIEW OF SEIZURE, EPILEPSY, AND EPILEPSY

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

The current classification systems build on newly formulated definitions of

308 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

310 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

relaxes. A generalized clonic seizure is characterized by bilateral and sustained

CASE 1-1 A 27-year-old man presented after experiencing an episode of loss of

312 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

314 APRIL 2019

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.